|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 3.0, |
|
"eval_steps": 500, |
|
"global_step": 21, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.14, |
|
"grad_norm": 1.3479268895743872, |
|
"learning_rate": 6.666666666666667e-06, |
|
"loss": 1.1957, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 0.71, |
|
"grad_norm": 2.1971208912415268, |
|
"learning_rate": 1.9396926207859085e-05, |
|
"loss": 1.1705, |
|
"step": 5 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_loss": 1.4559755325317383, |
|
"eval_runtime": 1.3739, |
|
"eval_samples_per_second": 5.823, |
|
"eval_steps_per_second": 0.728, |
|
"step": 7 |
|
}, |
|
{ |
|
"epoch": 1.43, |
|
"grad_norm": 0.9052399938436037, |
|
"learning_rate": 1.342020143325669e-05, |
|
"loss": 1.0304, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_loss": 1.4817352294921875, |
|
"eval_runtime": 1.4072, |
|
"eval_samples_per_second": 5.685, |
|
"eval_steps_per_second": 0.711, |
|
"step": 14 |
|
}, |
|
{ |
|
"epoch": 2.14, |
|
"grad_norm": 0.846886971725444, |
|
"learning_rate": 5.000000000000003e-06, |
|
"loss": 0.8943, |
|
"step": 15 |
|
}, |
|
{ |
|
"epoch": 2.86, |
|
"grad_norm": 0.6182004133132336, |
|
"learning_rate": 1.519224698779198e-07, |
|
"loss": 0.7906, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_loss": 1.4959019422531128, |
|
"eval_runtime": 1.4033, |
|
"eval_samples_per_second": 5.701, |
|
"eval_steps_per_second": 0.713, |
|
"step": 21 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"step": 21, |
|
"total_flos": 4344627855360.0, |
|
"train_loss": 0.9623228907585144, |
|
"train_runtime": 352.7017, |
|
"train_samples_per_second": 3.7, |
|
"train_steps_per_second": 0.06 |
|
} |
|
], |
|
"logging_steps": 5, |
|
"max_steps": 21, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 3, |
|
"save_steps": 500, |
|
"total_flos": 4344627855360.0, |
|
"train_batch_size": 4, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|