| { |
| "best_metric": 0.16112907230854034, |
| "best_model_checkpoint": "YELP_full/checkpoint-393", |
| "epoch": 5.0, |
| "eval_steps": 500, |
| "global_step": 655, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.9328, |
| "eval_loss": 0.17968642711639404, |
| "eval_runtime": 78.2206, |
| "eval_samples_per_second": 319.609, |
| "eval_steps_per_second": 1.675, |
| "step": 131 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.93844, |
| "eval_loss": 0.1635778844356537, |
| "eval_runtime": 78.3918, |
| "eval_samples_per_second": 318.911, |
| "eval_steps_per_second": 1.671, |
| "step": 262 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.9432, |
| "eval_loss": 0.16112907230854034, |
| "eval_runtime": 78.1143, |
| "eval_samples_per_second": 320.044, |
| "eval_steps_per_second": 1.677, |
| "step": 393 |
| }, |
| { |
| "epoch": 3.82, |
| "grad_norm": 4.371647834777832, |
| "learning_rate": 3.7099236641221374e-05, |
| "loss": 0.1759, |
| "step": 500 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.94228, |
| "eval_loss": 0.18924400210380554, |
| "eval_runtime": 78.0582, |
| "eval_samples_per_second": 320.274, |
| "eval_steps_per_second": 1.678, |
| "step": 524 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.94408, |
| "eval_loss": 0.17596399784088135, |
| "eval_runtime": 78.0645, |
| "eval_samples_per_second": 320.248, |
| "eval_steps_per_second": 1.678, |
| "step": 655 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 1310, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 10, |
| "save_steps": 500, |
| "total_flos": 6.5276215197696e+16, |
| "train_batch_size": 192, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
