{ | |
"best_metric": null, | |
"best_model_checkpoint": null, | |
"epoch": 4.9754266821553035, | |
"global_step": 870, | |
"is_hyper_param_search": false, | |
"is_local_process_zero": true, | |
"is_world_process_zero": true, | |
"log_history": [ | |
{ | |
"epoch": 0.01, | |
"learning_rate": 0, | |
"loss": 2.2973, | |
"step": 1 | |
}, | |
{ | |
"epoch": 2.86, | |
"learning_rate": 5e-05, | |
"loss": 1.0091, | |
"step": 500 | |
} | |
], | |
"max_steps": 870, | |
"num_train_epochs": 5, | |
"total_flos": 7313911250419712.0, | |
"trial_name": null, | |
"trial_params": null | |
} | |