| { | |
| "best_metric": 0.981159271079996, | |
| "best_model_checkpoint": "./output//42_roberta-large_nace_5__5e-6_0.01_0.06_07-21-22_10-40/checkpoint-12000", | |
| "epoch": 4.418262150220913, | |
| "global_step": 12000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 4.535840188014101e-06, | |
| "loss": 0.1726, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "eval_accuracy": 0.9712155725823152, | |
| "eval_f1": 0.9751469526657729, | |
| "eval_loss": 0.08084136247634888, | |
| "eval_roc_auc": 0.9750559610355007, | |
| "eval_runtime": 143.2301, | |
| "eval_samples_per_second": 33.715, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 3.7524481002741874e-06, | |
| "loss": 0.0668, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "eval_accuracy": 0.9726651480637813, | |
| "eval_f1": 0.9771228359439407, | |
| "eval_loss": 0.08218056708574295, | |
| "eval_roc_auc": 0.9770191055094254, | |
| "eval_runtime": 143.2878, | |
| "eval_samples_per_second": 33.701, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 2.969056012534274e-06, | |
| "loss": 0.0501, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "eval_accuracy": 0.9749430523917996, | |
| "eval_f1": 0.9794069192751236, | |
| "eval_loss": 0.07129524648189545, | |
| "eval_roc_auc": 0.9792922833616305, | |
| "eval_runtime": 143.1874, | |
| "eval_samples_per_second": 33.725, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 2.18566392479436e-06, | |
| "loss": 0.0377, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "eval_accuracy": 0.9757713812383516, | |
| "eval_f1": 0.9797813080255828, | |
| "eval_loss": 0.08654613047838211, | |
| "eval_roc_auc": 0.9797172685997191, | |
| "eval_runtime": 143.3834, | |
| "eval_samples_per_second": 33.679, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 1.4022718370544457e-06, | |
| "loss": 0.0254, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "eval_accuracy": 0.9743218057568854, | |
| "eval_f1": 0.9794111591517397, | |
| "eval_loss": 0.09397488832473755, | |
| "eval_roc_auc": 0.979291082596074, | |
| "eval_runtime": 143.2996, | |
| "eval_samples_per_second": 33.699, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.42, | |
| "learning_rate": 6.18879749314532e-07, | |
| "loss": 0.0196, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 4.42, | |
| "eval_accuracy": 0.9763926278732656, | |
| "eval_f1": 0.981159271079996, | |
| "eval_loss": 0.09825791418552399, | |
| "eval_roc_auc": 0.981051940958186, | |
| "eval_runtime": 143.2702, | |
| "eval_samples_per_second": 33.706, | |
| "step": 12000 | |
| } | |
| ], | |
| "max_steps": 13580, | |
| "num_train_epochs": 5, | |
| "total_flos": 6.9866971594752e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |