|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 0.32217236221378437, |
|
"global_step": 700, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.01, |
|
"learning_rate": 1.0305343511450381e-05, |
|
"loss": 13.0539, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"learning_rate": 2.118320610687023e-05, |
|
"loss": 20.9649, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"learning_rate": 3.206106870229007e-05, |
|
"loss": 21.1001, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 0.04, |
|
"learning_rate": 4.3511450381679383e-05, |
|
"loss": 5.5852, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 0.05, |
|
"learning_rate": 5.438931297709923e-05, |
|
"loss": 5.2002, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.05, |
|
"eval_loss": 3.5950469970703125, |
|
"eval_runtime": 194.8685, |
|
"eval_samples_per_second": 24.853, |
|
"eval_steps_per_second": 0.78, |
|
"eval_wer": 1.0, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.06, |
|
"learning_rate": 6.583969465648854e-05, |
|
"loss": 3.281, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 0.06, |
|
"learning_rate": 7.495301487862176e-05, |
|
"loss": 3.3442, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 0.07, |
|
"learning_rate": 7.472983555207517e-05, |
|
"loss": 3.5229, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 0.08, |
|
"learning_rate": 7.449490994518402e-05, |
|
"loss": 3.089, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 0.09, |
|
"learning_rate": 7.427173061863743e-05, |
|
"loss": 3.482, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.09, |
|
"eval_loss": 3.274627923965454, |
|
"eval_runtime": 194.5649, |
|
"eval_samples_per_second": 24.891, |
|
"eval_steps_per_second": 0.781, |
|
"eval_wer": 1.0, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"learning_rate": 7.403680501174628e-05, |
|
"loss": 3.1517, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 0.11, |
|
"learning_rate": 7.380187940485512e-05, |
|
"loss": 3.1617, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 0.12, |
|
"learning_rate": 7.357870007830852e-05, |
|
"loss": 3.2971, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 0.13, |
|
"learning_rate": 7.334377447141738e-05, |
|
"loss": 3.022, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 0.14, |
|
"learning_rate": 7.312059514487078e-05, |
|
"loss": 3.3322, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.14, |
|
"eval_loss": 3.0716073513031006, |
|
"eval_runtime": 196.1674, |
|
"eval_samples_per_second": 24.688, |
|
"eval_steps_per_second": 0.775, |
|
"eval_wer": 1.0, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.15, |
|
"learning_rate": 7.288566953797964e-05, |
|
"loss": 2.9964, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 0.16, |
|
"learning_rate": 7.265074393108849e-05, |
|
"loss": 3.0777, |
|
"step": 340 |
|
}, |
|
{ |
|
"epoch": 0.17, |
|
"learning_rate": 7.242756460454189e-05, |
|
"loss": 3.2703, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 0.17, |
|
"learning_rate": 7.219263899765073e-05, |
|
"loss": 3.1161, |
|
"step": 380 |
|
}, |
|
{ |
|
"epoch": 0.18, |
|
"learning_rate": 7.195771339075958e-05, |
|
"loss": 3.2435, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.18, |
|
"eval_loss": 3.063333034515381, |
|
"eval_runtime": 190.5938, |
|
"eval_samples_per_second": 25.41, |
|
"eval_steps_per_second": 0.798, |
|
"eval_wer": 1.0, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.19, |
|
"learning_rate": 7.172278778386843e-05, |
|
"loss": 2.9993, |
|
"step": 420 |
|
}, |
|
{ |
|
"epoch": 0.2, |
|
"learning_rate": 7.148786217697727e-05, |
|
"loss": 3.0555, |
|
"step": 440 |
|
}, |
|
{ |
|
"epoch": 0.21, |
|
"learning_rate": 7.126468285043069e-05, |
|
"loss": 3.1879, |
|
"step": 460 |
|
}, |
|
{ |
|
"epoch": 0.22, |
|
"learning_rate": 7.102975724353953e-05, |
|
"loss": 2.9861, |
|
"step": 480 |
|
}, |
|
{ |
|
"epoch": 0.23, |
|
"learning_rate": 7.080657791699295e-05, |
|
"loss": 3.0833, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.23, |
|
"eval_loss": 2.9769718647003174, |
|
"eval_runtime": 195.2639, |
|
"eval_samples_per_second": 24.802, |
|
"eval_steps_per_second": 0.778, |
|
"eval_wer": 1.0, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.24, |
|
"learning_rate": 7.05716523101018e-05, |
|
"loss": 3.1608, |
|
"step": 520 |
|
}, |
|
{ |
|
"epoch": 0.25, |
|
"learning_rate": 7.033672670321064e-05, |
|
"loss": 3.0206, |
|
"step": 540 |
|
}, |
|
{ |
|
"epoch": 0.26, |
|
"learning_rate": 7.01018010963195e-05, |
|
"loss": 3.0523, |
|
"step": 560 |
|
}, |
|
{ |
|
"epoch": 0.27, |
|
"learning_rate": 6.986687548942835e-05, |
|
"loss": 3.0222, |
|
"step": 580 |
|
}, |
|
{ |
|
"epoch": 0.28, |
|
"learning_rate": 6.964369616288175e-05, |
|
"loss": 3.0913, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 0.28, |
|
"eval_loss": 3.052091598510742, |
|
"eval_runtime": 189.8149, |
|
"eval_samples_per_second": 25.514, |
|
"eval_steps_per_second": 0.801, |
|
"eval_wer": 1.0, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 0.29, |
|
"learning_rate": 6.94087705559906e-05, |
|
"loss": 3.1987, |
|
"step": 620 |
|
}, |
|
{ |
|
"epoch": 0.29, |
|
"learning_rate": 6.917384494909944e-05, |
|
"loss": 3.1504, |
|
"step": 640 |
|
}, |
|
{ |
|
"epoch": 0.3, |
|
"learning_rate": 6.893891934220829e-05, |
|
"loss": 3.2406, |
|
"step": 660 |
|
}, |
|
{ |
|
"epoch": 0.31, |
|
"learning_rate": 6.870399373531715e-05, |
|
"loss": 3.0632, |
|
"step": 680 |
|
}, |
|
{ |
|
"epoch": 0.32, |
|
"learning_rate": 6.848081440877055e-05, |
|
"loss": 3.1487, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 0.32, |
|
"eval_loss": 3.1951305866241455, |
|
"eval_runtime": 196.8636, |
|
"eval_samples_per_second": 24.601, |
|
"eval_steps_per_second": 0.772, |
|
"eval_wer": 1.0, |
|
"step": 700 |
|
} |
|
], |
|
"max_steps": 6516, |
|
"num_train_epochs": 3, |
|
"total_flos": 1.3316117705308692e+19, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|