|
{ |
|
"best_metric": 26.64601084430674, |
|
"best_model_checkpoint": "./whisper-small-clean_v6/checkpoint-400", |
|
"epoch": 1.4020000000000001, |
|
"eval_steps": 100, |
|
"global_step": 500, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.05, |
|
"grad_norm": 24.222782135009766, |
|
"learning_rate": 4.2000000000000004e-06, |
|
"loss": 1.5789, |
|
"step": 25 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"grad_norm": 3.7403714656829834, |
|
"learning_rate": 9.200000000000002e-06, |
|
"loss": 0.2766, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.15, |
|
"grad_norm": 3.9098994731903076, |
|
"learning_rate": 9.533333333333334e-06, |
|
"loss": 0.1533, |
|
"step": 75 |
|
}, |
|
{ |
|
"epoch": 0.2, |
|
"grad_norm": 3.8178775310516357, |
|
"learning_rate": 8.977777777777778e-06, |
|
"loss": 0.1354, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.2, |
|
"eval_loss": 0.4426867365837097, |
|
"eval_runtime": 177.5025, |
|
"eval_samples_per_second": 1.792, |
|
"eval_steps_per_second": 0.056, |
|
"eval_wer": 34.77924089852827, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.25, |
|
"grad_norm": 3.6577646732330322, |
|
"learning_rate": 8.422222222222223e-06, |
|
"loss": 0.1327, |
|
"step": 125 |
|
}, |
|
{ |
|
"epoch": 0.3, |
|
"grad_norm": 3.8943512439727783, |
|
"learning_rate": 7.866666666666667e-06, |
|
"loss": 0.1225, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 0.35, |
|
"grad_norm": 3.568549394607544, |
|
"learning_rate": 7.3111111111111125e-06, |
|
"loss": 0.1349, |
|
"step": 175 |
|
}, |
|
{ |
|
"epoch": 0.4, |
|
"grad_norm": 2.832570791244507, |
|
"learning_rate": 6.755555555555556e-06, |
|
"loss": 0.1115, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.4, |
|
"eval_loss": 0.3909415304660797, |
|
"eval_runtime": 151.5663, |
|
"eval_samples_per_second": 2.098, |
|
"eval_steps_per_second": 0.066, |
|
"eval_wer": 27.73044151820294, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.45, |
|
"grad_norm": 3.3633787631988525, |
|
"learning_rate": 6.200000000000001e-06, |
|
"loss": 0.1162, |
|
"step": 225 |
|
}, |
|
{ |
|
"epoch": 0.5, |
|
"grad_norm": 3.3805878162384033, |
|
"learning_rate": 5.6444444444444445e-06, |
|
"loss": 0.1205, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 0.55, |
|
"grad_norm": 3.420900821685791, |
|
"learning_rate": 5.088888888888889e-06, |
|
"loss": 0.1055, |
|
"step": 275 |
|
}, |
|
{ |
|
"epoch": 1.002, |
|
"grad_norm": 1.5796862840652466, |
|
"learning_rate": 4.533333333333334e-06, |
|
"loss": 0.1105, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 1.002, |
|
"eval_loss": 0.37623485922813416, |
|
"eval_runtime": 149.2618, |
|
"eval_samples_per_second": 2.13, |
|
"eval_steps_per_second": 0.067, |
|
"eval_wer": 27.885360185902403, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 1.052, |
|
"grad_norm": 1.0628682374954224, |
|
"learning_rate": 3.977777777777778e-06, |
|
"loss": 0.0408, |
|
"step": 325 |
|
}, |
|
{ |
|
"epoch": 1.102, |
|
"grad_norm": 1.986166000366211, |
|
"learning_rate": 3.4222222222222224e-06, |
|
"loss": 0.039, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 1.152, |
|
"grad_norm": 1.9044853448867798, |
|
"learning_rate": 2.866666666666667e-06, |
|
"loss": 0.0476, |
|
"step": 375 |
|
}, |
|
{ |
|
"epoch": 1.202, |
|
"grad_norm": 1.8975799083709717, |
|
"learning_rate": 2.311111111111111e-06, |
|
"loss": 0.0392, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 1.202, |
|
"eval_loss": 0.38561537861824036, |
|
"eval_runtime": 148.1056, |
|
"eval_samples_per_second": 2.147, |
|
"eval_steps_per_second": 0.068, |
|
"eval_wer": 26.64601084430674, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 1.252, |
|
"grad_norm": 3.421165943145752, |
|
"learning_rate": 1.7555555555555556e-06, |
|
"loss": 0.0483, |
|
"step": 425 |
|
}, |
|
{ |
|
"epoch": 1.302, |
|
"grad_norm": 1.8972320556640625, |
|
"learning_rate": 1.2000000000000002e-06, |
|
"loss": 0.0664, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 1.3519999999999999, |
|
"grad_norm": 1.9235032796859741, |
|
"learning_rate": 6.444444444444445e-07, |
|
"loss": 0.039, |
|
"step": 475 |
|
}, |
|
{ |
|
"epoch": 1.4020000000000001, |
|
"grad_norm": 2.2619755268096924, |
|
"learning_rate": 8.88888888888889e-08, |
|
"loss": 0.0444, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 1.4020000000000001, |
|
"eval_loss": 0.37679705023765564, |
|
"eval_runtime": 157.3635, |
|
"eval_samples_per_second": 2.021, |
|
"eval_steps_per_second": 0.064, |
|
"eval_wer": 26.64601084430674, |
|
"step": 500 |
|
} |
|
], |
|
"logging_steps": 25, |
|
"max_steps": 500, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 9223372036854775807, |
|
"save_steps": 100, |
|
"stateful_callbacks": { |
|
"TrainerControl": { |
|
"args": { |
|
"should_epoch_stop": false, |
|
"should_evaluate": false, |
|
"should_log": false, |
|
"should_save": true, |
|
"should_training_stop": true |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 9.22347797962752e+18, |
|
"train_batch_size": 64, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|