|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 5.0, |
|
"global_step": 2500, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 1.0, |
|
"learning_rate": 4e-05, |
|
"loss": 0.3063, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.9211165048543689, |
|
"eval_loss": 0.2524273991584778, |
|
"eval_runtime": 7.3979, |
|
"eval_samples_per_second": 111.383, |
|
"eval_steps_per_second": 7.029, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"learning_rate": 3e-05, |
|
"loss": 0.136, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.9405339805825242, |
|
"eval_loss": 0.21534095704555511, |
|
"eval_runtime": 7.3273, |
|
"eval_samples_per_second": 112.457, |
|
"eval_steps_per_second": 7.097, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"learning_rate": 2e-05, |
|
"loss": 0.0567, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.9599514563106796, |
|
"eval_loss": 0.2191401720046997, |
|
"eval_runtime": 7.4908, |
|
"eval_samples_per_second": 110.002, |
|
"eval_steps_per_second": 6.942, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"learning_rate": 1e-05, |
|
"loss": 0.0189, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.9526699029126213, |
|
"eval_loss": 0.2866779863834381, |
|
"eval_runtime": 7.3916, |
|
"eval_samples_per_second": 111.478, |
|
"eval_steps_per_second": 7.035, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"learning_rate": 0.0, |
|
"loss": 0.0056, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"eval_accuracy": 0.9563106796116505, |
|
"eval_loss": 0.2526571452617645, |
|
"eval_runtime": 7.4135, |
|
"eval_samples_per_second": 111.148, |
|
"eval_steps_per_second": 7.014, |
|
"step": 2500 |
|
} |
|
], |
|
"max_steps": 2500, |
|
"num_train_epochs": 5, |
|
"total_flos": 1.052453670912e+16, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|