| { | |
| "best_metric": 0.9642899940083882, | |
| "best_model_checkpoint": "bert-base-uncased-finetuned-spam-real/run-0/checkpoint-1670", | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 1670, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.30469822883605957, | |
| "learning_rate": 3.878567975913952e-06, | |
| "loss": 0.3154, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9488915518274416, | |
| "eval_loss": 0.22129623591899872, | |
| "eval_runtime": 335.9128, | |
| "eval_samples_per_second": 49.686, | |
| "eval_steps_per_second": 3.108, | |
| "step": 835 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 0.06290946900844574, | |
| "learning_rate": 2.2210602938994423e-06, | |
| "loss": 0.1341, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 0.40462997555732727, | |
| "learning_rate": 5.635526118849332e-07, | |
| "loss": 0.1332, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9642899940083882, | |
| "eval_loss": 0.16416475176811218, | |
| "eval_runtime": 335.3225, | |
| "eval_samples_per_second": 49.773, | |
| "eval_steps_per_second": 3.113, | |
| "step": 1670 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1670, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "total_flos": 2178221399563560.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 5.536075657928461e-06, | |
| "num_train_epochs": 2, | |
| "per_device_train_batch_size": 4, | |
| "seed": 30 | |
| } | |
| } | |