{ "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-lam4e-5-s2-fb/select_scored.jsonl", "sha256": "06ca5e863bfc79fa27b0b9dcc16984d3faf131fb3e7b7014b2d4e7588a971c6b" }, "summary.json": { "commit": "996adf51777e618d115e27e9d7f8723cac857ab1", "dirty": false, "path": "/cluster/scratch/jingni/iclr/olmo3_rep/runs/s3_full_fb/osp-lam4e-5-s2-fb/summary.json", "sha256": "dd89ca466e7e57fe042b3ef065e245918ca8d563a7708f6dd7d21e7443439f9f" }, "val_scored.jsonl": { "commit": "996adf51777e618d115e27e9d7f8723cac857ab1", "dirty": false, "path": "/cluster/scratch/jingni/iclr/olmo3_rep/runs/s3_full_fb/osp-lam4e-5-s2-fb/val_scored.jsonl", "sha256": "29eb806d9e8dec74856a46ccdd2010054c7d89a224d92758a99ad0b4ccaadd79" } }, "n_dumped": { "select": 3804, "val": 1278 }, "n_select": 3804, "n_val": 1278, "path": "/cluster/scratch/jingni/iclr/olmo3_rep/runs/s3_full_fb/osp-lam4e-5-s2-fb", "world_size": 3, "writer_rank": 0 }, "n_global_steps_per_epoch": 5224, "parity_digests": { "args_digest": "2817192a4c41e83d", "param_layout": "0f8d0e7904b5ca53", "resume_epoch": "1feceb66ffc86f38", "resume_migration": "34234e98afe7498f", "resume_plan": "072b06957c0acf0a", "seed_lambda": "0b5d2c8652f4d9f8", "split_sizes": "320c275502e54dac", "train_digest": "2eef28cfe61f5ba8" }, "resume": { "args_digest_loaded": "52dfeeff99c9260e2ec961c5b57e972d1db78c15584988d50acd2c07692eea6e", "args_digest_now": "52dfeeff99c9260e2ec961c5b57e972d1db78c15584988d50acd2c07692eea6e", "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": "18dcb4d02da06a12", "n_chains": 1278, "n_scores": 73046, "split": "val", "world": 3 }, { "attempt": "18dcb4d02da06a12", "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": -0.1296359264409878, "world": 3 }, { "epoch": 2, "n_chains_contributing": 1278, "n_chains_expected": 1278, "n_chains_global": 1278, "n_chains_local": 426, "val_stat": -0.2484404359811535, "world": 3 }, { "epoch": 3, "n_chains_contributing": 1278, "n_chains_expected": 1278, "n_chains_global": 1278, "n_chains_local": 426, "val_stat": -0.497947067682557, "world": 3 }, { "epoch": 4, "n_chains_contributing": 1278, "n_chains_expected": 1278, "n_chains_global": 1278, "n_chains_local": 426, "val_stat": -0.48246892834015204, "world": 3 }, { "epoch": 5, "n_chains_contributing": 1278, "n_chains_expected": 1278, "n_chains_global": 1278, "n_chains_local": 426, "val_stat": -0.48677138899670147, "world": 3 }, { "epoch": 6, "n_chains_contributing": 1278, "n_chains_expected": 1278, "n_chains_global": 1278, "n_chains_local": 426, "val_stat": -0.46290617150215196, "world": 3 } ], "world_size": 3 }