{ | |
"best_metric": null, | |
"best_model_checkpoint": null, | |
"epoch": 2.0, | |
"eval_steps": 500, | |
"global_step": 504, | |
"is_hyper_param_search": true, | |
"is_local_process_zero": true, | |
"is_world_process_zero": true, | |
"log_history": [ | |
{ | |
"epoch": 1.98, | |
"learning_rate": 4.7901960965007784e-08, | |
"loss": 0.8118, | |
"step": 500 | |
} | |
], | |
"logging_steps": 500, | |
"max_steps": 504, | |
"num_train_epochs": 2, | |
"save_steps": 500, | |
"total_flos": 1058499625462272.0, | |
"trial_name": null, | |
"trial_params": { | |
"learning_rate": 6.035647081590981e-06, | |
"per_device_train_batch_size": 64 | |
} | |
} | |