| { | |
| "best_metric": 0.9277522935779816, | |
| "best_model_checkpoint": "electra-distilled-sst\\run-11\\checkpoint-3162", | |
| "epoch": 6.0, | |
| "eval_steps": 500, | |
| "global_step": 3162, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 14.338537216186523, | |
| "learning_rate": 5.0174854029835415e-05, | |
| "loss": 0.5621, | |
| "step": 527 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9071100917431193, | |
| "eval_loss": 0.39338329434394836, | |
| "eval_runtime": 1.4401, | |
| "eval_samples_per_second": 605.519, | |
| "eval_steps_per_second": 4.861, | |
| "step": 527 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 21.374542236328125, | |
| "learning_rate": 4.302061894724029e-05, | |
| "loss": 0.3067, | |
| "step": 1054 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9185779816513762, | |
| "eval_loss": 0.45042580366134644, | |
| "eval_runtime": 1.461, | |
| "eval_samples_per_second": 596.848, | |
| "eval_steps_per_second": 4.791, | |
| "step": 1054 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 5.032501697540283, | |
| "learning_rate": 3.5852782657263795e-05, | |
| "loss": 0.2271, | |
| "step": 1581 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9208715596330275, | |
| "eval_loss": 0.42918458580970764, | |
| "eval_runtime": 1.4441, | |
| "eval_samples_per_second": 603.85, | |
| "eval_steps_per_second": 4.847, | |
| "step": 1581 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 5.360136985778809, | |
| "learning_rate": 2.868494636728731e-05, | |
| "loss": 0.176, | |
| "step": 2108 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9208715596330275, | |
| "eval_loss": 0.4744165539741516, | |
| "eval_runtime": 1.4509, | |
| "eval_samples_per_second": 601.02, | |
| "eval_steps_per_second": 4.825, | |
| "step": 2108 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 11.912830352783203, | |
| "learning_rate": 2.1517110077310823e-05, | |
| "loss": 0.1486, | |
| "step": 2635 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9185779816513762, | |
| "eval_loss": 0.4420042037963867, | |
| "eval_runtime": 1.4442, | |
| "eval_samples_per_second": 603.815, | |
| "eval_steps_per_second": 4.847, | |
| "step": 2635 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 6.889047622680664, | |
| "learning_rate": 1.4349273787334336e-05, | |
| "loss": 0.1268, | |
| "step": 3162 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9277522935779816, | |
| "eval_loss": 0.4492184817790985, | |
| "eval_runtime": 1.4504, | |
| "eval_samples_per_second": 601.225, | |
| "eval_steps_per_second": 4.826, | |
| "step": 3162 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 4216, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 8, | |
| "save_steps": 500, | |
| "total_flos": 1123721737396596.0, | |
| "train_batch_size": 128, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.9346338538228292, | |
| "learning_rate": 5.73426903198119e-05, | |
| "num_train_epochs": 8, | |
| "temperature": 23 | |
| } | |
| } | |