| { | |
| "best_metric": 0.13642320036888123, | |
| "best_model_checkpoint": "wav2vec2-large-xlsr-mecita-coraa-portuguese-random-all-03/checkpoint-3182", | |
| "epoch": 57.0, | |
| "global_step": 4902, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 3e-05, | |
| "loss": 29.1535, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_cer": 1.0, | |
| "eval_loss": 3.1676151752471924, | |
| "eval_runtime": 5.9631, | |
| "eval_samples_per_second": 57.185, | |
| "eval_steps_per_second": 7.211, | |
| "eval_wer": 1.0, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 2.9661627906976745e-05, | |
| "loss": 7.924, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_cer": 1.0, | |
| "eval_loss": 2.9288899898529053, | |
| "eval_runtime": 6.0836, | |
| "eval_samples_per_second": 56.053, | |
| "eval_steps_per_second": 7.068, | |
| "eval_wer": 1.0, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 2.931279069767442e-05, | |
| "loss": 3.0115, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_cer": 1.0, | |
| "eval_loss": 2.8972272872924805, | |
| "eval_runtime": 6.4814, | |
| "eval_samples_per_second": 52.612, | |
| "eval_steps_per_second": 6.634, | |
| "eval_wer": 1.0, | |
| "step": 258 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "learning_rate": 2.8963953488372093e-05, | |
| "loss": 2.9304, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_cer": 1.0, | |
| "eval_loss": 2.887695550918579, | |
| "eval_runtime": 6.205, | |
| "eval_samples_per_second": 54.956, | |
| "eval_steps_per_second": 6.93, | |
| "eval_wer": 1.0, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "learning_rate": 2.861511627906977e-05, | |
| "loss": 2.9073, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_cer": 1.0, | |
| "eval_loss": 2.857363224029541, | |
| "eval_runtime": 5.8413, | |
| "eval_samples_per_second": 58.377, | |
| "eval_steps_per_second": 7.361, | |
| "eval_wer": 1.0, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 5.81, | |
| "learning_rate": 2.8266279069767444e-05, | |
| "loss": 2.7919, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_cer": 0.674222480032127, | |
| "eval_loss": 1.9143861532211304, | |
| "eval_runtime": 6.2962, | |
| "eval_samples_per_second": 54.159, | |
| "eval_steps_per_second": 6.829, | |
| "eval_wer": 1.0, | |
| "step": 516 | |
| }, | |
| { | |
| "epoch": 6.98, | |
| "learning_rate": 2.7917441860465118e-05, | |
| "loss": 1.6061, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_cer": 0.07661416268796573, | |
| "eval_loss": 0.543914258480072, | |
| "eval_runtime": 5.8137, | |
| "eval_samples_per_second": 58.654, | |
| "eval_steps_per_second": 7.396, | |
| "eval_wer": 0.2885666745899691, | |
| "step": 602 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_cer": 0.055776181339520774, | |
| "eval_loss": 0.37258943915367126, | |
| "eval_runtime": 6.0487, | |
| "eval_samples_per_second": 56.376, | |
| "eval_steps_per_second": 7.109, | |
| "eval_wer": 0.19491323983836464, | |
| "step": 688 | |
| }, | |
| { | |
| "epoch": 8.14, | |
| "learning_rate": 2.7568604651162792e-05, | |
| "loss": 0.7808, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_cer": 0.05167105439293204, | |
| "eval_loss": 0.2959502637386322, | |
| "eval_runtime": 5.8892, | |
| "eval_samples_per_second": 57.902, | |
| "eval_steps_per_second": 7.301, | |
| "eval_wer": 0.18183979082481577, | |
| "step": 774 | |
| }, | |
| { | |
| "epoch": 9.3, | |
| "learning_rate": 2.7219767441860466e-05, | |
| "loss": 0.5543, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_cer": 0.04774441122662978, | |
| "eval_loss": 0.259117066860199, | |
| "eval_runtime": 5.9755, | |
| "eval_samples_per_second": 57.066, | |
| "eval_steps_per_second": 7.196, | |
| "eval_wer": 0.16876634181126693, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 10.47, | |
| "learning_rate": 2.6870930232558143e-05, | |
| "loss": 0.4721, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_cer": 0.04265762348846549, | |
| "eval_loss": 0.23672477900981903, | |
| "eval_runtime": 6.0691, | |
| "eval_samples_per_second": 56.186, | |
| "eval_steps_per_second": 7.085, | |
| "eval_wer": 0.14452103636795816, | |
| "step": 946 | |
| }, | |
| { | |
| "epoch": 11.63, | |
| "learning_rate": 2.6522093023255817e-05, | |
| "loss": 0.414, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_cer": 0.037615456695372806, | |
| "eval_loss": 0.21672293543815613, | |
| "eval_runtime": 6.2518, | |
| "eval_samples_per_second": 54.545, | |
| "eval_steps_per_second": 6.878, | |
| "eval_wer": 0.12598050867601615, | |
| "step": 1032 | |
| }, | |
| { | |
| "epoch": 12.79, | |
| "learning_rate": 2.617325581395349e-05, | |
| "loss": 0.3819, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_cer": 0.03502744188121904, | |
| "eval_loss": 0.1979241967201233, | |
| "eval_runtime": 5.8651, | |
| "eval_samples_per_second": 58.141, | |
| "eval_steps_per_second": 7.332, | |
| "eval_wer": 0.11504635131922986, | |
| "step": 1118 | |
| }, | |
| { | |
| "epoch": 13.95, | |
| "learning_rate": 2.5824418604651165e-05, | |
| "loss": 0.3376, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_cer": 0.03462585337557449, | |
| "eval_loss": 0.1877206414937973, | |
| "eval_runtime": 6.3492, | |
| "eval_samples_per_second": 53.708, | |
| "eval_steps_per_second": 6.773, | |
| "eval_wer": 0.11694794390301878, | |
| "step": 1204 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_cer": 0.03346570880371246, | |
| "eval_loss": 0.17655867338180542, | |
| "eval_runtime": 6.0868, | |
| "eval_samples_per_second": 56.023, | |
| "eval_steps_per_second": 7.065, | |
| "eval_wer": 0.10839077727596863, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 15.12, | |
| "learning_rate": 2.547558139534884e-05, | |
| "loss": 0.3199, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_cer": 0.032260943286778816, | |
| "eval_loss": 0.175409734249115, | |
| "eval_runtime": 6.3852, | |
| "eval_samples_per_second": 53.405, | |
| "eval_steps_per_second": 6.734, | |
| "eval_wer": 0.10316139767054909, | |
| "step": 1376 | |
| }, | |
| { | |
| "epoch": 16.28, | |
| "learning_rate": 2.512674418604651e-05, | |
| "loss": 0.3174, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_cer": 0.031547008165632946, | |
| "eval_loss": 0.1697303056716919, | |
| "eval_runtime": 6.098, | |
| "eval_samples_per_second": 55.92, | |
| "eval_steps_per_second": 7.051, | |
| "eval_wer": 0.10173520323270739, | |
| "step": 1462 | |
| }, | |
| { | |
| "epoch": 17.44, | |
| "learning_rate": 2.4777906976744186e-05, | |
| "loss": 0.2747, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_cer": 0.030788452099415466, | |
| "eval_loss": 0.1668318510055542, | |
| "eval_runtime": 6.0816, | |
| "eval_samples_per_second": 56.071, | |
| "eval_steps_per_second": 7.071, | |
| "eval_wer": 0.09626812455431424, | |
| "step": 1548 | |
| }, | |
| { | |
| "epoch": 18.6, | |
| "learning_rate": 2.442906976744186e-05, | |
| "loss": 0.2618, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_cer": 0.0300745169782696, | |
| "eval_loss": 0.16257551312446594, | |
| "eval_runtime": 5.9751, | |
| "eval_samples_per_second": 57.07, | |
| "eval_steps_per_second": 7.197, | |
| "eval_wer": 0.09365343475160447, | |
| "step": 1634 | |
| }, | |
| { | |
| "epoch": 19.77, | |
| "learning_rate": 2.4080232558139534e-05, | |
| "loss": 0.2557, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_cer": 0.029851412252911515, | |
| "eval_loss": 0.15971209108829498, | |
| "eval_runtime": 6.1983, | |
| "eval_samples_per_second": 55.015, | |
| "eval_steps_per_second": 6.937, | |
| "eval_wer": 0.09460423104349892, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 20.93, | |
| "learning_rate": 2.3731395348837208e-05, | |
| "loss": 0.2578, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_cer": 0.029628307527553435, | |
| "eval_loss": 0.15854881703853607, | |
| "eval_runtime": 6.6998, | |
| "eval_samples_per_second": 50.897, | |
| "eval_steps_per_second": 6.418, | |
| "eval_wer": 0.09436653197052532, | |
| "step": 1806 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_cer": 0.030163758868412834, | |
| "eval_loss": 0.15489967167377472, | |
| "eval_runtime": 5.8895, | |
| "eval_samples_per_second": 57.899, | |
| "eval_steps_per_second": 7.301, | |
| "eval_wer": 0.09650582362728785, | |
| "step": 1892 | |
| }, | |
| { | |
| "epoch": 22.09, | |
| "learning_rate": 2.3382558139534882e-05, | |
| "loss": 0.2288, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_cer": 0.02837892106554817, | |
| "eval_loss": 0.15013234317302704, | |
| "eval_runtime": 5.9578, | |
| "eval_samples_per_second": 57.236, | |
| "eval_steps_per_second": 7.217, | |
| "eval_wer": 0.09389113382457809, | |
| "step": 1978 | |
| }, | |
| { | |
| "epoch": 23.26, | |
| "learning_rate": 2.303372093023256e-05, | |
| "loss": 0.2269, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_cer": 0.029092856186694036, | |
| "eval_loss": 0.15236006677150726, | |
| "eval_runtime": 6.4802, | |
| "eval_samples_per_second": 52.622, | |
| "eval_steps_per_second": 6.636, | |
| "eval_wer": 0.09436653197052532, | |
| "step": 2064 | |
| }, | |
| { | |
| "epoch": 24.42, | |
| "learning_rate": 2.2684883720930233e-05, | |
| "loss": 0.2327, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_cer": 0.028111195395118468, | |
| "eval_loss": 0.14761987328529358, | |
| "eval_runtime": 5.9425, | |
| "eval_samples_per_second": 57.383, | |
| "eval_steps_per_second": 7.236, | |
| "eval_wer": 0.09032564772997385, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 25.58, | |
| "learning_rate": 2.2336046511627907e-05, | |
| "loss": 0.2024, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_cer": 0.028423542010619786, | |
| "eval_loss": 0.14807896316051483, | |
| "eval_runtime": 6.0178, | |
| "eval_samples_per_second": 56.665, | |
| "eval_steps_per_second": 7.145, | |
| "eval_wer": 0.09032564772997385, | |
| "step": 2236 | |
| }, | |
| { | |
| "epoch": 26.74, | |
| "learning_rate": 2.198720930232558e-05, | |
| "loss": 0.2056, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_cer": 0.02837892106554817, | |
| "eval_loss": 0.1433907002210617, | |
| "eval_runtime": 6.0685, | |
| "eval_samples_per_second": 56.192, | |
| "eval_steps_per_second": 7.086, | |
| "eval_wer": 0.09246493938673639, | |
| "step": 2322 | |
| }, | |
| { | |
| "epoch": 27.91, | |
| "learning_rate": 2.1638372093023255e-05, | |
| "loss": 0.2097, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_cer": 0.028021953504975234, | |
| "eval_loss": 0.1468418389558792, | |
| "eval_runtime": 7.6911, | |
| "eval_samples_per_second": 44.337, | |
| "eval_steps_per_second": 5.591, | |
| "eval_wer": 0.0893748514380794, | |
| "step": 2408 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_cer": 0.02730801838382937, | |
| "eval_loss": 0.14347784221172333, | |
| "eval_runtime": 6.2489, | |
| "eval_samples_per_second": 54.569, | |
| "eval_steps_per_second": 6.881, | |
| "eval_wer": 0.08604706441644877, | |
| "step": 2494 | |
| }, | |
| { | |
| "epoch": 29.07, | |
| "learning_rate": 2.128953488372093e-05, | |
| "loss": 0.2177, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_cer": 0.028111195395118468, | |
| "eval_loss": 0.14982478320598602, | |
| "eval_runtime": 6.7618, | |
| "eval_samples_per_second": 50.43, | |
| "eval_steps_per_second": 6.359, | |
| "eval_wer": 0.08771095792726408, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 30.23, | |
| "learning_rate": 2.0940697674418606e-05, | |
| "loss": 0.1935, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_cer": 0.027798848779617153, | |
| "eval_loss": 0.14517097175121307, | |
| "eval_runtime": 5.9019, | |
| "eval_samples_per_second": 57.778, | |
| "eval_steps_per_second": 7.286, | |
| "eval_wer": 0.08913715236510578, | |
| "step": 2666 | |
| }, | |
| { | |
| "epoch": 31.4, | |
| "learning_rate": 2.059186046511628e-05, | |
| "loss": 0.1918, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_cer": 0.027486502164115835, | |
| "eval_loss": 0.14662335813045502, | |
| "eval_runtime": 6.9556, | |
| "eval_samples_per_second": 49.026, | |
| "eval_steps_per_second": 6.182, | |
| "eval_wer": 0.08485856905158069, | |
| "step": 2752 | |
| }, | |
| { | |
| "epoch": 32.56, | |
| "learning_rate": 2.0243023255813954e-05, | |
| "loss": 0.1805, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_cer": 0.028155816340190085, | |
| "eval_loss": 0.14373721182346344, | |
| "eval_runtime": 6.2565, | |
| "eval_samples_per_second": 54.504, | |
| "eval_steps_per_second": 6.873, | |
| "eval_wer": 0.08889945329213216, | |
| "step": 2838 | |
| }, | |
| { | |
| "epoch": 33.72, | |
| "learning_rate": 1.9894186046511628e-05, | |
| "loss": 0.1805, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_cer": 0.027441881219044218, | |
| "eval_loss": 0.14093732833862305, | |
| "eval_runtime": 6.6898, | |
| "eval_samples_per_second": 50.973, | |
| "eval_steps_per_second": 6.428, | |
| "eval_wer": 0.08699786070834324, | |
| "step": 2924 | |
| }, | |
| { | |
| "epoch": 34.88, | |
| "learning_rate": 1.9545348837209302e-05, | |
| "loss": 0.1835, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_cer": 0.027129534603542903, | |
| "eval_loss": 0.14223171770572662, | |
| "eval_runtime": 6.4219, | |
| "eval_samples_per_second": 53.099, | |
| "eval_steps_per_second": 6.696, | |
| "eval_wer": 0.08557166627050154, | |
| "step": 3010 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_cer": 0.026415599482397037, | |
| "eval_loss": 0.13766661286354065, | |
| "eval_runtime": 7.5399, | |
| "eval_samples_per_second": 45.226, | |
| "eval_steps_per_second": 5.703, | |
| "eval_wer": 0.08509626812455431, | |
| "step": 3096 | |
| }, | |
| { | |
| "epoch": 36.05, | |
| "learning_rate": 1.919651162790698e-05, | |
| "loss": 0.1787, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_cer": 0.026683325152826738, | |
| "eval_loss": 0.13642320036888123, | |
| "eval_runtime": 6.5226, | |
| "eval_samples_per_second": 52.28, | |
| "eval_steps_per_second": 6.592, | |
| "eval_wer": 0.08438317090563346, | |
| "step": 3182 | |
| }, | |
| { | |
| "epoch": 37.21, | |
| "learning_rate": 1.8847674418604653e-05, | |
| "loss": 0.1695, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_cer": 0.02677256704296997, | |
| "eval_loss": 0.14176321029663086, | |
| "eval_runtime": 6.3748, | |
| "eval_samples_per_second": 53.492, | |
| "eval_steps_per_second": 6.745, | |
| "eval_wer": 0.08485856905158069, | |
| "step": 3268 | |
| }, | |
| { | |
| "epoch": 38.37, | |
| "learning_rate": 1.8498837209302327e-05, | |
| "loss": 0.1775, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_cer": 0.02704029271339967, | |
| "eval_loss": 0.1400807499885559, | |
| "eval_runtime": 6.1197, | |
| "eval_samples_per_second": 55.721, | |
| "eval_steps_per_second": 7.026, | |
| "eval_wer": 0.08438317090563346, | |
| "step": 3354 | |
| }, | |
| { | |
| "epoch": 39.53, | |
| "learning_rate": 1.815e-05, | |
| "loss": 0.1763, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_cer": 0.02650484137254027, | |
| "eval_loss": 0.14022426307201385, | |
| "eval_runtime": 5.9433, | |
| "eval_samples_per_second": 57.375, | |
| "eval_steps_per_second": 7.235, | |
| "eval_wer": 0.08153078202995008, | |
| "step": 3440 | |
| }, | |
| { | |
| "epoch": 40.7, | |
| "learning_rate": 1.7801162790697675e-05, | |
| "loss": 0.1702, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_cer": 0.026415599482397037, | |
| "eval_loss": 0.14177961647510529, | |
| "eval_runtime": 6.7266, | |
| "eval_samples_per_second": 50.694, | |
| "eval_steps_per_second": 6.393, | |
| "eval_wer": 0.08295697646779178, | |
| "step": 3526 | |
| }, | |
| { | |
| "epoch": 41.86, | |
| "learning_rate": 1.7452325581395352e-05, | |
| "loss": 0.1569, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_cer": 0.02583552719646602, | |
| "eval_loss": 0.13996584713459015, | |
| "eval_runtime": 5.888, | |
| "eval_samples_per_second": 57.914, | |
| "eval_steps_per_second": 7.303, | |
| "eval_wer": 0.08248157832184455, | |
| "step": 3612 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_cer": 0.026192494757038953, | |
| "eval_loss": 0.14005370438098907, | |
| "eval_runtime": 5.8324, | |
| "eval_samples_per_second": 58.466, | |
| "eval_steps_per_second": 7.373, | |
| "eval_wer": 0.08153078202995008, | |
| "step": 3698 | |
| }, | |
| { | |
| "epoch": 43.02, | |
| "learning_rate": 1.7103488372093026e-05, | |
| "loss": 0.1617, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_cer": 0.026192494757038953, | |
| "eval_loss": 0.14058516919612885, | |
| "eval_runtime": 6.1068, | |
| "eval_samples_per_second": 55.839, | |
| "eval_steps_per_second": 7.041, | |
| "eval_wer": 0.07915379130021392, | |
| "step": 3784 | |
| }, | |
| { | |
| "epoch": 44.19, | |
| "learning_rate": 1.67546511627907e-05, | |
| "loss": 0.1596, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_cer": 0.02637097853732542, | |
| "eval_loss": 0.1394573301076889, | |
| "eval_runtime": 6.5093, | |
| "eval_samples_per_second": 52.386, | |
| "eval_steps_per_second": 6.606, | |
| "eval_wer": 0.0817684811029237, | |
| "step": 3870 | |
| }, | |
| { | |
| "epoch": 45.35, | |
| "learning_rate": 1.6405813953488374e-05, | |
| "loss": 0.1431, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_cer": 0.026192494757038953, | |
| "eval_loss": 0.13823828101158142, | |
| "eval_runtime": 6.0067, | |
| "eval_samples_per_second": 56.77, | |
| "eval_steps_per_second": 7.159, | |
| "eval_wer": 0.08153078202995008, | |
| "step": 3956 | |
| }, | |
| { | |
| "epoch": 46.51, | |
| "learning_rate": 1.6056976744186047e-05, | |
| "loss": 0.158, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_cer": 0.02650484137254027, | |
| "eval_loss": 0.13912436366081238, | |
| "eval_runtime": 6.0425, | |
| "eval_samples_per_second": 56.433, | |
| "eval_steps_per_second": 7.116, | |
| "eval_wer": 0.08129308295697647, | |
| "step": 4042 | |
| }, | |
| { | |
| "epoch": 47.67, | |
| "learning_rate": 1.570813953488372e-05, | |
| "loss": 0.1552, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_cer": 0.026594083262683504, | |
| "eval_loss": 0.13926950097084045, | |
| "eval_runtime": 6.5183, | |
| "eval_samples_per_second": 52.314, | |
| "eval_steps_per_second": 6.597, | |
| "eval_wer": 0.08248157832184455, | |
| "step": 4128 | |
| }, | |
| { | |
| "epoch": 48.84, | |
| "learning_rate": 1.5359302325581395e-05, | |
| "loss": 0.1379, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_cer": 0.025612422471107937, | |
| "eval_loss": 0.13708749413490295, | |
| "eval_runtime": 6.0023, | |
| "eval_samples_per_second": 56.811, | |
| "eval_steps_per_second": 7.164, | |
| "eval_wer": 0.08105538388400285, | |
| "step": 4214 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 1.5010465116279071e-05, | |
| "loss": 0.145, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_cer": 0.02556780152603632, | |
| "eval_loss": 0.1392103135585785, | |
| "eval_runtime": 7.2777, | |
| "eval_samples_per_second": 46.855, | |
| "eval_steps_per_second": 5.908, | |
| "eval_wer": 0.0801045875921084, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 51.0, | |
| "eval_cer": 0.02623711570211057, | |
| "eval_loss": 0.14161016047000885, | |
| "eval_runtime": 6.4267, | |
| "eval_samples_per_second": 53.06, | |
| "eval_steps_per_second": 6.691, | |
| "eval_wer": 0.08200618017589731, | |
| "step": 4386 | |
| }, | |
| { | |
| "epoch": 51.16, | |
| "learning_rate": 1.4661627906976745e-05, | |
| "loss": 0.1647, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_cer": 0.02556780152603632, | |
| "eval_loss": 0.13915704190731049, | |
| "eval_runtime": 6.1093, | |
| "eval_samples_per_second": 55.816, | |
| "eval_steps_per_second": 7.038, | |
| "eval_wer": 0.07891609222724032, | |
| "step": 4472 | |
| }, | |
| { | |
| "epoch": 52.33, | |
| "learning_rate": 1.4312790697674419e-05, | |
| "loss": 0.1493, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 53.0, | |
| "eval_cer": 0.02570166436125117, | |
| "eval_loss": 0.14251555502414703, | |
| "eval_runtime": 6.4895, | |
| "eval_samples_per_second": 52.546, | |
| "eval_steps_per_second": 6.626, | |
| "eval_wer": 0.07939149037318755, | |
| "step": 4558 | |
| }, | |
| { | |
| "epoch": 53.49, | |
| "learning_rate": 1.3963953488372094e-05, | |
| "loss": 0.1492, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "eval_cer": 0.025746285306322787, | |
| "eval_loss": 0.1419229656457901, | |
| "eval_runtime": 5.9077, | |
| "eval_samples_per_second": 57.721, | |
| "eval_steps_per_second": 7.279, | |
| "eval_wer": 0.07962918944616117, | |
| "step": 4644 | |
| }, | |
| { | |
| "epoch": 54.65, | |
| "learning_rate": 1.3615116279069768e-05, | |
| "loss": 0.139, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "eval_cer": 0.024987729240105304, | |
| "eval_loss": 0.13996940851211548, | |
| "eval_runtime": 5.9605, | |
| "eval_samples_per_second": 57.21, | |
| "eval_steps_per_second": 7.214, | |
| "eval_wer": 0.07582600427858331, | |
| "step": 4730 | |
| }, | |
| { | |
| "epoch": 55.81, | |
| "learning_rate": 1.3266279069767444e-05, | |
| "loss": 0.1385, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_cer": 0.02534469680067824, | |
| "eval_loss": 0.1423511654138565, | |
| "eval_runtime": 6.5323, | |
| "eval_samples_per_second": 52.203, | |
| "eval_steps_per_second": 6.583, | |
| "eval_wer": 0.07915379130021392, | |
| "step": 4816 | |
| }, | |
| { | |
| "epoch": 56.98, | |
| "learning_rate": 1.2917441860465116e-05, | |
| "loss": 0.128, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 57.0, | |
| "eval_cer": 0.025255454910535005, | |
| "eval_loss": 0.1403101086616516, | |
| "eval_runtime": 6.4807, | |
| "eval_samples_per_second": 52.618, | |
| "eval_steps_per_second": 6.635, | |
| "eval_wer": 0.08057998573805562, | |
| "step": 4902 | |
| }, | |
| { | |
| "epoch": 57.0, | |
| "step": 4902, | |
| "total_flos": 4.0970110980881424e+19, | |
| "train_loss": 0.6410711561475954, | |
| "train_runtime": 7770.6868, | |
| "train_samples_per_second": 35.299, | |
| "train_steps_per_second": 1.107 | |
| } | |
| ], | |
| "max_steps": 8600, | |
| "num_train_epochs": 100, | |
| "total_flos": 4.0970110980881424e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |