|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 2.0, |
|
"global_step": 5972, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.0, |
|
"learning_rate": 2.9994976557267246e-05, |
|
"loss": 0.0444, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 0.33, |
|
"learning_rate": 2.500167448091092e-05, |
|
"loss": 0.01, |
|
"step": 995 |
|
}, |
|
{ |
|
"epoch": 0.33, |
|
"eval_loss": 0.01052522286772728, |
|
"eval_runtime": 579.4194, |
|
"eval_samples_per_second": 88.345, |
|
"eval_steps_per_second": 1.381, |
|
"step": 995 |
|
}, |
|
{ |
|
"epoch": 0.67, |
|
"learning_rate": 2.0003348961821836e-05, |
|
"loss": 0.0087, |
|
"step": 1990 |
|
}, |
|
{ |
|
"epoch": 0.67, |
|
"eval_loss": 0.010309631004929543, |
|
"eval_runtime": 580.7641, |
|
"eval_samples_per_second": 88.141, |
|
"eval_steps_per_second": 1.377, |
|
"step": 1990 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"learning_rate": 1.5005023442732755e-05, |
|
"loss": 0.0084, |
|
"step": 2985 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_loss": 0.01057281345129013, |
|
"eval_runtime": 580.4744, |
|
"eval_samples_per_second": 88.185, |
|
"eval_steps_per_second": 1.378, |
|
"step": 2985 |
|
}, |
|
{ |
|
"epoch": 1.33, |
|
"learning_rate": 1.000669792364367e-05, |
|
"loss": 0.0074, |
|
"step": 3980 |
|
}, |
|
{ |
|
"epoch": 1.33, |
|
"eval_loss": 0.01067323051393032, |
|
"eval_runtime": 580.3889, |
|
"eval_samples_per_second": 88.198, |
|
"eval_steps_per_second": 1.378, |
|
"step": 3980 |
|
}, |
|
{ |
|
"epoch": 1.67, |
|
"learning_rate": 5.008372404554588e-06, |
|
"loss": 0.0073, |
|
"step": 4975 |
|
}, |
|
{ |
|
"epoch": 1.67, |
|
"eval_loss": 0.01240999810397625, |
|
"eval_runtime": 579.3778, |
|
"eval_samples_per_second": 88.352, |
|
"eval_steps_per_second": 1.381, |
|
"step": 4975 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"learning_rate": 1.0046885465505694e-08, |
|
"loss": 0.007, |
|
"step": 5970 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_loss": 0.01129997055977583, |
|
"eval_runtime": 579.8667, |
|
"eval_samples_per_second": 88.277, |
|
"eval_steps_per_second": 1.38, |
|
"step": 5970 |
|
} |
|
], |
|
"max_steps": 5972, |
|
"num_train_epochs": 2, |
|
"total_flos": 2.01113970314453e+17, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|