|
{ |
|
"best_metric": 2.5020904541015625, |
|
"best_model_checkpoint": "./outputs/checkpoint-500", |
|
"epoch": 0.36429872495446264, |
|
"eval_steps": 100, |
|
"global_step": 500, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.07, |
|
"learning_rate": 0.0002, |
|
"loss": 2.7401, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.07, |
|
"eval_loss": 2.6424834728240967, |
|
"eval_runtime": 206.3704, |
|
"eval_samples_per_second": 30.402, |
|
"eval_steps_per_second": 3.804, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.15, |
|
"learning_rate": 0.0002, |
|
"loss": 2.6053, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.15, |
|
"eval_loss": 2.592982053756714, |
|
"eval_runtime": 206.2608, |
|
"eval_samples_per_second": 30.418, |
|
"eval_steps_per_second": 3.806, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.22, |
|
"learning_rate": 0.0002, |
|
"loss": 2.5637, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.22, |
|
"eval_loss": 2.558687448501587, |
|
"eval_runtime": 206.6302, |
|
"eval_samples_per_second": 30.363, |
|
"eval_steps_per_second": 3.799, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.29, |
|
"learning_rate": 0.0002, |
|
"loss": 2.5384, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.29, |
|
"eval_loss": 2.5261261463165283, |
|
"eval_runtime": 206.3988, |
|
"eval_samples_per_second": 30.397, |
|
"eval_steps_per_second": 3.803, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.36, |
|
"learning_rate": 0.0002, |
|
"loss": 2.4965, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.36, |
|
"eval_loss": 2.5020904541015625, |
|
"eval_runtime": 273.7006, |
|
"eval_samples_per_second": 22.923, |
|
"eval_steps_per_second": 2.868, |
|
"step": 500 |
|
} |
|
], |
|
"logging_steps": 100, |
|
"max_steps": 4116, |
|
"num_train_epochs": 3, |
|
"save_steps": 100, |
|
"total_flos": 1.459702746464256e+16, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|