|
{ |
|
"best_metric": 102.28571428571429, |
|
"best_model_checkpoint": "./medium_TH/checkpoint-1000", |
|
"epoch": 19.23076923076923, |
|
"global_step": 1000, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.48, |
|
"learning_rate": 4.4e-07, |
|
"loss": 0.782, |
|
"step": 25 |
|
}, |
|
{ |
|
"epoch": 0.96, |
|
"learning_rate": 9.200000000000001e-07, |
|
"loss": 0.6976, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 1.44, |
|
"learning_rate": 1.42e-06, |
|
"loss": 0.6697, |
|
"step": 75 |
|
}, |
|
{ |
|
"epoch": 1.92, |
|
"learning_rate": 1.9200000000000003e-06, |
|
"loss": 0.5716, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 2.4, |
|
"learning_rate": 2.42e-06, |
|
"loss": 0.569, |
|
"step": 125 |
|
}, |
|
{ |
|
"epoch": 2.88, |
|
"learning_rate": 2.92e-06, |
|
"loss": 0.4445, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 3.37, |
|
"learning_rate": 3.4200000000000007e-06, |
|
"loss": 0.3945, |
|
"step": 175 |
|
}, |
|
{ |
|
"epoch": 3.85, |
|
"learning_rate": 3.920000000000001e-06, |
|
"loss": 0.3925, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 4.33, |
|
"learning_rate": 4.42e-06, |
|
"loss": 0.3319, |
|
"step": 225 |
|
}, |
|
{ |
|
"epoch": 4.81, |
|
"learning_rate": 4.92e-06, |
|
"loss": 0.289, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 5.29, |
|
"learning_rate": 5.420000000000001e-06, |
|
"loss": 0.2305, |
|
"step": 275 |
|
}, |
|
{ |
|
"epoch": 5.77, |
|
"learning_rate": 5.92e-06, |
|
"loss": 0.2545, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 6.25, |
|
"learning_rate": 6.42e-06, |
|
"loss": 0.2123, |
|
"step": 325 |
|
}, |
|
{ |
|
"epoch": 6.73, |
|
"learning_rate": 6.92e-06, |
|
"loss": 0.1256, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 7.21, |
|
"learning_rate": 7.420000000000001e-06, |
|
"loss": 0.1196, |
|
"step": 375 |
|
}, |
|
{ |
|
"epoch": 7.69, |
|
"learning_rate": 7.92e-06, |
|
"loss": 0.1217, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 8.17, |
|
"learning_rate": 8.42e-06, |
|
"loss": 0.1214, |
|
"step": 425 |
|
}, |
|
{ |
|
"epoch": 8.65, |
|
"learning_rate": 8.920000000000001e-06, |
|
"loss": 0.1197, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 9.13, |
|
"learning_rate": 9.42e-06, |
|
"loss": 0.1044, |
|
"step": 475 |
|
}, |
|
{ |
|
"epoch": 9.62, |
|
"learning_rate": 9.920000000000002e-06, |
|
"loss": 0.1093, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 10.1, |
|
"learning_rate": 9.940000000000001e-06, |
|
"loss": 0.124, |
|
"step": 525 |
|
}, |
|
{ |
|
"epoch": 10.58, |
|
"learning_rate": 9.86857142857143e-06, |
|
"loss": 0.0515, |
|
"step": 550 |
|
}, |
|
{ |
|
"epoch": 11.06, |
|
"learning_rate": 9.797142857142858e-06, |
|
"loss": 0.1163, |
|
"step": 575 |
|
}, |
|
{ |
|
"epoch": 11.54, |
|
"learning_rate": 9.725714285714287e-06, |
|
"loss": 0.0971, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 12.02, |
|
"learning_rate": 9.654285714285716e-06, |
|
"loss": 0.0775, |
|
"step": 625 |
|
}, |
|
{ |
|
"epoch": 12.5, |
|
"learning_rate": 9.582857142857143e-06, |
|
"loss": 0.0775, |
|
"step": 650 |
|
}, |
|
{ |
|
"epoch": 12.98, |
|
"learning_rate": 9.511428571428572e-06, |
|
"loss": 0.0842, |
|
"step": 675 |
|
}, |
|
{ |
|
"epoch": 13.46, |
|
"learning_rate": 9.440000000000001e-06, |
|
"loss": 0.0781, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 13.94, |
|
"learning_rate": 9.368571428571428e-06, |
|
"loss": 0.0481, |
|
"step": 725 |
|
}, |
|
{ |
|
"epoch": 14.42, |
|
"learning_rate": 9.297142857142857e-06, |
|
"loss": 0.0649, |
|
"step": 750 |
|
}, |
|
{ |
|
"epoch": 14.9, |
|
"learning_rate": 9.225714285714286e-06, |
|
"loss": 0.0269, |
|
"step": 775 |
|
}, |
|
{ |
|
"epoch": 15.38, |
|
"learning_rate": 9.154285714285715e-06, |
|
"loss": 0.0721, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 15.87, |
|
"learning_rate": 9.082857142857143e-06, |
|
"loss": 0.0459, |
|
"step": 825 |
|
}, |
|
{ |
|
"epoch": 16.35, |
|
"learning_rate": 9.011428571428572e-06, |
|
"loss": 0.0361, |
|
"step": 850 |
|
}, |
|
{ |
|
"epoch": 16.83, |
|
"learning_rate": 8.94e-06, |
|
"loss": 0.0598, |
|
"step": 875 |
|
}, |
|
{ |
|
"epoch": 17.31, |
|
"learning_rate": 8.86857142857143e-06, |
|
"loss": 0.0479, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 17.79, |
|
"learning_rate": 8.797142857142857e-06, |
|
"loss": 0.046, |
|
"step": 925 |
|
}, |
|
{ |
|
"epoch": 18.27, |
|
"learning_rate": 8.725714285714286e-06, |
|
"loss": 0.0197, |
|
"step": 950 |
|
}, |
|
{ |
|
"epoch": 18.75, |
|
"learning_rate": 8.654285714285715e-06, |
|
"loss": 0.0377, |
|
"step": 975 |
|
}, |
|
{ |
|
"epoch": 19.23, |
|
"learning_rate": 8.582857142857144e-06, |
|
"loss": 0.052, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 19.23, |
|
"eval_loss": 0.38925179839134216, |
|
"eval_runtime": 14.2835, |
|
"eval_samples_per_second": 1.61, |
|
"eval_steps_per_second": 0.42, |
|
"eval_wer": 102.28571428571429, |
|
"step": 1000 |
|
} |
|
], |
|
"max_steps": 4000, |
|
"num_train_epochs": 77, |
|
"total_flos": 2.02181832179712e+18, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|