| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 8.333333333333334, | |
| "eval_steps": 25, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 2.0833333333333335, | |
| "grad_norm": 0.33437681198120117, | |
| "learning_rate": 0.00015151515151515152, | |
| "loss": 2.0496, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.0833333333333335, | |
| "eval_loss": 1.1585125923156738, | |
| "eval_runtime": 1.2434, | |
| "eval_samples_per_second": 9.651, | |
| "eval_steps_per_second": 1.608, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 4.166666666666667, | |
| "grad_norm": 0.3806462287902832, | |
| "learning_rate": 0.00010101010101010102, | |
| "loss": 1.0128, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.166666666666667, | |
| "eval_loss": 0.8326404094696045, | |
| "eval_runtime": 1.129, | |
| "eval_samples_per_second": 10.629, | |
| "eval_steps_per_second": 1.772, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "grad_norm": 0.427622526884079, | |
| "learning_rate": 5.050505050505051e-05, | |
| "loss": 0.733, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "eval_loss": 0.7211847901344299, | |
| "eval_runtime": 1.1314, | |
| "eval_samples_per_second": 10.606, | |
| "eval_steps_per_second": 1.768, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 8.333333333333334, | |
| "grad_norm": 0.4516367018222809, | |
| "learning_rate": 0.0, | |
| "loss": 0.6339, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 8.333333333333334, | |
| "eval_loss": 0.6972503662109375, | |
| "eval_runtime": 1.1255, | |
| "eval_samples_per_second": 10.662, | |
| "eval_steps_per_second": 1.777, | |
| "step": 100 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 263984573337600.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |