|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 14.634146341463415, |
|
"eval_steps": 500, |
|
"global_step": 150, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.98, |
|
"learning_rate": 4.9863047384206835e-05, |
|
"loss": 2.7294, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 1.95, |
|
"learning_rate": 4.9453690018345144e-05, |
|
"loss": 3.5475, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 2.93, |
|
"learning_rate": 4.9453690018345144e-05, |
|
"loss": 5.3042, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 3.9, |
|
"learning_rate": 4.9214579028215776e-05, |
|
"loss": 6.5608, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 4.88, |
|
"learning_rate": 4.9214579028215776e-05, |
|
"loss": 12.436, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 5.85, |
|
"learning_rate": 4.9214579028215776e-05, |
|
"loss": 8.1905, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 6.83, |
|
"learning_rate": 4.9214579028215776e-05, |
|
"loss": 5.0973, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 7.8, |
|
"learning_rate": 4.9214579028215776e-05, |
|
"loss": 3.3543, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 8.78, |
|
"learning_rate": 4.893298743830168e-05, |
|
"loss": 9.9021, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 9.76, |
|
"learning_rate": 4.893298743830168e-05, |
|
"loss": 16.1257, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 10.73, |
|
"learning_rate": 4.893298743830168e-05, |
|
"loss": 15.988, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 11.71, |
|
"learning_rate": 4.877641290737884e-05, |
|
"loss": 5.743, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 12.68, |
|
"learning_rate": 4.860940925593703e-05, |
|
"loss": 4.5123, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 13.66, |
|
"learning_rate": 4.860940925593703e-05, |
|
"loss": 8.7191, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 14.63, |
|
"learning_rate": 4.860940925593703e-05, |
|
"loss": 8.9813, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 14.63, |
|
"step": 150, |
|
"total_flos": 3.550292723048448e+16, |
|
"train_loss": 7.812765197753906, |
|
"train_runtime": 716.3495, |
|
"train_samples_per_second": 3.413, |
|
"train_steps_per_second": 0.209 |
|
} |
|
], |
|
"logging_steps": 10, |
|
"max_steps": 150, |
|
"num_train_epochs": 15, |
|
"save_steps": 1000, |
|
"total_flos": 3.550292723048448e+16, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|