| { | |
| "best_metric": 0.8434029546612328, | |
| "best_model_checkpoint": "./outputs/bert-base-uncased/fine_tuned_models/checkpoint-36816", | |
| "epoch": 5.0, | |
| "global_step": 61360, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.7388593694651082e-05, | |
| "loss": 0.5875, | |
| "step": 12272 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8329088130412634, | |
| "eval_loss": 0.4355992078781128, | |
| "eval_runtime": 24.4348, | |
| "eval_samples_per_second": 401.682, | |
| "eval_steps_per_second": 50.215, | |
| "step": 12272 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.3041445270988312e-05, | |
| "loss": 0.371, | |
| "step": 24544 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8348446255731024, | |
| "eval_loss": 0.438580721616745, | |
| "eval_runtime": 24.4254, | |
| "eval_samples_per_second": 401.836, | |
| "eval_steps_per_second": 50.235, | |
| "step": 24544 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 8.694296847325541e-06, | |
| "loss": 0.248, | |
| "step": 36816 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8434029546612328, | |
| "eval_loss": 0.4767991900444031, | |
| "eval_runtime": 24.4585, | |
| "eval_samples_per_second": 401.293, | |
| "eval_steps_per_second": 50.167, | |
| "step": 36816 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.3471484236627705e-06, | |
| "loss": 0.1643, | |
| "step": 49088 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8407539480387163, | |
| "eval_loss": 0.5648263096809387, | |
| "eval_runtime": 24.4437, | |
| "eval_samples_per_second": 401.536, | |
| "eval_steps_per_second": 50.197, | |
| "step": 49088 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.1136, | |
| "step": 61360 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8391237901171676, | |
| "eval_loss": 0.7124413847923279, | |
| "eval_runtime": 24.4342, | |
| "eval_samples_per_second": 401.691, | |
| "eval_steps_per_second": 50.217, | |
| "step": 61360 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 61360, | |
| "total_flos": 1.6510003421181696e+17, | |
| "train_loss": 0.29687764193896676, | |
| "train_runtime": 13416.1217, | |
| "train_samples_per_second": 146.355, | |
| "train_steps_per_second": 4.574 | |
| } | |
| ], | |
| "max_steps": 61360, | |
| "num_train_epochs": 5, | |
| "total_flos": 1.6510003421181696e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |