|
{ |
|
"best_metric": 15.718055032039203, |
|
"best_model_checkpoint": "./checkpoint-1000", |
|
"epoch": 19.012, |
|
"global_step": 1000, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.03, |
|
"learning_rate": 4.6000000000000004e-07, |
|
"loss": 1.2023, |
|
"step": 25 |
|
}, |
|
{ |
|
"epoch": 0.05, |
|
"learning_rate": 9.600000000000001e-07, |
|
"loss": 0.8225, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 1.02, |
|
"learning_rate": 1.46e-06, |
|
"loss": 0.4219, |
|
"step": 75 |
|
}, |
|
{ |
|
"epoch": 1.05, |
|
"learning_rate": 1.9600000000000003e-06, |
|
"loss": 0.2197, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 2.02, |
|
"learning_rate": 2.46e-06, |
|
"loss": 0.2024, |
|
"step": 125 |
|
}, |
|
{ |
|
"epoch": 2.05, |
|
"learning_rate": 2.96e-06, |
|
"loss": 0.1199, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 3.02, |
|
"learning_rate": 3.46e-06, |
|
"loss": 0.1207, |
|
"step": 175 |
|
}, |
|
{ |
|
"epoch": 3.04, |
|
"learning_rate": 3.96e-06, |
|
"loss": 0.0724, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 4.02, |
|
"learning_rate": 4.4600000000000005e-06, |
|
"loss": 0.062, |
|
"step": 225 |
|
}, |
|
{ |
|
"epoch": 4.04, |
|
"learning_rate": 4.960000000000001e-06, |
|
"loss": 0.0386, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 5.01, |
|
"learning_rate": 5.460000000000001e-06, |
|
"loss": 0.0326, |
|
"step": 275 |
|
}, |
|
{ |
|
"epoch": 5.04, |
|
"learning_rate": 5.9600000000000005e-06, |
|
"loss": 0.0206, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 6.01, |
|
"learning_rate": 6.460000000000001e-06, |
|
"loss": 0.0183, |
|
"step": 325 |
|
}, |
|
{ |
|
"epoch": 6.04, |
|
"learning_rate": 6.96e-06, |
|
"loss": 0.0136, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 7.01, |
|
"learning_rate": 7.4600000000000006e-06, |
|
"loss": 0.0121, |
|
"step": 375 |
|
}, |
|
{ |
|
"epoch": 7.04, |
|
"learning_rate": 7.960000000000002e-06, |
|
"loss": 0.0107, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 8.01, |
|
"learning_rate": 8.46e-06, |
|
"loss": 0.0105, |
|
"step": 425 |
|
}, |
|
{ |
|
"epoch": 8.03, |
|
"learning_rate": 8.96e-06, |
|
"loss": 0.0101, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 9.01, |
|
"learning_rate": 9.460000000000001e-06, |
|
"loss": 0.0116, |
|
"step": 475 |
|
}, |
|
{ |
|
"epoch": 9.03, |
|
"learning_rate": 9.960000000000001e-06, |
|
"loss": 0.011, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 10.01, |
|
"learning_rate": 9.54e-06, |
|
"loss": 0.0128, |
|
"step": 525 |
|
}, |
|
{ |
|
"epoch": 10.03, |
|
"learning_rate": 9.040000000000002e-06, |
|
"loss": 0.0104, |
|
"step": 550 |
|
}, |
|
{ |
|
"epoch": 11.0, |
|
"learning_rate": 8.540000000000001e-06, |
|
"loss": 0.0083, |
|
"step": 575 |
|
}, |
|
{ |
|
"epoch": 11.03, |
|
"learning_rate": 8.040000000000001e-06, |
|
"loss": 0.0055, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 12.0, |
|
"learning_rate": 7.540000000000001e-06, |
|
"loss": 0.0037, |
|
"step": 625 |
|
}, |
|
{ |
|
"epoch": 12.03, |
|
"learning_rate": 7.04e-06, |
|
"loss": 0.0041, |
|
"step": 650 |
|
}, |
|
{ |
|
"epoch": 12.05, |
|
"learning_rate": 6.540000000000001e-06, |
|
"loss": 0.0032, |
|
"step": 675 |
|
}, |
|
{ |
|
"epoch": 13.02, |
|
"learning_rate": 6.040000000000001e-06, |
|
"loss": 0.0025, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 13.05, |
|
"learning_rate": 5.540000000000001e-06, |
|
"loss": 0.0021, |
|
"step": 725 |
|
}, |
|
{ |
|
"epoch": 14.02, |
|
"learning_rate": 5.04e-06, |
|
"loss": 0.0015, |
|
"step": 750 |
|
}, |
|
{ |
|
"epoch": 14.05, |
|
"learning_rate": 4.540000000000001e-06, |
|
"loss": 0.0011, |
|
"step": 775 |
|
}, |
|
{ |
|
"epoch": 15.02, |
|
"learning_rate": 4.04e-06, |
|
"loss": 0.0011, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 15.04, |
|
"learning_rate": 3.54e-06, |
|
"loss": 0.0009, |
|
"step": 825 |
|
}, |
|
{ |
|
"epoch": 16.02, |
|
"learning_rate": 3.04e-06, |
|
"loss": 0.001, |
|
"step": 850 |
|
}, |
|
{ |
|
"epoch": 16.04, |
|
"learning_rate": 2.5400000000000002e-06, |
|
"loss": 0.0009, |
|
"step": 875 |
|
}, |
|
{ |
|
"epoch": 17.02, |
|
"learning_rate": 2.04e-06, |
|
"loss": 0.0009, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 17.04, |
|
"learning_rate": 1.54e-06, |
|
"loss": 0.0008, |
|
"step": 925 |
|
}, |
|
{ |
|
"epoch": 18.01, |
|
"learning_rate": 1.04e-06, |
|
"loss": 0.0009, |
|
"step": 950 |
|
}, |
|
{ |
|
"epoch": 18.04, |
|
"learning_rate": 5.4e-07, |
|
"loss": 0.0008, |
|
"step": 975 |
|
}, |
|
{ |
|
"epoch": 19.01, |
|
"learning_rate": 4e-08, |
|
"loss": 0.0009, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 19.01, |
|
"eval_loss": 0.30098843574523926, |
|
"eval_runtime": 1385.7195, |
|
"eval_samples_per_second": 1.23, |
|
"eval_steps_per_second": 0.039, |
|
"eval_wer": 15.718055032039203, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 19.01, |
|
"step": 1000, |
|
"total_flos": 6.477575227047936e+19, |
|
"train_loss": 0.08742457270808518, |
|
"train_runtime": 16211.9832, |
|
"train_samples_per_second": 3.948, |
|
"train_steps_per_second": 0.062 |
|
} |
|
], |
|
"max_steps": 1000, |
|
"num_train_epochs": 9223372036854775807, |
|
"total_flos": 6.477575227047936e+19, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|