JingweiNi's picture
Publish original trained checkpoints and run metadata
085c688 verified
Raw History Blame Contribute Delete
3.96 kB
{
"backend": "nccl",
"chains_per_step_is_global_batch": true,
"dist_timeout_s": 10800,
"export": {
"files": {
"select_scored.jsonl": {
"commit": "996adf51777e618d115e27e9d7f8723cac857ab1",
"dirty": false,
"path": "/cluster/scratch/jingni/iclr/olmo3_rep/runs/s3_full_fb/osp-lam2e-5-s0-fb/select_scored.jsonl",
"sha256": "fd7e1e51a59db209f467cf2ab7976ab7d0146509aa0c2c865fe6f7714e7cb6bb"
},
"summary.json": {
"commit": "996adf51777e618d115e27e9d7f8723cac857ab1",
"dirty": false,
"path": "/cluster/scratch/jingni/iclr/olmo3_rep/runs/s3_full_fb/osp-lam2e-5-s0-fb/summary.json",
"sha256": "4216ef3f511c24fded63db21f037e2a60295ce1b960bb0e512f27e6cc4c99e1c"
},
"val_scored.jsonl": {
"commit": "996adf51777e618d115e27e9d7f8723cac857ab1",
"dirty": false,
"path": "/cluster/scratch/jingni/iclr/olmo3_rep/runs/s3_full_fb/osp-lam2e-5-s0-fb/val_scored.jsonl",
"sha256": "b74f14cdd3388bc03dad993f7e5c1f9e930b46d375e4dd445ccf6bc0af1ebbac"
}
},
"n_dumped": {
"select": 3804,
"val": 1278
},
"n_select": 3804,
"n_val": 1278,
"path": "/cluster/scratch/jingni/iclr/olmo3_rep/runs/s3_full_fb/osp-lam2e-5-s0-fb",
"world_size": 3,
"writer_rank": 0
},
"n_global_steps_per_epoch": 5224,
"parity_digests": {
"args_digest": "0a6faf080a22f613",
"param_layout": "0f8d0e7904b5ca53",
"resume_epoch": "1feceb66ffc86f38",
"resume_migration": "34234e98afe7498f",
"resume_plan": "072b06957c0acf0a",
"seed_lambda": "1b4769f78bb8e610",
"split_sizes": "320c275502e54dac",
"train_digest": "2eef28cfe61f5ba8"
},
"resume": {
"args_digest_loaded": "dea98fcc255a1ff29b0bd10148bcd5b0fcbbfb15a964005dc685f9375edf1b08",
"args_digest_now": "dea98fcc255a1ff29b0bd10148bcd5b0fcbbfb15a964005dc685f9375edf1b08",
"checkpoint_found": true,
"enabled": true,
"migration": null,
"plan": {
"complete": false,
"completed_epochs": 1,
"early_stopped": false,
"remaining": 6,
"start_epoch": 1
},
"resumed_from_epoch": 0,
"rng_replay_verified": true,
"train_digest": "2f23490618025ec10b7a8bd77eb7f04fec74326d3b0800e896fa58d63ce9dcc6"
},
"schema": "ospruner-execution/v1",
"score_shards": [
{
"attempt": "18dc73001ca4338f",
"n_chains": 1278,
"n_scores": 73046,
"split": "val",
"world": 3
},
{
"attempt": "18dc73001ca4338f",
"n_chains": 3801,
"n_scores": 158686,
"split": "select",
"world": 3
}
],
"validation": [
{
"epoch": 1,
"n_chains_contributing": 1278,
"n_chains_expected": 1278,
"n_chains_global": 1278,
"n_chains_local": 426,
"val_stat": -0.3255031144691716,
"world": 3
},
{
"epoch": 2,
"n_chains_contributing": 1278,
"n_chains_expected": 1278,
"n_chains_global": 1278,
"n_chains_local": 426,
"val_stat": -0.4972470615624635,
"world": 3
},
{
"epoch": 3,
"n_chains_contributing": 1278,
"n_chains_expected": 1278,
"n_chains_global": 1278,
"n_chains_local": 426,
"val_stat": -0.571000778436929,
"world": 3
},
{
"epoch": 4,
"n_chains_contributing": 1278,
"n_chains_expected": 1278,
"n_chains_global": 1278,
"n_chains_local": 426,
"val_stat": -0.5730022685864652,
"world": 3
},
{
"epoch": 5,
"n_chains_contributing": 1278,
"n_chains_expected": 1278,
"n_chains_global": 1278,
"n_chains_local": 426,
"val_stat": -0.5124424087708591,
"world": 3
},
{
"epoch": 6,
"n_chains_contributing": 1278,
"n_chains_expected": 1278,
"n_chains_global": 1278,
"n_chains_local": 426,
"val_stat": -0.5222622259058706,
"world": 3
}
],
"world_size": 3
}