| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.716981132075472, | |
| "eval_steps": 500, | |
| "global_step": 1500, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6445161290322581, | |
| "eval_loss": 0.29249975085258484, | |
| "eval_runtime": 5.1868, | |
| "eval_samples_per_second": 597.67, | |
| "eval_steps_per_second": 12.532, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.5723270440251573, | |
| "grad_norm": 0.6633772850036621, | |
| "learning_rate": 1.550763701707098e-05, | |
| "loss": 0.4626, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8380645161290322, | |
| "eval_loss": 0.12248263508081436, | |
| "eval_runtime": 5.5326, | |
| "eval_samples_per_second": 560.313, | |
| "eval_steps_per_second": 11.749, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8887096774193548, | |
| "eval_loss": 0.07343784719705582, | |
| "eval_runtime": 5.2077, | |
| "eval_samples_per_second": 595.269, | |
| "eval_steps_per_second": 12.481, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.1446540880503147, | |
| "grad_norm": 0.59869384765625, | |
| "learning_rate": 1.101527403414196e-05, | |
| "loss": 0.1438, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9045161290322581, | |
| "eval_loss": 0.05444410815834999, | |
| "eval_runtime": 5.2524, | |
| "eval_samples_per_second": 590.204, | |
| "eval_steps_per_second": 12.375, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.716981132075472, | |
| "grad_norm": 0.4075869023799896, | |
| "learning_rate": 6.522911051212939e-06, | |
| "loss": 0.0845, | |
| "step": 1500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2226, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 723386288618448.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.08526764417742227, | |
| "num_train_epochs": 7, | |
| "temperature": 3 | |
| } | |
| } | |