| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.9964267483409905, | |
| "eval_steps": 500, | |
| "global_step": 1220, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.9964267483409903, | |
| "eval_loss": 0.6944473385810852, | |
| "eval_runtime": 114.7819, | |
| "eval_samples_per_second": 7.188, | |
| "eval_steps_per_second": 0.906, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 1.0040837161817253, | |
| "grad_norm": 90.0, | |
| "learning_rate": 1.7553956834532375e-05, | |
| "loss": 13.1822, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 1.9964267483409903, | |
| "eval_loss": 0.688672661781311, | |
| "eval_runtime": 83.9218, | |
| "eval_samples_per_second": 9.831, | |
| "eval_steps_per_second": 1.239, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 2.0081674323634506, | |
| "grad_norm": 40.5, | |
| "learning_rate": 1.4028776978417267e-05, | |
| "loss": 11.0456, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.9964267483409905, | |
| "eval_loss": 0.7161328196525574, | |
| "eval_runtime": 85.1515, | |
| "eval_samples_per_second": 9.689, | |
| "eval_steps_per_second": 1.221, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 3.012251148545176, | |
| "grad_norm": 77.0, | |
| "learning_rate": 1.0503597122302158e-05, | |
| "loss": 8.2545, | |
| "step": 735 | |
| }, | |
| { | |
| "epoch": 3.9964267483409905, | |
| "eval_loss": 1.2752673625946045, | |
| "eval_runtime": 83.1659, | |
| "eval_samples_per_second": 9.92, | |
| "eval_steps_per_second": 1.251, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 4.016334864726901, | |
| "grad_norm": 23.25, | |
| "learning_rate": 6.978417266187051e-06, | |
| "loss": 1.473, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 4.9964267483409905, | |
| "eval_loss": 2.4596149921417236, | |
| "eval_runtime": 98.7355, | |
| "eval_samples_per_second": 8.356, | |
| "eval_steps_per_second": 1.053, | |
| "step": 1220 | |
| } | |
| ], | |
| "logging_steps": 245, | |
| "max_steps": 1464, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 6, | |
| "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": 1.1909651086511596e+18, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |