|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 2.834734952281962, |
|
"eval_steps": 500, |
|
"global_step": 30000, |
|
"is_hyper_param_search": true, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 1.42, |
|
"learning_rate": 5.403141605002034e-06, |
|
"loss": 0.9472, |
|
"step": 15000 |
|
}, |
|
{ |
|
"epoch": 2.83, |
|
"learning_rate": 5.642184409307157e-07, |
|
"loss": 0.5511, |
|
"step": 30000 |
|
} |
|
], |
|
"logging_steps": 15000, |
|
"max_steps": 31749, |
|
"num_train_epochs": 3, |
|
"save_steps": 500, |
|
"total_flos": 216724668715296.0, |
|
"trial_name": null, |
|
"trial_params": { |
|
"learning_rate": 1.0242064769073352e-05, |
|
"per_device_train_batch_size": 1 |
|
} |
|
} |
|
|