| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "global_step": 4626, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 1.7838305231301342e-05, | |
| "loss": 2.2294, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 1.5676610462602683e-05, | |
| "loss": 2.1683, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 1.3514915693904023e-05, | |
| "loss": 2.1713, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.009819507598877, | |
| "eval_runtime": 82.235, | |
| "eval_samples_per_second": 63.148, | |
| "eval_steps_per_second": 7.904, | |
| "step": 1542 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 1.1353220925205362e-05, | |
| "loss": 2.1142, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 9.191526156506702e-06, | |
| "loss": 2.0871, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 7.029831387808041e-06, | |
| "loss": 2.0736, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.9852919578552246, | |
| "eval_runtime": 82.023, | |
| "eval_samples_per_second": 63.312, | |
| "eval_steps_per_second": 7.925, | |
| "step": 3084 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 4.8681366191093824e-06, | |
| "loss": 2.0674, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 2.706441850410722e-06, | |
| "loss": 2.0482, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 5.447470817120623e-07, | |
| "loss": 2.0543, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 2.013369083404541, | |
| "eval_runtime": 82.0996, | |
| "eval_samples_per_second": 63.252, | |
| "eval_steps_per_second": 7.917, | |
| "step": 4626 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 4626, | |
| "total_flos": 7.791321717294797e+16, | |
| "train_loss": 2.110191111022448, | |
| "train_runtime": 17635.5568, | |
| "train_samples_per_second": 16.785, | |
| "train_steps_per_second": 0.262 | |
| } | |
| ], | |
| "max_steps": 4626, | |
| "num_train_epochs": 3, | |
| "total_flos": 7.791321717294797e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |