|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 0.0008078196946441554, |
|
"eval_steps": 3, |
|
"global_step": 10, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 8.078196946441554e-05, |
|
"grad_norm": 10.836163520812988, |
|
"learning_rate": 2e-05, |
|
"loss": 5.2973, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 8.078196946441554e-05, |
|
"eval_loss": 6.08513069152832, |
|
"eval_runtime": 190.4047, |
|
"eval_samples_per_second": 27.379, |
|
"eval_steps_per_second": 13.692, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 0.00016156393892883107, |
|
"grad_norm": 15.810860633850098, |
|
"learning_rate": 4e-05, |
|
"loss": 6.7627, |
|
"step": 2 |
|
}, |
|
{ |
|
"epoch": 0.00024234590839324662, |
|
"grad_norm": 11.094259262084961, |
|
"learning_rate": 6e-05, |
|
"loss": 5.5142, |
|
"step": 3 |
|
}, |
|
{ |
|
"epoch": 0.00024234590839324662, |
|
"eval_loss": 5.897503852844238, |
|
"eval_runtime": 185.1619, |
|
"eval_samples_per_second": 28.154, |
|
"eval_steps_per_second": 14.08, |
|
"step": 3 |
|
}, |
|
{ |
|
"epoch": 0.00032312787785766214, |
|
"grad_norm": 13.849967956542969, |
|
"learning_rate": 8e-05, |
|
"loss": 6.0094, |
|
"step": 4 |
|
}, |
|
{ |
|
"epoch": 0.0004039098473220777, |
|
"grad_norm": 11.20494556427002, |
|
"learning_rate": 0.0001, |
|
"loss": 5.0195, |
|
"step": 5 |
|
}, |
|
{ |
|
"epoch": 0.00048469181678649324, |
|
"grad_norm": 10.72175407409668, |
|
"learning_rate": 0.00012, |
|
"loss": 4.6667, |
|
"step": 6 |
|
}, |
|
{ |
|
"epoch": 0.00048469181678649324, |
|
"eval_loss": 3.9255988597869873, |
|
"eval_runtime": 186.96, |
|
"eval_samples_per_second": 27.883, |
|
"eval_steps_per_second": 13.944, |
|
"step": 6 |
|
}, |
|
{ |
|
"epoch": 0.0005654737862509088, |
|
"grad_norm": 9.823970794677734, |
|
"learning_rate": 0.00014, |
|
"loss": 3.7467, |
|
"step": 7 |
|
}, |
|
{ |
|
"epoch": 0.0006462557557153243, |
|
"grad_norm": 9.715559959411621, |
|
"learning_rate": 0.00016, |
|
"loss": 3.0583, |
|
"step": 8 |
|
}, |
|
{ |
|
"epoch": 0.0007270377251797399, |
|
"grad_norm": 9.468438148498535, |
|
"learning_rate": 0.00018, |
|
"loss": 2.6625, |
|
"step": 9 |
|
}, |
|
{ |
|
"epoch": 0.0007270377251797399, |
|
"eval_loss": 1.930371642112732, |
|
"eval_runtime": 84.7272, |
|
"eval_samples_per_second": 61.527, |
|
"eval_steps_per_second": 30.769, |
|
"step": 9 |
|
}, |
|
{ |
|
"epoch": 0.0008078196946441554, |
|
"grad_norm": 9.221755981445312, |
|
"learning_rate": 0.0002, |
|
"loss": 2.0402, |
|
"step": 10 |
|
} |
|
], |
|
"logging_steps": 1, |
|
"max_steps": 10, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 1, |
|
"save_steps": 3, |
|
"stateful_callbacks": { |
|
"TrainerControl": { |
|
"args": { |
|
"should_epoch_stop": false, |
|
"should_evaluate": false, |
|
"should_log": false, |
|
"should_save": true, |
|
"should_training_stop": true |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 489477793382400.0, |
|
"train_batch_size": 2, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|