| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 2397, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 14.102521896362305, | |
| "learning_rate": 3.3125e-05, | |
| "loss": 1.5474, | |
| "step": 159 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 11.06152057647705, | |
| "learning_rate": 4.8191933240611966e-05, | |
| "loss": 1.3906, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 11.38102912902832, | |
| "learning_rate": 4.4506258692628656e-05, | |
| "loss": 1.3152, | |
| "step": 477 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 10.112909317016602, | |
| "learning_rate": 4.082058414464534e-05, | |
| "loss": 1.2965, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 8.64354133605957, | |
| "learning_rate": 3.713490959666203e-05, | |
| "loss": 1.2541, | |
| "step": 795 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "grad_norm": 8.374133110046387, | |
| "learning_rate": 3.344923504867872e-05, | |
| "loss": 1.1896, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "grad_norm": 13.710701942443848, | |
| "learning_rate": 2.9763560500695413e-05, | |
| "loss": 1.1773, | |
| "step": 1113 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "grad_norm": 11.795424461364746, | |
| "learning_rate": 2.60778859527121e-05, | |
| "loss": 1.1184, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "grad_norm": 9.229485511779785, | |
| "learning_rate": 2.239221140472879e-05, | |
| "loss": 1.1397, | |
| "step": 1431 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "grad_norm": 18.048118591308594, | |
| "learning_rate": 1.870653685674548e-05, | |
| "loss": 1.0934, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "grad_norm": 12.560694694519043, | |
| "learning_rate": 1.5020862308762171e-05, | |
| "loss": 1.0125, | |
| "step": 1749 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "grad_norm": 28.996366500854492, | |
| "learning_rate": 1.1335187760778861e-05, | |
| "loss": 1.0356, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "grad_norm": 17.208913803100586, | |
| "learning_rate": 7.64951321279555e-06, | |
| "loss": 0.9445, | |
| "step": 2067 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "grad_norm": 16.612789154052734, | |
| "learning_rate": 3.96383866481224e-06, | |
| "loss": 1.0326, | |
| "step": 2226 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "grad_norm": 24.243637084960938, | |
| "learning_rate": 2.7816411682892905e-07, | |
| "loss": 0.8955, | |
| "step": 2385 | |
| } | |
| ], | |
| "logging_steps": 159, | |
| "max_steps": 2397, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 1.4854311755525652e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |