|
{ |
|
"best_metric": 23.047330380041966, |
|
"best_model_checkpoint": "./whisper-small-clean_6-v5/checkpoint-500", |
|
"epoch": 0.945179584120983, |
|
"eval_steps": 500, |
|
"global_step": 500, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.04725897920604915, |
|
"grad_norm": 4.044429302215576, |
|
"learning_rate": 4.4e-06, |
|
"loss": 0.6829, |
|
"step": 25 |
|
}, |
|
{ |
|
"epoch": 0.0945179584120983, |
|
"grad_norm": 4.314411640167236, |
|
"learning_rate": 9.4e-06, |
|
"loss": 0.2473, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.14177693761814744, |
|
"grad_norm": 3.5965349674224854, |
|
"learning_rate": 9.848275862068966e-06, |
|
"loss": 0.201, |
|
"step": 75 |
|
}, |
|
{ |
|
"epoch": 0.1890359168241966, |
|
"grad_norm": 3.4847946166992188, |
|
"learning_rate": 9.675862068965518e-06, |
|
"loss": 0.1812, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.23629489603024575, |
|
"grad_norm": 3.8979456424713135, |
|
"learning_rate": 9.503448275862069e-06, |
|
"loss": 0.1815, |
|
"step": 125 |
|
}, |
|
{ |
|
"epoch": 0.2835538752362949, |
|
"grad_norm": 4.511611461639404, |
|
"learning_rate": 9.331034482758623e-06, |
|
"loss": 0.1789, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 0.33081285444234404, |
|
"grad_norm": 4.727020263671875, |
|
"learning_rate": 9.158620689655173e-06, |
|
"loss": 0.1709, |
|
"step": 175 |
|
}, |
|
{ |
|
"epoch": 0.3780718336483932, |
|
"grad_norm": 3.503211259841919, |
|
"learning_rate": 8.986206896551725e-06, |
|
"loss": 0.1761, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.42533081285444235, |
|
"grad_norm": 3.376284599304199, |
|
"learning_rate": 8.813793103448277e-06, |
|
"loss": 0.1673, |
|
"step": 225 |
|
}, |
|
{ |
|
"epoch": 0.4725897920604915, |
|
"grad_norm": 4.138789176940918, |
|
"learning_rate": 8.641379310344828e-06, |
|
"loss": 0.172, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 0.5198487712665406, |
|
"grad_norm": 3.522365093231201, |
|
"learning_rate": 8.46896551724138e-06, |
|
"loss": 0.1627, |
|
"step": 275 |
|
}, |
|
{ |
|
"epoch": 0.5671077504725898, |
|
"grad_norm": 2.7766571044921875, |
|
"learning_rate": 8.296551724137932e-06, |
|
"loss": 0.1605, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.6143667296786389, |
|
"grad_norm": 3.5490145683288574, |
|
"learning_rate": 8.124137931034484e-06, |
|
"loss": 0.1574, |
|
"step": 325 |
|
}, |
|
{ |
|
"epoch": 0.6616257088846881, |
|
"grad_norm": 3.641836643218994, |
|
"learning_rate": 7.951724137931034e-06, |
|
"loss": 0.1637, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 0.7088846880907372, |
|
"grad_norm": 3.28399920463562, |
|
"learning_rate": 7.779310344827586e-06, |
|
"loss": 0.156, |
|
"step": 375 |
|
}, |
|
{ |
|
"epoch": 0.7561436672967864, |
|
"grad_norm": 3.2820281982421875, |
|
"learning_rate": 7.6068965517241385e-06, |
|
"loss": 0.1578, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.8034026465028355, |
|
"grad_norm": 3.589008092880249, |
|
"learning_rate": 7.4344827586206906e-06, |
|
"loss": 0.1573, |
|
"step": 425 |
|
}, |
|
{ |
|
"epoch": 0.8506616257088847, |
|
"grad_norm": 3.9363508224487305, |
|
"learning_rate": 7.262068965517242e-06, |
|
"loss": 0.1527, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 0.8979206049149339, |
|
"grad_norm": 3.573969841003418, |
|
"learning_rate": 7.089655172413794e-06, |
|
"loss": 0.1561, |
|
"step": 475 |
|
}, |
|
{ |
|
"epoch": 0.945179584120983, |
|
"grad_norm": 3.35899019241333, |
|
"learning_rate": 6.917241379310345e-06, |
|
"loss": 0.1461, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.945179584120983, |
|
"eval_loss": 0.27461186051368713, |
|
"eval_runtime": 652.5217, |
|
"eval_samples_per_second": 2.337, |
|
"eval_steps_per_second": 0.074, |
|
"eval_wer": 23.047330380041966, |
|
"step": 500 |
|
} |
|
], |
|
"logging_steps": 25, |
|
"max_steps": 1500, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 3, |
|
"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": 9.23473281024e+18, |
|
"train_batch_size": 64, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|