|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 9.433962264150944, |
|
"eval_steps": 500, |
|
"global_step": 3000, |
|
"is_hyper_param_search": true, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.6106451612903225, |
|
"eval_loss": 0.2180573046207428, |
|
"eval_runtime": 5.2234, |
|
"eval_samples_per_second": 593.484, |
|
"eval_steps_per_second": 12.444, |
|
"step": 318 |
|
}, |
|
{ |
|
"epoch": 1.5723270440251573, |
|
"grad_norm": 0.5000836253166199, |
|
"learning_rate": 1.685534591194969e-05, |
|
"loss": 0.3508, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.8367741935483871, |
|
"eval_loss": 0.10006564110517502, |
|
"eval_runtime": 5.3217, |
|
"eval_samples_per_second": 582.516, |
|
"eval_steps_per_second": 12.214, |
|
"step": 636 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.8880645161290323, |
|
"eval_loss": 0.06387896835803986, |
|
"eval_runtime": 5.4086, |
|
"eval_samples_per_second": 573.166, |
|
"eval_steps_per_second": 12.018, |
|
"step": 954 |
|
}, |
|
{ |
|
"epoch": 3.1446540880503147, |
|
"grad_norm": 0.5663716197013855, |
|
"learning_rate": 1.371069182389937e-05, |
|
"loss": 0.1169, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.9, |
|
"eval_loss": 0.04769841209053993, |
|
"eval_runtime": 5.4917, |
|
"eval_samples_per_second": 564.492, |
|
"eval_steps_per_second": 11.836, |
|
"step": 1272 |
|
}, |
|
{ |
|
"epoch": 4.716981132075472, |
|
"grad_norm": 0.3216884136199951, |
|
"learning_rate": 1.0566037735849058e-05, |
|
"loss": 0.0714, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"eval_accuracy": 0.9170967741935484, |
|
"eval_loss": 0.0384916327893734, |
|
"eval_runtime": 5.6475, |
|
"eval_samples_per_second": 548.917, |
|
"eval_steps_per_second": 11.51, |
|
"step": 1590 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"eval_accuracy": 0.9183870967741935, |
|
"eval_loss": 0.0333557203412056, |
|
"eval_runtime": 5.3094, |
|
"eval_samples_per_second": 583.872, |
|
"eval_steps_per_second": 12.242, |
|
"step": 1908 |
|
}, |
|
{ |
|
"epoch": 6.289308176100629, |
|
"grad_norm": 0.24820531904697418, |
|
"learning_rate": 7.421383647798742e-06, |
|
"loss": 0.055, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"eval_accuracy": 0.9245161290322581, |
|
"eval_loss": 0.030584245920181274, |
|
"eval_runtime": 5.6073, |
|
"eval_samples_per_second": 552.85, |
|
"eval_steps_per_second": 11.592, |
|
"step": 2226 |
|
}, |
|
{ |
|
"epoch": 7.861635220125786, |
|
"grad_norm": 0.21355891227722168, |
|
"learning_rate": 4.276729559748428e-06, |
|
"loss": 0.0474, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"eval_accuracy": 0.9251612903225807, |
|
"eval_loss": 0.028302712365984917, |
|
"eval_runtime": 5.4604, |
|
"eval_samples_per_second": 567.72, |
|
"eval_steps_per_second": 11.904, |
|
"step": 2544 |
|
}, |
|
{ |
|
"epoch": 9.0, |
|
"eval_accuracy": 0.9270967741935484, |
|
"eval_loss": 0.027429578825831413, |
|
"eval_runtime": 5.4597, |
|
"eval_samples_per_second": 567.795, |
|
"eval_steps_per_second": 11.905, |
|
"step": 2862 |
|
}, |
|
{ |
|
"epoch": 9.433962264150944, |
|
"grad_norm": 0.22619187831878662, |
|
"learning_rate": 1.1320754716981133e-06, |
|
"loss": 0.0438, |
|
"step": 3000 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 3180, |
|
"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": false |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 780738843279612.0, |
|
"train_batch_size": 48, |
|
"trial_name": null, |
|
"trial_params": { |
|
"alpha": 0.5433913699935149, |
|
"num_train_epochs": 10, |
|
"temperature": 6 |
|
} |
|
} |
|
|