|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 0.3229527104959631, |
|
"global_step": 700, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.0, |
|
"learning_rate": 5.725190839694656e-06, |
|
"loss": 12.7187, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"learning_rate": 1.1450381679389312e-05, |
|
"loss": 13.2124, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"learning_rate": 1.6603053435114505e-05, |
|
"loss": 14.1489, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"learning_rate": 2.1755725190839692e-05, |
|
"loss": 25.9008, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"learning_rate": 2.6908396946564882e-05, |
|
"loss": 31.3861, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"learning_rate": 3.263358778625954e-05, |
|
"loss": 8.9557, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"learning_rate": 3.835877862595419e-05, |
|
"loss": 5.8443, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 0.04, |
|
"learning_rate": 4.408396946564885e-05, |
|
"loss": 4.3173, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 0.04, |
|
"learning_rate": 4.980916030534351e-05, |
|
"loss": 4.7444, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 0.05, |
|
"learning_rate": 5.49618320610687e-05, |
|
"loss": 4.542, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.05, |
|
"eval_loss": 3.579599618911743, |
|
"eval_runtime": 132.6751, |
|
"eval_samples_per_second": 34.822, |
|
"eval_steps_per_second": 1.093, |
|
"eval_wer": 1.0, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.05, |
|
"learning_rate": 6.0687022900763354e-05, |
|
"loss": 3.3873, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 0.06, |
|
"learning_rate": 6.641221374045802e-05, |
|
"loss": 3.1783, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 0.06, |
|
"learning_rate": 7.213740458015267e-05, |
|
"loss": 3.1672, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 0.06, |
|
"learning_rate": 7.494113029827315e-05, |
|
"loss": 3.4915, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 0.07, |
|
"learning_rate": 7.483516483516483e-05, |
|
"loss": 3.7276, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 0.07, |
|
"learning_rate": 7.471742543171114e-05, |
|
"loss": 3.2435, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 0.08, |
|
"learning_rate": 7.459968602825744e-05, |
|
"loss": 3.0634, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 0.08, |
|
"learning_rate": 7.448194662480377e-05, |
|
"loss": 3.0859, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 0.09, |
|
"learning_rate": 7.436420722135007e-05, |
|
"loss": 3.3592, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 0.09, |
|
"learning_rate": 7.425824175824176e-05, |
|
"loss": 3.6018, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.09, |
|
"eval_loss": 3.3504831790924072, |
|
"eval_runtime": 130.1255, |
|
"eval_samples_per_second": 35.504, |
|
"eval_steps_per_second": 1.114, |
|
"eval_wer": 1.0, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"learning_rate": 7.414050235478806e-05, |
|
"loss": 3.1612, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"learning_rate": 7.402276295133437e-05, |
|
"loss": 3.0375, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 0.11, |
|
"learning_rate": 7.390502354788069e-05, |
|
"loss": 3.08, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 0.11, |
|
"learning_rate": 7.3787284144427e-05, |
|
"loss": 3.2809, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 0.12, |
|
"learning_rate": 7.368131868131867e-05, |
|
"loss": 3.5351, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 0.12, |
|
"learning_rate": 7.356357927786499e-05, |
|
"loss": 3.01, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 0.12, |
|
"learning_rate": 7.34458398744113e-05, |
|
"loss": 3.0006, |
|
"step": 270 |
|
}, |
|
{ |
|
"epoch": 0.13, |
|
"learning_rate": 7.33281004709576e-05, |
|
"loss": 3.0397, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 0.13, |
|
"learning_rate": 7.321036106750392e-05, |
|
"loss": 3.1733, |
|
"step": 290 |
|
}, |
|
{ |
|
"epoch": 0.14, |
|
"learning_rate": 7.310439560439559e-05, |
|
"loss": 3.5007, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.14, |
|
"eval_loss": 3.068483591079712, |
|
"eval_runtime": 129.9806, |
|
"eval_samples_per_second": 35.544, |
|
"eval_steps_per_second": 1.116, |
|
"eval_wer": 1.0, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.14, |
|
"learning_rate": 7.29866562009419e-05, |
|
"loss": 2.9984, |
|
"step": 310 |
|
}, |
|
{ |
|
"epoch": 0.15, |
|
"learning_rate": 7.286891679748822e-05, |
|
"loss": 3.0006, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 0.15, |
|
"learning_rate": 7.275117739403453e-05, |
|
"loss": 3.029, |
|
"step": 330 |
|
}, |
|
{ |
|
"epoch": 0.16, |
|
"learning_rate": 7.263343799058085e-05, |
|
"loss": 3.158, |
|
"step": 340 |
|
}, |
|
{ |
|
"epoch": 0.16, |
|
"learning_rate": 7.252747252747252e-05, |
|
"loss": 3.4909, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 0.17, |
|
"learning_rate": 7.240973312401883e-05, |
|
"loss": 3.001, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 0.17, |
|
"learning_rate": 7.229199372056515e-05, |
|
"loss": 2.9925, |
|
"step": 370 |
|
}, |
|
{ |
|
"epoch": 0.18, |
|
"learning_rate": 7.217425431711145e-05, |
|
"loss": 3.0009, |
|
"step": 380 |
|
}, |
|
{ |
|
"epoch": 0.18, |
|
"learning_rate": 7.205651491365776e-05, |
|
"loss": 3.1263, |
|
"step": 390 |
|
}, |
|
{ |
|
"epoch": 0.18, |
|
"learning_rate": 7.195054945054944e-05, |
|
"loss": 3.421, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.18, |
|
"eval_loss": 3.0357792377471924, |
|
"eval_runtime": 130.0328, |
|
"eval_samples_per_second": 35.529, |
|
"eval_steps_per_second": 1.115, |
|
"eval_wer": 1.0, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.19, |
|
"learning_rate": 7.183281004709575e-05, |
|
"loss": 3.0018, |
|
"step": 410 |
|
}, |
|
{ |
|
"epoch": 0.19, |
|
"learning_rate": 7.171507064364207e-05, |
|
"loss": 2.9947, |
|
"step": 420 |
|
}, |
|
{ |
|
"epoch": 0.2, |
|
"learning_rate": 7.159733124018838e-05, |
|
"loss": 2.9935, |
|
"step": 430 |
|
}, |
|
{ |
|
"epoch": 0.2, |
|
"learning_rate": 7.147959183673469e-05, |
|
"loss": 3.0818, |
|
"step": 440 |
|
}, |
|
{ |
|
"epoch": 0.21, |
|
"learning_rate": 7.137362637362637e-05, |
|
"loss": 3.3386, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 0.21, |
|
"learning_rate": 7.125588697017268e-05, |
|
"loss": 3.0803, |
|
"step": 460 |
|
}, |
|
{ |
|
"epoch": 0.22, |
|
"learning_rate": 7.1138147566719e-05, |
|
"loss": 3.1655, |
|
"step": 470 |
|
}, |
|
{ |
|
"epoch": 0.22, |
|
"learning_rate": 7.10204081632653e-05, |
|
"loss": 3.0899, |
|
"step": 480 |
|
}, |
|
{ |
|
"epoch": 0.23, |
|
"learning_rate": 7.090266875981161e-05, |
|
"loss": 3.1638, |
|
"step": 490 |
|
}, |
|
{ |
|
"epoch": 0.23, |
|
"learning_rate": 7.079670329670328e-05, |
|
"loss": 3.3518, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.23, |
|
"eval_loss": 2.9412434101104736, |
|
"eval_runtime": 130.1575, |
|
"eval_samples_per_second": 35.495, |
|
"eval_steps_per_second": 1.114, |
|
"eval_wer": 1.0, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.24, |
|
"learning_rate": 7.06789638932496e-05, |
|
"loss": 2.9352, |
|
"step": 510 |
|
}, |
|
{ |
|
"epoch": 0.24, |
|
"learning_rate": 7.056122448979591e-05, |
|
"loss": 2.9405, |
|
"step": 520 |
|
}, |
|
{ |
|
"epoch": 0.24, |
|
"learning_rate": 7.044348508634222e-05, |
|
"loss": 2.9125, |
|
"step": 530 |
|
}, |
|
{ |
|
"epoch": 0.25, |
|
"learning_rate": 7.032574568288854e-05, |
|
"loss": 2.9401, |
|
"step": 540 |
|
}, |
|
{ |
|
"epoch": 0.25, |
|
"learning_rate": 7.021978021978021e-05, |
|
"loss": 3.1443, |
|
"step": 550 |
|
}, |
|
{ |
|
"epoch": 0.26, |
|
"learning_rate": 7.010204081632653e-05, |
|
"loss": 2.7692, |
|
"step": 560 |
|
}, |
|
{ |
|
"epoch": 0.26, |
|
"learning_rate": 6.998430141287284e-05, |
|
"loss": 2.7574, |
|
"step": 570 |
|
}, |
|
{ |
|
"epoch": 0.27, |
|
"learning_rate": 6.986656200941914e-05, |
|
"loss": 2.706, |
|
"step": 580 |
|
}, |
|
{ |
|
"epoch": 0.27, |
|
"learning_rate": 6.974882260596545e-05, |
|
"loss": 2.7155, |
|
"step": 590 |
|
}, |
|
{ |
|
"epoch": 0.28, |
|
"learning_rate": 6.964285714285713e-05, |
|
"loss": 2.8747, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 0.28, |
|
"eval_loss": 2.5006003379821777, |
|
"eval_runtime": 129.3111, |
|
"eval_samples_per_second": 35.728, |
|
"eval_steps_per_second": 1.121, |
|
"eval_wer": 1.0, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 0.28, |
|
"learning_rate": 6.952511773940345e-05, |
|
"loss": 2.5368, |
|
"step": 610 |
|
}, |
|
{ |
|
"epoch": 0.29, |
|
"learning_rate": 6.940737833594976e-05, |
|
"loss": 2.5942, |
|
"step": 620 |
|
}, |
|
{ |
|
"epoch": 0.29, |
|
"learning_rate": 6.928963893249607e-05, |
|
"loss": 2.4907, |
|
"step": 630 |
|
}, |
|
{ |
|
"epoch": 0.3, |
|
"learning_rate": 6.917189952904238e-05, |
|
"loss": 2.472, |
|
"step": 640 |
|
}, |
|
{ |
|
"epoch": 0.3, |
|
"learning_rate": 6.907770800627943e-05, |
|
"loss": 2.6826, |
|
"step": 650 |
|
}, |
|
{ |
|
"epoch": 0.3, |
|
"learning_rate": 6.895996860282574e-05, |
|
"loss": 2.3167, |
|
"step": 660 |
|
}, |
|
{ |
|
"epoch": 0.31, |
|
"learning_rate": 6.884222919937205e-05, |
|
"loss": 2.3457, |
|
"step": 670 |
|
}, |
|
{ |
|
"epoch": 0.31, |
|
"learning_rate": 6.872448979591836e-05, |
|
"loss": 2.3532, |
|
"step": 680 |
|
}, |
|
{ |
|
"epoch": 0.32, |
|
"learning_rate": 6.860675039246468e-05, |
|
"loss": 2.4822, |
|
"step": 690 |
|
}, |
|
{ |
|
"epoch": 0.32, |
|
"learning_rate": 6.850078492935636e-05, |
|
"loss": 3.1002, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 0.32, |
|
"eval_loss": 2.819200277328491, |
|
"eval_runtime": 129.5044, |
|
"eval_samples_per_second": 35.674, |
|
"eval_steps_per_second": 1.12, |
|
"eval_wer": 0.9999694058618369, |
|
"step": 700 |
|
} |
|
], |
|
"max_steps": 6501, |
|
"num_train_epochs": 3, |
|
"total_flos": 1.3228060515474647e+19, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|