| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 1590, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5616129032258065, | |
| "eval_loss": 0.20086993277072906, | |
| "eval_runtime": 5.6351, | |
| "eval_samples_per_second": 550.121, | |
| "eval_steps_per_second": 11.535, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.5723270440251573, | |
| "grad_norm": 0.5363451838493347, | |
| "learning_rate": 1.371069182389937e-05, | |
| "loss": 0.3151, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8009677419354839, | |
| "eval_loss": 0.10518894344568253, | |
| "eval_runtime": 6.034, | |
| "eval_samples_per_second": 513.759, | |
| "eval_steps_per_second": 10.772, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8529032258064516, | |
| "eval_loss": 0.07637829333543777, | |
| "eval_runtime": 5.7334, | |
| "eval_samples_per_second": 540.689, | |
| "eval_steps_per_second": 11.337, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.1446540880503147, | |
| "grad_norm": 0.4681667983531952, | |
| "learning_rate": 7.421383647798742e-06, | |
| "loss": 0.1211, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8761290322580645, | |
| "eval_loss": 0.06463249772787094, | |
| "eval_runtime": 6.1819, | |
| "eval_samples_per_second": 501.467, | |
| "eval_steps_per_second": 10.515, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.716981132075472, | |
| "grad_norm": 0.34897753596305847, | |
| "learning_rate": 1.1320754716981133e-06, | |
| "loss": 0.0878, | |
| "step": 1500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1590, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 391368939443328.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.38686283208351424, | |
| "num_train_epochs": 5, | |
| "temperature": 17 | |
| } | |
| } | |