|
{ |
|
"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": 10.547944577355347, |
|
"learning_rate": 6.666666666666667e-06, |
|
"loss": 1.7384, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 0.71, |
|
"grad_norm": 13.625247977537818, |
|
"learning_rate": 1.9396926207859085e-05, |
|
"loss": 1.7589, |
|
"step": 5 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_loss": 1.7180463075637817, |
|
"eval_runtime": 1.7598, |
|
"eval_samples_per_second": 7.387, |
|
"eval_steps_per_second": 0.568, |
|
"step": 7 |
|
}, |
|
{ |
|
"epoch": 1.43, |
|
"grad_norm": 5.744275140398176, |
|
"learning_rate": 1.342020143325669e-05, |
|
"loss": 1.4041, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_loss": 1.763495683670044, |
|
"eval_runtime": 1.6856, |
|
"eval_samples_per_second": 7.712, |
|
"eval_steps_per_second": 0.593, |
|
"step": 14 |
|
}, |
|
{ |
|
"epoch": 2.14, |
|
"grad_norm": 7.096816156714056, |
|
"learning_rate": 5.000000000000003e-06, |
|
"loss": 1.0218, |
|
"step": 15 |
|
}, |
|
{ |
|
"epoch": 2.86, |
|
"grad_norm": 4.77063469960506, |
|
"learning_rate": 1.519224698779198e-07, |
|
"loss": 0.6946, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_loss": 1.8394839763641357, |
|
"eval_runtime": 1.7001, |
|
"eval_samples_per_second": 7.647, |
|
"eval_steps_per_second": 0.588, |
|
"step": 21 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"step": 21, |
|
"total_flos": 4344627855360.0, |
|
"train_loss": 1.1876622920944577, |
|
"train_runtime": 427.6941, |
|
"train_samples_per_second": 3.065, |
|
"train_steps_per_second": 0.049 |
|
} |
|
], |
|
"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 |
|
} |
|
|