|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 10.0, |
|
"eval_steps": 500, |
|
"global_step": 2850, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 1.0, |
|
"eval_bleu": 0.03086976985092356, |
|
"eval_loss": 4.82624626159668, |
|
"eval_runtime": 97.3301, |
|
"eval_samples_per_second": 11.682, |
|
"eval_steps_per_second": 0.74, |
|
"step": 285 |
|
}, |
|
{ |
|
"epoch": 1.7543859649122808, |
|
"grad_norm": 0.6388371586799622, |
|
"learning_rate": 1.9840000000000003e-05, |
|
"loss": 5.7131, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_bleu": 1.2576014507933267e-08, |
|
"eval_loss": 1.6568931341171265, |
|
"eval_runtime": 344.4207, |
|
"eval_samples_per_second": 3.301, |
|
"eval_steps_per_second": 0.209, |
|
"step": 570 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_bleu": 0.0, |
|
"eval_loss": 1.6353662014007568, |
|
"eval_runtime": 344.9175, |
|
"eval_samples_per_second": 3.296, |
|
"eval_steps_per_second": 0.209, |
|
"step": 855 |
|
}, |
|
{ |
|
"epoch": 3.5087719298245617, |
|
"grad_norm": 0.41950201988220215, |
|
"learning_rate": 1.5778723404255322e-05, |
|
"loss": 1.6327, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_bleu": 0.028380647856164357, |
|
"eval_loss": 1.6304912567138672, |
|
"eval_runtime": 341.8623, |
|
"eval_samples_per_second": 3.326, |
|
"eval_steps_per_second": 0.211, |
|
"step": 1140 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"eval_bleu": 2.3806283652785307, |
|
"eval_loss": 1.6316522359848022, |
|
"eval_runtime": 316.8037, |
|
"eval_samples_per_second": 3.589, |
|
"eval_steps_per_second": 0.227, |
|
"step": 1425 |
|
}, |
|
{ |
|
"epoch": 5.2631578947368425, |
|
"grad_norm": 0.46960917115211487, |
|
"learning_rate": 1.152340425531915e-05, |
|
"loss": 1.5867, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"eval_bleu": 5.280724757906034, |
|
"eval_loss": 1.632155179977417, |
|
"eval_runtime": 173.2701, |
|
"eval_samples_per_second": 6.562, |
|
"eval_steps_per_second": 0.416, |
|
"step": 1710 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"eval_bleu": 5.34771509760483, |
|
"eval_loss": 1.6327825784683228, |
|
"eval_runtime": 237.2864, |
|
"eval_samples_per_second": 4.792, |
|
"eval_steps_per_second": 0.303, |
|
"step": 1995 |
|
}, |
|
{ |
|
"epoch": 7.017543859649122, |
|
"grad_norm": 0.33166804909706116, |
|
"learning_rate": 7.268085106382979e-06, |
|
"loss": 1.5654, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"eval_bleu": 4.833246227017286, |
|
"eval_loss": 1.634158730506897, |
|
"eval_runtime": 173.1237, |
|
"eval_samples_per_second": 6.568, |
|
"eval_steps_per_second": 0.416, |
|
"step": 2280 |
|
}, |
|
{ |
|
"epoch": 8.771929824561404, |
|
"grad_norm": 0.2613541781902313, |
|
"learning_rate": 3.012765957446809e-06, |
|
"loss": 1.5528, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 9.0, |
|
"eval_bleu": 5.348525572887838, |
|
"eval_loss": 1.6349856853485107, |
|
"eval_runtime": 140.3041, |
|
"eval_samples_per_second": 8.104, |
|
"eval_steps_per_second": 0.513, |
|
"step": 2565 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"eval_bleu": 5.1657380844985274, |
|
"eval_loss": 1.6354198455810547, |
|
"eval_runtime": 113.6949, |
|
"eval_samples_per_second": 10.0, |
|
"eval_steps_per_second": 0.633, |
|
"step": 2850 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 2850, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 10, |
|
"save_steps": 500, |
|
"stateful_callbacks": { |
|
"TrainerControl": { |
|
"args": { |
|
"should_epoch_stop": false, |
|
"should_evaluate": false, |
|
"should_log": false, |
|
"should_save": true, |
|
"should_training_stop": true |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 2.463997932601344e+16, |
|
"train_batch_size": 16, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|