|
{ |
|
"best_metric": 45.769943585670354, |
|
"best_model_checkpoint": "./whisper_finetuned/checkpoint-500", |
|
"epoch": 0.16, |
|
"eval_steps": 500, |
|
"global_step": 500, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.01, |
|
"learning_rate": 4.2000000000000006e-07, |
|
"loss": 3.6327, |
|
"step": 25 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"learning_rate": 9.200000000000001e-07, |
|
"loss": 2.9524, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"learning_rate": 1.42e-06, |
|
"loss": 2.1796, |
|
"step": 75 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"learning_rate": 1.9200000000000003e-06, |
|
"loss": 1.4059, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.04, |
|
"learning_rate": 2.42e-06, |
|
"loss": 1.0336, |
|
"step": 125 |
|
}, |
|
{ |
|
"epoch": 0.05, |
|
"learning_rate": 2.92e-06, |
|
"loss": 0.9054, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 0.06, |
|
"learning_rate": 3.4200000000000007e-06, |
|
"loss": 0.8159, |
|
"step": 175 |
|
}, |
|
{ |
|
"epoch": 0.06, |
|
"learning_rate": 3.920000000000001e-06, |
|
"loss": 0.7266, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.07, |
|
"learning_rate": 4.42e-06, |
|
"loss": 0.6528, |
|
"step": 225 |
|
}, |
|
{ |
|
"epoch": 0.08, |
|
"learning_rate": 4.92e-06, |
|
"loss": 0.6193, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 0.09, |
|
"learning_rate": 5.420000000000001e-06, |
|
"loss": 0.4741, |
|
"step": 275 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"learning_rate": 5.92e-06, |
|
"loss": 0.3247, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"learning_rate": 6.42e-06, |
|
"loss": 0.2819, |
|
"step": 325 |
|
}, |
|
{ |
|
"epoch": 0.11, |
|
"learning_rate": 6.92e-06, |
|
"loss": 0.2899, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 0.12, |
|
"learning_rate": 7.420000000000001e-06, |
|
"loss": 0.286, |
|
"step": 375 |
|
}, |
|
{ |
|
"epoch": 0.13, |
|
"learning_rate": 7.92e-06, |
|
"loss": 0.278, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.14, |
|
"learning_rate": 8.42e-06, |
|
"loss": 0.2817, |
|
"step": 425 |
|
}, |
|
{ |
|
"epoch": 0.14, |
|
"learning_rate": 8.920000000000001e-06, |
|
"loss": 0.2855, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 0.15, |
|
"learning_rate": 9.42e-06, |
|
"loss": 0.2825, |
|
"step": 475 |
|
}, |
|
{ |
|
"epoch": 0.16, |
|
"learning_rate": 9.920000000000002e-06, |
|
"loss": 0.2856, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.16, |
|
"eval_loss": 0.3059645891189575, |
|
"eval_runtime": 10209.9182, |
|
"eval_samples_per_second": 0.979, |
|
"eval_steps_per_second": 0.122, |
|
"eval_wer": 45.769943585670354, |
|
"step": 500 |
|
} |
|
], |
|
"logging_steps": 25, |
|
"max_steps": 15000, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 5, |
|
"save_steps": 500, |
|
"total_flos": 2.30868320256e+18, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|