| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 7.861635220125786, | |
| "eval_steps": 500, | |
| "global_step": 2500, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6680645161290323, | |
| "eval_loss": 0.4126000702381134, | |
| "eval_runtime": 5.3541, | |
| "eval_samples_per_second": 578.999, | |
| "eval_steps_per_second": 12.14, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.5723270440251573, | |
| "grad_norm": 0.9286167621612549, | |
| "learning_rate": 1.606918238993711e-05, | |
| "loss": 0.6457, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8387096774193549, | |
| "eval_loss": 0.14377787709236145, | |
| "eval_runtime": 5.7959, | |
| "eval_samples_per_second": 534.861, | |
| "eval_steps_per_second": 11.215, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8964516129032258, | |
| "eval_loss": 0.0756388008594513, | |
| "eval_runtime": 5.5995, | |
| "eval_samples_per_second": 553.618, | |
| "eval_steps_per_second": 11.608, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.1446540880503147, | |
| "grad_norm": 0.782314121723175, | |
| "learning_rate": 1.2138364779874214e-05, | |
| "loss": 0.1679, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9119354838709678, | |
| "eval_loss": 0.054538920521736145, | |
| "eval_runtime": 5.4328, | |
| "eval_samples_per_second": 570.605, | |
| "eval_steps_per_second": 11.964, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.716981132075472, | |
| "grad_norm": 0.44344785809516907, | |
| "learning_rate": 8.207547169811321e-06, | |
| "loss": 0.0851, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9258064516129032, | |
| "eval_loss": 0.044873543083667755, | |
| "eval_runtime": 5.5592, | |
| "eval_samples_per_second": 557.632, | |
| "eval_steps_per_second": 11.692, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.927741935483871, | |
| "eval_loss": 0.04030356556177139, | |
| "eval_runtime": 5.6057, | |
| "eval_samples_per_second": 553.01, | |
| "eval_steps_per_second": 11.595, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.289308176100629, | |
| "grad_norm": 0.3489580750465393, | |
| "learning_rate": 4.276729559748428e-06, | |
| "loss": 0.065, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.927741935483871, | |
| "eval_loss": 0.03819738328456879, | |
| "eval_runtime": 5.427, | |
| "eval_samples_per_second": 571.223, | |
| "eval_steps_per_second": 11.977, | |
| "step": 2226 | |
| }, | |
| { | |
| "epoch": 7.861635220125786, | |
| "grad_norm": 0.3018985688686371, | |
| "learning_rate": 3.459119496855346e-07, | |
| "loss": 0.058, | |
| "step": 2500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2544, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 8, | |
| "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": 853150315276476.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.0171008380469381, | |
| "num_train_epochs": 8, | |
| "temperature": 2 | |
| } | |
| } | |