| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.19809825673534073, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 3.846153846153847e-06, | |
| "loss": 2.7796, | |
| "theoretical_loss": 3.3518567762894107, | |
| "tokens_seen": 2969305088 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 2.4642, | |
| "theoretical_loss": 3.3517550749686795, | |
| "tokens_seen": 2970484736 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 7.692307692307693e-05, | |
| "loss": 1.9063, | |
| "theoretical_loss": 3.351642134115535, | |
| "tokens_seen": 2971795456 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 9.983987189751803e-05, | |
| "loss": 1.4437, | |
| "theoretical_loss": 3.351529257004948, | |
| "tokens_seen": 2973106176 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 9.943955164131305e-05, | |
| "loss": 1.273, | |
| "theoretical_loss": 3.3514164435728655, | |
| "tokens_seen": 2974416896 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 9.90392313851081e-05, | |
| "loss": 1.2074, | |
| "theoretical_loss": 3.3513036937553267, | |
| "tokens_seen": 2975727616 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 9.863891112890312e-05, | |
| "loss": 1.2079, | |
| "theoretical_loss": 3.3511910074884628, | |
| "tokens_seen": 2977038336 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "objective/train/docs_used": 997783, | |
| "objective/train/instantaneous_batch_size": 8, | |
| "objective/train/instantaneous_microbatch_size": 8192, | |
| "objective/train/original_loss": 0.9412798285484314, | |
| "objective/train/theoretical_loss": 3.3511628458440637, | |
| "objective/train/tokens_used": 28650976, | |
| "theoretical_loss": 3.3511628458440637, | |
| "tokens_seen": 2977366016 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 9.823859087269817e-05, | |
| "loss": 1.2032, | |
| "theoretical_loss": 3.3510783847084977, | |
| "tokens_seen": 2978349056 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 9.78382706164932e-05, | |
| "loss": 1.2178, | |
| "theoretical_loss": 3.350965825351748, | |
| "tokens_seen": 2979659776 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 9.743795036028824e-05, | |
| "loss": 1.189, | |
| "theoretical_loss": 3.3508533293546208, | |
| "tokens_seen": 2980970496 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 9.703763010408326e-05, | |
| "loss": 1.2086, | |
| "theoretical_loss": 3.3507408966536154, | |
| "tokens_seen": 2982281216 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 9.663730984787832e-05, | |
| "loss": 1.1797, | |
| "theoretical_loss": 3.3506285271853233, | |
| "tokens_seen": 2983591936 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 9.623698959167334e-05, | |
| "loss": 1.1709, | |
| "theoretical_loss": 3.350516220886427, | |
| "tokens_seen": 2984902656 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "objective/train/docs_used": 1000073, | |
| "objective/train/instantaneous_batch_size": 8, | |
| "objective/train/instantaneous_microbatch_size": 8192, | |
| "objective/train/original_loss": 1.0221836566925049, | |
| "objective/train/theoretical_loss": 3.3504600914057403, | |
| "objective/train/tokens_used": 36842976, | |
| "theoretical_loss": 3.3504600914057403, | |
| "tokens_seen": 2985558016 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 9.583666933546838e-05, | |
| "loss": 1.1658, | |
| "theoretical_loss": 3.3504039776936994, | |
| "tokens_seen": 2986213376 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 9.543634907926342e-05, | |
| "loss": 1.1823, | |
| "theoretical_loss": 3.350291797544005, | |
| "tokens_seen": 2987524096 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 9.503602882305846e-05, | |
| "loss": 1.1859, | |
| "theoretical_loss": 3.3501796803742994, | |
| "tokens_seen": 2988834816 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 9.463570856685348e-05, | |
| "loss": 1.1906, | |
| "theoretical_loss": 3.3500676261216285, | |
| "tokens_seen": 2990145536 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 9.423538831064852e-05, | |
| "loss": 1.185, | |
| "theoretical_loss": 3.3499556347231287, | |
| "tokens_seen": 2991456256 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 9.383506805444356e-05, | |
| "loss": 1.2042, | |
| "theoretical_loss": 3.349843706116027, | |
| "tokens_seen": 2992766976 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "objective/train/docs_used": 1002769, | |
| "objective/train/instantaneous_batch_size": 8, | |
| "objective/train/instantaneous_microbatch_size": 8192, | |
| "objective/train/original_loss": 1.227379560470581, | |
| "objective/train/theoretical_loss": 3.349759800829844, | |
| "objective/train/tokens_used": 45034976, | |
| "theoretical_loss": 3.349759800829844, | |
| "tokens_seen": 2993750016 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 9.34347477982386e-05, | |
| "loss": 1.1807, | |
| "theoretical_loss": 3.3497318402376397, | |
| "tokens_seen": 2994077696 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 9.303442754203363e-05, | |
| "loss": 1.1862, | |
| "theoretical_loss": 3.349620037025374, | |
| "tokens_seen": 2995388416 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 9.263410728582867e-05, | |
| "loss": 1.178, | |
| "theoretical_loss": 3.349508296416727, | |
| "tokens_seen": 2996699136 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 9.22337870296237e-05, | |
| "loss": 1.1656, | |
| "theoretical_loss": 3.3493966183492847, | |
| "tokens_seen": 2998009856 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 9.183346677341874e-05, | |
| "loss": 1.1769, | |
| "theoretical_loss": 3.349285002760723, | |
| "tokens_seen": 2999320576 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.143314651721377e-05, | |
| "loss": 1.1513, | |
| "theoretical_loss": 3.3491734495888066, | |
| "tokens_seen": 3000631296 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "objective/train/docs_used": 1005438, | |
| "objective/train/instantaneous_batch_size": 8, | |
| "objective/train/instantaneous_microbatch_size": 8192, | |
| "objective/train/original_loss": 1.1515272855758667, | |
| "objective/train/theoretical_loss": 3.3490619587713906, | |
| "objective/train/tokens_used": 53226976, | |
| "theoretical_loss": 3.3490619587713906, | |
| "tokens_seen": 3001942016 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.103282626100881e-05, | |
| "loss": 1.1793, | |
| "theoretical_loss": 3.3490619587713906, | |
| "tokens_seen": 3001942016 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.063250600480385e-05, | |
| "loss": 1.1534, | |
| "theoretical_loss": 3.348950530246418, | |
| "tokens_seen": 3003252736 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 9.023218574859889e-05, | |
| "loss": 1.1928, | |
| "theoretical_loss": 3.348839163951921, | |
| "tokens_seen": 3004563456 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 8.983186549239391e-05, | |
| "loss": 1.1678, | |
| "theoretical_loss": 3.3487278598260195, | |
| "tokens_seen": 3005874176 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 8.943154523618895e-05, | |
| "loss": 1.1477, | |
| "theoretical_loss": 3.348616617806923, | |
| "tokens_seen": 3007184896 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 8.903122497998399e-05, | |
| "loss": 1.1773, | |
| "theoretical_loss": 3.34850543783293, | |
| "tokens_seen": 3008495616 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 8.863090472377903e-05, | |
| "loss": 1.185, | |
| "theoretical_loss": 3.3483943198424253, | |
| "tokens_seen": 3009806336 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "objective/train/docs_used": 1008334, | |
| "objective/train/instantaneous_batch_size": 8, | |
| "objective/train/instantaneous_microbatch_size": 8192, | |
| "objective/train/original_loss": 1.3460177183151245, | |
| "objective/train/theoretical_loss": 3.348366550022507, | |
| "objective/train/tokens_used": 61418976, | |
| "theoretical_loss": 3.348366550022507, | |
| "tokens_seen": 3010134016 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 8.823058446757406e-05, | |
| "loss": 1.1798, | |
| "theoretical_loss": 3.348283263773882, | |
| "tokens_seen": 3011117056 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 8.78302642113691e-05, | |
| "loss": 1.1794, | |
| "theoretical_loss": 3.3481722695658633, | |
| "tokens_seen": 3012427776 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 8.742994395516414e-05, | |
| "loss": 1.1759, | |
| "theoretical_loss": 3.3480613371570165, | |
| "tokens_seen": 3013738496 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 8.702962369895917e-05, | |
| "loss": 1.1839, | |
| "theoretical_loss": 3.347950466486079, | |
| "tokens_seen": 3015049216 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 8.66293034427542e-05, | |
| "loss": 1.1996, | |
| "theoretical_loss": 3.3478396574918747, | |
| "tokens_seen": 3016359936 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 8.622898318654924e-05, | |
| "loss": 1.1897, | |
| "theoretical_loss": 3.347728910113315, | |
| "tokens_seen": 3017670656 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "objective/train/docs_used": 1011213, | |
| "objective/train/instantaneous_batch_size": 8, | |
| "objective/train/instantaneous_microbatch_size": 8192, | |
| "objective/train/original_loss": 1.227089524269104, | |
| "objective/train/theoretical_loss": 3.3476735595108353, | |
| "objective/train/tokens_used": 69610976, | |
| "theoretical_loss": 3.3476735595108353, | |
| "tokens_seen": 3018326016 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 8.582866293034428e-05, | |
| "loss": 1.1741, | |
| "theoretical_loss": 3.3476182242893975, | |
| "tokens_seen": 3018981376 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 8.542834267413932e-05, | |
| "loss": 1.191, | |
| "theoretical_loss": 3.3475075999592074, | |
| "tokens_seen": 3020292096 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 8.502802241793434e-05, | |
| "loss": 1.1924, | |
| "theoretical_loss": 3.347397037061916, | |
| "tokens_seen": 3021602816 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 8.462770216172938e-05, | |
| "loss": 1.188, | |
| "theoretical_loss": 3.3472865355367825, | |
| "tokens_seen": 3022913536 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 8.422738190552442e-05, | |
| "loss": 1.1754, | |
| "theoretical_loss": 3.3471760953231504, | |
| "tokens_seen": 3024224256 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 8.382706164931946e-05, | |
| "loss": 1.2086, | |
| "theoretical_loss": 3.3470657163604507, | |
| "tokens_seen": 3025534976 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "objective/train/docs_used": 1013697, | |
| "objective/train/instantaneous_batch_size": 8, | |
| "objective/train/instantaneous_microbatch_size": 8192, | |
| "objective/train/original_loss": 1.1135369539260864, | |
| "objective/train/theoretical_loss": 3.3469829722979627, | |
| "objective/train/tokens_used": 77802976, | |
| "theoretical_loss": 3.3469829722979627, | |
| "tokens_seen": 3026518016 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 8.34267413931145e-05, | |
| "loss": 1.1969, | |
| "theoretical_loss": 3.3469553985882, | |
| "tokens_seen": 3026845696 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 8.302642113690953e-05, | |
| "loss": 1.2176, | |
| "theoretical_loss": 3.3468451419460017, | |
| "tokens_seen": 3028156416 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 8.262610088070457e-05, | |
| "loss": 1.188, | |
| "theoretical_loss": 3.3467349463735436, | |
| "tokens_seen": 3029467136 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 8.22257806244996e-05, | |
| "loss": 1.1834, | |
| "theoretical_loss": 3.3466248118105995, | |
| "tokens_seen": 3030777856 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 8.182546036829464e-05, | |
| "loss": 1.1977, | |
| "theoretical_loss": 3.3465147381970293, | |
| "tokens_seen": 3032088576 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 8.142514011208967e-05, | |
| "loss": 1.1889, | |
| "theoretical_loss": 3.346404725472777, | |
| "tokens_seen": 3033399296 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "objective/train/docs_used": 1016482, | |
| "objective/train/instantaneous_batch_size": 8, | |
| "objective/train/instantaneous_microbatch_size": 8192, | |
| "objective/train/original_loss": 1.4041112661361694, | |
| "objective/train/theoretical_loss": 3.3462947735778727, | |
| "objective/train/tokens_used": 85994976, | |
| "theoretical_loss": 3.3462947735778727, | |
| "tokens_seen": 3034710016 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 8.102481985588471e-05, | |
| "loss": 1.1983, | |
| "theoretical_loss": 3.3462947735778727, | |
| "tokens_seen": 3034710016 | |
| } | |
| ], | |
| "max_steps": 2524, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 3.3445380096e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |