|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 4.485623003194888, |
|
"global_step": 350, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.26, |
|
"learning_rate": 0.0007307692307692308, |
|
"loss": 4.2559, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.51, |
|
"learning_rate": 0.0007115384615384615, |
|
"loss": 3.068, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.64, |
|
"eval_loss": 3.058549165725708, |
|
"eval_runtime": 3.3353, |
|
"eval_samples_per_second": 29.982, |
|
"eval_steps_per_second": 1.199, |
|
"eval_wer": 1.0, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.77, |
|
"learning_rate": 0.0006923076923076924, |
|
"loss": 3.0637, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 1.03, |
|
"learning_rate": 0.0006730769230769232, |
|
"loss": 3.0752, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 1.28, |
|
"learning_rate": 0.0006538461538461538, |
|
"loss": 3.0083, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 1.28, |
|
"eval_loss": 3.1246652603149414, |
|
"eval_runtime": 3.2285, |
|
"eval_samples_per_second": 30.974, |
|
"eval_steps_per_second": 1.239, |
|
"eval_wer": 1.0, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 1.54, |
|
"learning_rate": 0.0006346153846153846, |
|
"loss": 3.0159, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 1.79, |
|
"learning_rate": 0.0006153846153846154, |
|
"loss": 2.977, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 1.92, |
|
"eval_loss": 2.973355770111084, |
|
"eval_runtime": 3.1716, |
|
"eval_samples_per_second": 31.53, |
|
"eval_steps_per_second": 1.261, |
|
"eval_wer": 1.0, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 2.05, |
|
"learning_rate": 0.0005961538461538461, |
|
"loss": 3.0056, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 2.31, |
|
"learning_rate": 0.000576923076923077, |
|
"loss": 2.9591, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 2.56, |
|
"learning_rate": 0.0005576923076923078, |
|
"loss": 2.9398, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 2.56, |
|
"eval_loss": 2.940675735473633, |
|
"eval_runtime": 3.1691, |
|
"eval_samples_per_second": 31.554, |
|
"eval_steps_per_second": 1.262, |
|
"eval_wer": 1.0, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 2.82, |
|
"learning_rate": 0.0005384615384615384, |
|
"loss": 2.9295, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 3.08, |
|
"learning_rate": 0.0005192307692307692, |
|
"loss": 2.9465, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 3.2, |
|
"eval_loss": 2.835913896560669, |
|
"eval_runtime": 3.1824, |
|
"eval_samples_per_second": 31.423, |
|
"eval_steps_per_second": 1.257, |
|
"eval_wer": 1.0, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 3.33, |
|
"learning_rate": 0.0005, |
|
"loss": 2.9053, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 3.59, |
|
"learning_rate": 0.0004807692307692308, |
|
"loss": 2.899, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 3.84, |
|
"learning_rate": 0.0004615384615384616, |
|
"loss": 2.8861, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 3.84, |
|
"eval_loss": 2.8373122215270996, |
|
"eval_runtime": 3.1455, |
|
"eval_samples_per_second": 31.791, |
|
"eval_steps_per_second": 1.272, |
|
"eval_wer": 1.0, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 4.1, |
|
"learning_rate": 0.0004423076923076923, |
|
"loss": 2.9255, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 4.36, |
|
"learning_rate": 0.0004230769230769231, |
|
"loss": 2.884, |
|
"step": 340 |
|
}, |
|
{ |
|
"epoch": 4.49, |
|
"eval_loss": 2.815220832824707, |
|
"eval_runtime": 3.2038, |
|
"eval_samples_per_second": 31.213, |
|
"eval_steps_per_second": 1.249, |
|
"eval_wer": 1.0476190476190477, |
|
"step": 350 |
|
} |
|
], |
|
"max_steps": 780, |
|
"num_train_epochs": 10, |
|
"total_flos": 5.238613498111745e+18, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|