| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "global_step": 1587, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0001320754716981132, | |
| "loss": 1.3208, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.00019285714285714286, | |
| "loss": 1.2357, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.0001781512605042017, | |
| "loss": 1.1761, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 0.0001634453781512605, | |
| "loss": 1.1779, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 0.00014873949579831934, | |
| "loss": 1.1687, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 0.00013403361344537814, | |
| "loss": 1.1464, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 0.00011932773109243697, | |
| "loss": 1.1646, | |
| "step": 735 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 0.00010462184873949581, | |
| "loss": 1.1561, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 8.991596638655462e-05, | |
| "loss": 1.1641, | |
| "step": 945 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 7.521008403361345e-05, | |
| "loss": 1.1321, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 6.0504201680672267e-05, | |
| "loss": 1.1328, | |
| "step": 1155 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 4.579831932773109e-05, | |
| "loss": 1.1272, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 3.1092436974789916e-05, | |
| "loss": 1.1187, | |
| "step": 1365 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 1.638655462184874e-05, | |
| "loss": 1.1136, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 1.680672268907563e-06, | |
| "loss": 1.112, | |
| "step": 1575 | |
| } | |
| ], | |
| "max_steps": 1587, | |
| "num_train_epochs": 3, | |
| "total_flos": 2.6322813809432986e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |