| { |
| "best_metric": 1.117621898651123, |
| "best_model_checkpoint": "clean_output/bert-large-cased/l_context/checkpoint-586", |
| "epoch": 2.0, |
| "global_step": 586, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.49786324786324787, |
| "eval_loss": 1.1456269025802612, |
| "eval_runtime": 10.4236, |
| "eval_samples_per_second": 224.491, |
| "eval_steps_per_second": 28.109, |
| "step": 293 |
| }, |
| { |
| "epoch": 1.71, |
| "learning_rate": 7.680097680097681e-06, |
| "loss": 1.2045, |
| "step": 500 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.5205128205128206, |
| "eval_loss": 1.117621898651123, |
| "eval_runtime": 10.4232, |
| "eval_samples_per_second": 224.499, |
| "eval_steps_per_second": 28.11, |
| "step": 586 |
| } |
| ], |
| "max_steps": 1758, |
| "num_train_epochs": 6, |
| "total_flos": 6812609916643200.0, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
