| { | |
| "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": 0.5741935483870968, | |
| "eval_loss": 0.21353885531425476, | |
| "eval_runtime": 5.4329, | |
| "eval_samples_per_second": 570.601, | |
| "eval_steps_per_second": 11.964, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.5723270440251573, | |
| "grad_norm": 0.553505003452301, | |
| "learning_rate": 1.371069182389937e-05, | |
| "loss": 0.3345, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8087096774193548, | |
| "eval_loss": 0.10993640869855881, | |
| "eval_runtime": 6.0275, | |
| "eval_samples_per_second": 514.313, | |
| "eval_steps_per_second": 10.784, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8596774193548387, | |
| "eval_loss": 0.0786653459072113, | |
| "eval_runtime": 5.5267, | |
| "eval_samples_per_second": 560.911, | |
| "eval_steps_per_second": 11.761, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.1446540880503147, | |
| "grad_norm": 0.48568108677864075, | |
| "learning_rate": 7.421383647798742e-06, | |
| "loss": 0.1269, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8790322580645161, | |
| "eval_loss": 0.06603369116783142, | |
| "eval_runtime": 5.5422, | |
| "eval_samples_per_second": 559.348, | |
| "eval_steps_per_second": 11.728, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.716981132075472, | |
| "grad_norm": 0.3628057539463043, | |
| "learning_rate": 1.1320754716981133e-06, | |
| "loss": 0.0909, | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 391368939443328.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.5191171844972077, | |
| "num_train_epochs": 5, | |
| "temperature": 9 | |
| } | |
| } | |