| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.807200929152149, | |
| "eval_steps": 100, | |
| "global_step": 30000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.019357336430507164, | |
| "eval_loss": 3.567659854888916, | |
| "eval_runtime": 160.9588, | |
| "eval_samples_per_second": 35.139, | |
| "eval_steps_per_second": 4.392, | |
| "eval_wer": 1.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.03871467286101433, | |
| "eval_loss": 3.0471677780151367, | |
| "eval_runtime": 158.7973, | |
| "eval_samples_per_second": 35.618, | |
| "eval_steps_per_second": 4.452, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.05807200929152149, | |
| "eval_loss": 2.9665186405181885, | |
| "eval_runtime": 159.308, | |
| "eval_samples_per_second": 35.504, | |
| "eval_steps_per_second": 4.438, | |
| "eval_wer": 1.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.07742934572202866, | |
| "eval_loss": 2.464332103729248, | |
| "eval_runtime": 159.8297, | |
| "eval_samples_per_second": 35.388, | |
| "eval_steps_per_second": 4.423, | |
| "eval_wer": 0.9813195101988413, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.09678668215253582, | |
| "grad_norm": 6.005111217498779, | |
| "learning_rate": 0.00029759999999999997, | |
| "loss": 4.1279, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.09678668215253582, | |
| "eval_loss": 1.625333547592163, | |
| "eval_runtime": 160.6655, | |
| "eval_samples_per_second": 35.204, | |
| "eval_steps_per_second": 4.4, | |
| "eval_wer": 0.9345380430421595, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.11614401858304298, | |
| "eval_loss": 1.24808931350708, | |
| "eval_runtime": 160.4402, | |
| "eval_samples_per_second": 35.253, | |
| "eval_steps_per_second": 4.407, | |
| "eval_wer": 0.8190528157147213, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.13550135501355012, | |
| "eval_loss": 1.0997203588485718, | |
| "eval_runtime": 161.1872, | |
| "eval_samples_per_second": 35.09, | |
| "eval_steps_per_second": 4.386, | |
| "eval_wer": 0.7769735680698432, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.1548586914440573, | |
| "eval_loss": 1.0475263595581055, | |
| "eval_runtime": 161.0626, | |
| "eval_samples_per_second": 35.117, | |
| "eval_steps_per_second": 4.39, | |
| "eval_wer": 0.7339795541718156, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.17421602787456447, | |
| "eval_loss": 0.9692754149436951, | |
| "eval_runtime": 161.3683, | |
| "eval_samples_per_second": 35.05, | |
| "eval_steps_per_second": 4.381, | |
| "eval_wer": 0.7012565999582738, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.19357336430507163, | |
| "grad_norm": 2.414348602294922, | |
| "learning_rate": 0.0002949762711864406, | |
| "loss": 1.0598, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.19357336430507163, | |
| "eval_loss": 0.911480724811554, | |
| "eval_runtime": 162.0835, | |
| "eval_samples_per_second": 34.896, | |
| "eval_steps_per_second": 4.362, | |
| "eval_wer": 0.6749049124552647, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.2129307007355788, | |
| "eval_loss": 0.8823792338371277, | |
| "eval_runtime": 161.5426, | |
| "eval_samples_per_second": 35.012, | |
| "eval_steps_per_second": 4.377, | |
| "eval_wer": 0.65625651971562, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.23228803716608595, | |
| "eval_loss": 0.8609552383422852, | |
| "eval_runtime": 161.6778, | |
| "eval_samples_per_second": 34.983, | |
| "eval_steps_per_second": 4.373, | |
| "eval_wer": 0.6431288215563865, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.2516453735965931, | |
| "eval_loss": 0.8330459594726562, | |
| "eval_runtime": 161.7974, | |
| "eval_samples_per_second": 34.957, | |
| "eval_steps_per_second": 4.37, | |
| "eval_wer": 0.6114169247805363, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.27100271002710025, | |
| "eval_loss": 0.8172780871391296, | |
| "eval_runtime": 161.4977, | |
| "eval_samples_per_second": 35.022, | |
| "eval_steps_per_second": 4.378, | |
| "eval_wer": 0.6017396607340598, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.29036004645760743, | |
| "grad_norm": 4.460846424102783, | |
| "learning_rate": 0.0002898915254237288, | |
| "loss": 0.8546, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.29036004645760743, | |
| "eval_loss": 0.8102588653564453, | |
| "eval_runtime": 161.397, | |
| "eval_samples_per_second": 35.044, | |
| "eval_steps_per_second": 4.381, | |
| "eval_wer": 0.6139204955786298, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.3097173828881146, | |
| "eval_loss": 0.7860382795333862, | |
| "eval_runtime": 162.0586, | |
| "eval_samples_per_second": 34.901, | |
| "eval_steps_per_second": 4.363, | |
| "eval_wer": 0.6077739082986953, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.32907471931862176, | |
| "eval_loss": 0.857550323009491, | |
| "eval_runtime": 161.2906, | |
| "eval_samples_per_second": 35.067, | |
| "eval_steps_per_second": 4.383, | |
| "eval_wer": 0.5990114105053682, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.34843205574912894, | |
| "eval_loss": 0.7555657029151917, | |
| "eval_runtime": 161.7718, | |
| "eval_samples_per_second": 34.963, | |
| "eval_steps_per_second": 4.37, | |
| "eval_wer": 0.5773298454526488, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.3677893921796361, | |
| "eval_loss": 0.7365372180938721, | |
| "eval_runtime": 162.1133, | |
| "eval_samples_per_second": 34.889, | |
| "eval_steps_per_second": 4.361, | |
| "eval_wer": 0.5825777150101908, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.38714672861014326, | |
| "grad_norm": 3.4646999835968018, | |
| "learning_rate": 0.0002848067796610169, | |
| "loss": 0.7776, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.38714672861014326, | |
| "eval_loss": 0.7291606068611145, | |
| "eval_runtime": 162.354, | |
| "eval_samples_per_second": 34.837, | |
| "eval_steps_per_second": 4.355, | |
| "eval_wer": 0.5551989215387332, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.4065040650406504, | |
| "eval_loss": 0.716595470905304, | |
| "eval_runtime": 162.461, | |
| "eval_samples_per_second": 34.815, | |
| "eval_steps_per_second": 4.352, | |
| "eval_wer": 0.5385726436744716, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.4258614014711576, | |
| "eval_loss": 0.7117305397987366, | |
| "eval_runtime": 161.9033, | |
| "eval_samples_per_second": 34.934, | |
| "eval_steps_per_second": 4.367, | |
| "eval_wer": 0.5401774967501726, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.4452187379016647, | |
| "eval_loss": 0.7060667872428894, | |
| "eval_runtime": 162.0901, | |
| "eval_samples_per_second": 34.894, | |
| "eval_steps_per_second": 4.362, | |
| "eval_wer": 0.5388294201665838, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.4645760743321719, | |
| "eval_loss": 0.7044907212257385, | |
| "eval_runtime": 162.5576, | |
| "eval_samples_per_second": 34.794, | |
| "eval_steps_per_second": 4.349, | |
| "eval_wer": 0.5364060920222754, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.48393341076267904, | |
| "grad_norm": 2.70296573638916, | |
| "learning_rate": 0.00027972203389830505, | |
| "loss": 0.706, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.48393341076267904, | |
| "eval_loss": 0.7062936425209045, | |
| "eval_runtime": 162.4753, | |
| "eval_samples_per_second": 34.811, | |
| "eval_steps_per_second": 4.351, | |
| "eval_wer": 0.5428736499173501, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.5032907471931862, | |
| "eval_loss": 0.6941363215446472, | |
| "eval_runtime": 162.6699, | |
| "eval_samples_per_second": 34.77, | |
| "eval_steps_per_second": 4.346, | |
| "eval_wer": 0.5433872029015744, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.5226480836236934, | |
| "eval_loss": 0.6840428113937378, | |
| "eval_runtime": 162.5617, | |
| "eval_samples_per_second": 34.793, | |
| "eval_steps_per_second": 4.349, | |
| "eval_wer": 0.5203094156729952, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.5420054200542005, | |
| "eval_loss": 0.6902298331260681, | |
| "eval_runtime": 162.7532, | |
| "eval_samples_per_second": 34.752, | |
| "eval_steps_per_second": 4.344, | |
| "eval_wer": 0.5593715395355555, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.5613627564847077, | |
| "eval_loss": 0.6594961881637573, | |
| "eval_runtime": 163.1259, | |
| "eval_samples_per_second": 34.673, | |
| "eval_steps_per_second": 4.334, | |
| "eval_wer": 0.5149171093386401, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.5807200929152149, | |
| "grad_norm": 4.962900161743164, | |
| "learning_rate": 0.0002746372881355932, | |
| "loss": 0.7002, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.5807200929152149, | |
| "eval_loss": 0.6767885088920593, | |
| "eval_runtime": 162.9945, | |
| "eval_samples_per_second": 34.701, | |
| "eval_steps_per_second": 4.338, | |
| "eval_wer": 0.525284460207668, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.6000774293457221, | |
| "eval_loss": 0.6656874418258667, | |
| "eval_runtime": 163.0572, | |
| "eval_samples_per_second": 34.687, | |
| "eval_steps_per_second": 4.336, | |
| "eval_wer": 0.5063953395066682, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.6194347657762292, | |
| "eval_loss": 0.6758668422698975, | |
| "eval_runtime": 163.796, | |
| "eval_samples_per_second": 34.531, | |
| "eval_steps_per_second": 4.316, | |
| "eval_wer": 0.5409478262265089, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.6387921022067363, | |
| "eval_loss": 0.6709346175193787, | |
| "eval_runtime": 162.7448, | |
| "eval_samples_per_second": 34.754, | |
| "eval_steps_per_second": 4.344, | |
| "eval_wer": 0.5090914926738457, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 0.6581494386372435, | |
| "eval_loss": 0.6478992104530334, | |
| "eval_runtime": 163.2374, | |
| "eval_samples_per_second": 34.649, | |
| "eval_steps_per_second": 4.331, | |
| "eval_wer": 0.5037473319317617, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.6775067750677507, | |
| "grad_norm": 3.27418851852417, | |
| "learning_rate": 0.0002695525423728813, | |
| "loss": 0.685, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.6775067750677507, | |
| "eval_loss": 0.6378278136253357, | |
| "eval_runtime": 162.9066, | |
| "eval_samples_per_second": 34.719, | |
| "eval_steps_per_second": 4.34, | |
| "eval_wer": 0.5033782157243505, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.6968641114982579, | |
| "eval_loss": 0.6492822170257568, | |
| "eval_runtime": 162.8688, | |
| "eval_samples_per_second": 34.727, | |
| "eval_steps_per_second": 4.341, | |
| "eval_wer": 0.49883648152011684, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.716221447928765, | |
| "eval_loss": 0.6340391635894775, | |
| "eval_runtime": 163.0198, | |
| "eval_samples_per_second": 34.695, | |
| "eval_steps_per_second": 4.337, | |
| "eval_wer": 0.4832694066858179, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 0.7355787843592722, | |
| "eval_loss": 0.6226627826690674, | |
| "eval_runtime": 164.0506, | |
| "eval_samples_per_second": 34.477, | |
| "eval_steps_per_second": 4.31, | |
| "eval_wer": 0.47354399704707034, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 0.7549361207897793, | |
| "eval_loss": 0.6257476210594177, | |
| "eval_runtime": 167.0907, | |
| "eval_samples_per_second": 33.85, | |
| "eval_steps_per_second": 4.231, | |
| "eval_wer": 0.49068382789555615, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 0.7742934572202865, | |
| "grad_norm": 5.494376182556152, | |
| "learning_rate": 0.0002644677966101695, | |
| "loss": 0.6655, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.7742934572202865, | |
| "eval_loss": 0.6420141458511353, | |
| "eval_runtime": 163.4141, | |
| "eval_samples_per_second": 34.611, | |
| "eval_steps_per_second": 4.326, | |
| "eval_wer": 0.49987963601932245, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.7936507936507936, | |
| "eval_loss": 0.6111469268798828, | |
| "eval_runtime": 163.5321, | |
| "eval_samples_per_second": 34.586, | |
| "eval_steps_per_second": 4.323, | |
| "eval_wer": 0.4790646916274815, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 0.8130081300813008, | |
| "eval_loss": 0.6136205196380615, | |
| "eval_runtime": 163.9442, | |
| "eval_samples_per_second": 34.5, | |
| "eval_steps_per_second": 4.312, | |
| "eval_wer": 0.48073373882621045, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 0.832365466511808, | |
| "eval_loss": 0.6218396425247192, | |
| "eval_runtime": 163.3391, | |
| "eval_samples_per_second": 34.627, | |
| "eval_steps_per_second": 4.328, | |
| "eval_wer": 0.48596555985299544, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 0.8517228029423152, | |
| "eval_loss": 0.6084252595901489, | |
| "eval_runtime": 162.8945, | |
| "eval_samples_per_second": 34.722, | |
| "eval_steps_per_second": 4.34, | |
| "eval_wer": 0.4585386207892667, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 0.8710801393728222, | |
| "grad_norm": 3.0379676818847656, | |
| "learning_rate": 0.0002593830508474576, | |
| "loss": 0.6431, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.8710801393728222, | |
| "eval_loss": 0.6008957624435425, | |
| "eval_runtime": 163.8125, | |
| "eval_samples_per_second": 34.527, | |
| "eval_steps_per_second": 4.316, | |
| "eval_wer": 0.4627593843783602, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.8904374758033294, | |
| "eval_loss": 0.6009930968284607, | |
| "eval_runtime": 163.1039, | |
| "eval_samples_per_second": 34.677, | |
| "eval_steps_per_second": 4.335, | |
| "eval_wer": 0.46295196674744427, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 0.9097948122338366, | |
| "eval_loss": 0.5823432207107544, | |
| "eval_runtime": 163.3804, | |
| "eval_samples_per_second": 34.619, | |
| "eval_steps_per_second": 4.327, | |
| "eval_wer": 0.45035387010319206, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 0.9291521486643438, | |
| "eval_loss": 0.6118789315223694, | |
| "eval_runtime": 163.7255, | |
| "eval_samples_per_second": 34.546, | |
| "eval_steps_per_second": 4.318, | |
| "eval_wer": 0.4630001123397153, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 0.948509485094851, | |
| "eval_loss": 0.6001989245414734, | |
| "eval_runtime": 163.3492, | |
| "eval_samples_per_second": 34.625, | |
| "eval_steps_per_second": 4.328, | |
| "eval_wer": 0.4600150856189116, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 0.9678668215253581, | |
| "grad_norm": 3.1605985164642334, | |
| "learning_rate": 0.00025430847457627115, | |
| "loss": 0.6235, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.9678668215253581, | |
| "eval_loss": 0.5892329216003418, | |
| "eval_runtime": 163.5255, | |
| "eval_samples_per_second": 34.588, | |
| "eval_steps_per_second": 4.323, | |
| "eval_wer": 0.4551844778610518, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.9872241579558653, | |
| "eval_loss": 0.5673592686653137, | |
| "eval_runtime": 163.343, | |
| "eval_samples_per_second": 34.627, | |
| "eval_steps_per_second": 4.328, | |
| "eval_wer": 0.44889345380430423, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 1.0065814943863725, | |
| "eval_loss": 0.5792257785797119, | |
| "eval_runtime": 162.8592, | |
| "eval_samples_per_second": 34.729, | |
| "eval_steps_per_second": 4.341, | |
| "eval_wer": 0.43167338030203334, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 1.0259388308168795, | |
| "eval_loss": 0.5752869844436646, | |
| "eval_runtime": 162.7856, | |
| "eval_samples_per_second": 34.745, | |
| "eval_steps_per_second": 4.343, | |
| "eval_wer": 0.43331033043924827, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 1.0452961672473868, | |
| "eval_loss": 0.5698733925819397, | |
| "eval_runtime": 161.9949, | |
| "eval_samples_per_second": 34.915, | |
| "eval_steps_per_second": 4.364, | |
| "eval_wer": 0.44619730063712665, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 1.064653503677894, | |
| "grad_norm": 0.8791279792785645, | |
| "learning_rate": 0.0002492338983050847, | |
| "loss": 0.5527, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.064653503677894, | |
| "eval_loss": 0.5666691660881042, | |
| "eval_runtime": 162.0517, | |
| "eval_samples_per_second": 34.902, | |
| "eval_steps_per_second": 4.363, | |
| "eval_wer": 0.43639164834459404, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.084010840108401, | |
| "eval_loss": 0.5558171272277832, | |
| "eval_runtime": 161.9847, | |
| "eval_samples_per_second": 34.917, | |
| "eval_steps_per_second": 4.365, | |
| "eval_wer": 0.42945868305756607, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 1.1033681765389083, | |
| "eval_loss": 0.5602455139160156, | |
| "eval_runtime": 162.701, | |
| "eval_samples_per_second": 34.763, | |
| "eval_steps_per_second": 4.345, | |
| "eval_wer": 0.422349183932211, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 1.1227255129694154, | |
| "eval_loss": 0.559140145778656, | |
| "eval_runtime": 162.4402, | |
| "eval_samples_per_second": 34.819, | |
| "eval_steps_per_second": 4.352, | |
| "eval_wer": 0.41942835133443535, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 1.1420828493999227, | |
| "eval_loss": 0.5399234890937805, | |
| "eval_runtime": 162.3316, | |
| "eval_samples_per_second": 34.842, | |
| "eval_steps_per_second": 4.355, | |
| "eval_wer": 0.418818507165669, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 1.1614401858304297, | |
| "grad_norm": 0.9803772568702698, | |
| "learning_rate": 0.00024414915254237287, | |
| "loss": 0.533, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.1614401858304297, | |
| "eval_loss": 0.545900821685791, | |
| "eval_runtime": 161.6822, | |
| "eval_samples_per_second": 34.982, | |
| "eval_steps_per_second": 4.373, | |
| "eval_wer": 0.431063536133267, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.1807975222609368, | |
| "eval_loss": 0.5347985625267029, | |
| "eval_runtime": 161.8121, | |
| "eval_samples_per_second": 34.954, | |
| "eval_steps_per_second": 4.369, | |
| "eval_wer": 0.41183739628637, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 1.2001548586914441, | |
| "eval_loss": 0.5453631281852722, | |
| "eval_runtime": 161.8802, | |
| "eval_samples_per_second": 34.939, | |
| "eval_steps_per_second": 4.367, | |
| "eval_wer": 0.4176309158896503, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 1.2195121951219512, | |
| "eval_loss": 0.5442932844161987, | |
| "eval_runtime": 162.1767, | |
| "eval_samples_per_second": 34.876, | |
| "eval_steps_per_second": 4.359, | |
| "eval_wer": 0.42157885445587456, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 1.2388695315524583, | |
| "eval_loss": 0.5382806658744812, | |
| "eval_runtime": 161.5364, | |
| "eval_samples_per_second": 35.014, | |
| "eval_steps_per_second": 4.377, | |
| "eval_wer": 0.40962269904190274, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 1.2582268679829656, | |
| "grad_norm": 2.7026009559631348, | |
| "learning_rate": 0.00023906440677966102, | |
| "loss": 0.5228, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.2582268679829656, | |
| "eval_loss": 0.540704071521759, | |
| "eval_runtime": 161.6157, | |
| "eval_samples_per_second": 34.997, | |
| "eval_steps_per_second": 4.375, | |
| "eval_wer": 0.41260772576270643, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.2775842044134726, | |
| "eval_loss": 0.5527251362800598, | |
| "eval_runtime": 161.6568, | |
| "eval_samples_per_second": 34.988, | |
| "eval_steps_per_second": 4.373, | |
| "eval_wer": 0.41426072443067835, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 1.29694154084398, | |
| "eval_loss": 0.5312824845314026, | |
| "eval_runtime": 161.7821, | |
| "eval_samples_per_second": 34.961, | |
| "eval_steps_per_second": 4.37, | |
| "eval_wer": 0.40811413715074385, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 1.316298877274487, | |
| "eval_loss": 0.533909797668457, | |
| "eval_runtime": 161.8361, | |
| "eval_samples_per_second": 34.949, | |
| "eval_steps_per_second": 4.369, | |
| "eval_wer": 0.4150471024377718, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 1.3356562137049943, | |
| "eval_loss": 0.523649275302887, | |
| "eval_runtime": 161.5476, | |
| "eval_samples_per_second": 35.011, | |
| "eval_steps_per_second": 4.376, | |
| "eval_wer": 0.4120781242477251, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 1.3550135501355014, | |
| "grad_norm": 0.709751546382904, | |
| "learning_rate": 0.00023397966101694912, | |
| "loss": 0.5204, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.3550135501355014, | |
| "eval_loss": 0.5527586340904236, | |
| "eval_runtime": 162.1137, | |
| "eval_samples_per_second": 34.889, | |
| "eval_steps_per_second": 4.361, | |
| "eval_wer": 0.4165877613904447, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.3743708865660085, | |
| "eval_loss": 0.5330629944801331, | |
| "eval_runtime": 161.681, | |
| "eval_samples_per_second": 34.982, | |
| "eval_steps_per_second": 4.373, | |
| "eval_wer": 0.40559451782189343, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 1.3937282229965158, | |
| "eval_loss": 0.5242415070533752, | |
| "eval_runtime": 162.2742, | |
| "eval_samples_per_second": 34.855, | |
| "eval_steps_per_second": 4.357, | |
| "eval_wer": 0.4058833913755196, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 1.4130855594270229, | |
| "eval_loss": 0.5309507250785828, | |
| "eval_runtime": 163.2224, | |
| "eval_samples_per_second": 34.652, | |
| "eval_steps_per_second": 4.332, | |
| "eval_wer": 0.4092856798960055, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 1.43244289585753, | |
| "eval_loss": 0.5278186798095703, | |
| "eval_runtime": 162.9755, | |
| "eval_samples_per_second": 34.705, | |
| "eval_steps_per_second": 4.338, | |
| "eval_wer": 0.4063006531752018, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 1.4518002322880372, | |
| "grad_norm": 0.9905166029930115, | |
| "learning_rate": 0.00022889491525423728, | |
| "loss": 0.5199, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.4518002322880372, | |
| "eval_loss": 0.5168124437332153, | |
| "eval_runtime": 162.1336, | |
| "eval_samples_per_second": 34.885, | |
| "eval_steps_per_second": 4.361, | |
| "eval_wer": 0.3955802346295197, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.4711575687185443, | |
| "eval_loss": 0.5236623287200928, | |
| "eval_runtime": 161.1426, | |
| "eval_samples_per_second": 35.099, | |
| "eval_steps_per_second": 4.387, | |
| "eval_wer": 0.40235271460897754, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 1.4905149051490514, | |
| "eval_loss": 0.5316073894500732, | |
| "eval_runtime": 162.1382, | |
| "eval_samples_per_second": 34.884, | |
| "eval_steps_per_second": 4.36, | |
| "eval_wer": 0.4179358379740335, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 1.5098722415795587, | |
| "eval_loss": 0.5182381868362427, | |
| "eval_runtime": 161.7911, | |
| "eval_samples_per_second": 34.959, | |
| "eval_steps_per_second": 4.37, | |
| "eval_wer": 0.40326748086212705, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 1.5292295780100658, | |
| "eval_loss": 0.5175392627716064, | |
| "eval_runtime": 161.5725, | |
| "eval_samples_per_second": 35.006, | |
| "eval_steps_per_second": 4.376, | |
| "eval_wer": 0.3983887275119963, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 1.5485869144405728, | |
| "grad_norm": 0.8261615037918091, | |
| "learning_rate": 0.00022382033898305084, | |
| "loss": 0.5066, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.5485869144405728, | |
| "eval_loss": 0.5138476490974426, | |
| "eval_runtime": 162.4044, | |
| "eval_samples_per_second": 34.827, | |
| "eval_steps_per_second": 4.353, | |
| "eval_wer": 0.39492224486848226, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.5679442508710801, | |
| "eval_loss": 0.515566885471344, | |
| "eval_runtime": 162.5299, | |
| "eval_samples_per_second": 34.8, | |
| "eval_steps_per_second": 4.35, | |
| "eval_wer": 0.4016305307249121, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 1.5873015873015874, | |
| "eval_loss": 0.5131089091300964, | |
| "eval_runtime": 162.835, | |
| "eval_samples_per_second": 34.735, | |
| "eval_steps_per_second": 4.342, | |
| "eval_wer": 0.39793936865080004, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 1.6066589237320945, | |
| "eval_loss": 0.5139849185943604, | |
| "eval_runtime": 162.5015, | |
| "eval_samples_per_second": 34.806, | |
| "eval_steps_per_second": 4.351, | |
| "eval_wer": 0.39413586686138885, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 1.6260162601626016, | |
| "eval_loss": 0.5224258303642273, | |
| "eval_runtime": 162.1349, | |
| "eval_samples_per_second": 34.885, | |
| "eval_steps_per_second": 4.361, | |
| "eval_wer": 0.39853316428880936, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 1.645373596593109, | |
| "grad_norm": 1.0760446786880493, | |
| "learning_rate": 0.00021873559322033897, | |
| "loss": 0.502, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.645373596593109, | |
| "eval_loss": 0.5274536609649658, | |
| "eval_runtime": 162.3162, | |
| "eval_samples_per_second": 34.846, | |
| "eval_steps_per_second": 4.356, | |
| "eval_wer": 0.40023430854905234, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.664730933023616, | |
| "eval_loss": 0.5054244995117188, | |
| "eval_runtime": 162.7623, | |
| "eval_samples_per_second": 34.75, | |
| "eval_steps_per_second": 4.344, | |
| "eval_wer": 0.3860955529521272, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 1.684088269454123, | |
| "eval_loss": 0.5144466161727905, | |
| "eval_runtime": 162.1114, | |
| "eval_samples_per_second": 34.89, | |
| "eval_steps_per_second": 4.361, | |
| "eval_wer": 0.3912631798558842, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 1.7034456058846303, | |
| "eval_loss": 0.5017980933189392, | |
| "eval_runtime": 162.1144, | |
| "eval_samples_per_second": 34.889, | |
| "eval_steps_per_second": 4.361, | |
| "eval_wer": 0.3860955529521272, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 1.7228029423151374, | |
| "eval_loss": 0.5001707673072815, | |
| "eval_runtime": 162.7958, | |
| "eval_samples_per_second": 34.743, | |
| "eval_steps_per_second": 4.343, | |
| "eval_wer": 0.39978494968785605, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 1.7421602787456445, | |
| "grad_norm": 2.3791110515594482, | |
| "learning_rate": 0.00021366101694915253, | |
| "loss": 0.4965, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.7421602787456445, | |
| "eval_loss": 0.5074877142906189, | |
| "eval_runtime": 162.5487, | |
| "eval_samples_per_second": 34.796, | |
| "eval_steps_per_second": 4.349, | |
| "eval_wer": 0.38896823995763186, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.7615176151761518, | |
| "eval_loss": 0.4928957521915436, | |
| "eval_runtime": 162.2035, | |
| "eval_samples_per_second": 34.87, | |
| "eval_steps_per_second": 4.359, | |
| "eval_wer": 0.3865449118133235, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 1.7808749516066589, | |
| "eval_loss": 0.49622705578804016, | |
| "eval_runtime": 162.8808, | |
| "eval_samples_per_second": 34.725, | |
| "eval_steps_per_second": 4.341, | |
| "eval_wer": 0.38559804849866, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 1.800232288037166, | |
| "eval_loss": 0.49036508798599243, | |
| "eval_runtime": 162.3886, | |
| "eval_samples_per_second": 34.83, | |
| "eval_steps_per_second": 4.354, | |
| "eval_wer": 0.3759689300444544, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 1.8195896244676733, | |
| "eval_loss": 0.49964088201522827, | |
| "eval_runtime": 162.526, | |
| "eval_samples_per_second": 34.801, | |
| "eval_steps_per_second": 4.35, | |
| "eval_wer": 0.3901237341721365, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 1.8389469608981805, | |
| "grad_norm": 1.2548748254776, | |
| "learning_rate": 0.00020857627118644066, | |
| "loss": 0.4776, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.8389469608981805, | |
| "eval_loss": 0.4899130165576935, | |
| "eval_runtime": 162.5723, | |
| "eval_samples_per_second": 34.791, | |
| "eval_steps_per_second": 4.349, | |
| "eval_wer": 0.37616151241353857, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.8583042973286876, | |
| "eval_loss": 0.4918155074119568, | |
| "eval_runtime": 162.1914, | |
| "eval_samples_per_second": 34.872, | |
| "eval_steps_per_second": 4.359, | |
| "eval_wer": 0.37948355828023944, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 1.8776616337591947, | |
| "eval_loss": 0.49148374795913696, | |
| "eval_runtime": 162.2813, | |
| "eval_samples_per_second": 34.853, | |
| "eval_steps_per_second": 4.357, | |
| "eval_wer": 0.37980452889537963, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 1.897018970189702, | |
| "eval_loss": 0.4841060936450958, | |
| "eval_runtime": 162.2556, | |
| "eval_samples_per_second": 34.859, | |
| "eval_steps_per_second": 4.357, | |
| "eval_wer": 0.37060872077161333, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 1.916376306620209, | |
| "eval_loss": 0.4834117293357849, | |
| "eval_runtime": 163.5813, | |
| "eval_samples_per_second": 34.576, | |
| "eval_steps_per_second": 4.322, | |
| "eval_wer": 0.37728490956652916, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 1.9357336430507162, | |
| "grad_norm": 1.011767029762268, | |
| "learning_rate": 0.00020349152542372878, | |
| "loss": 0.4752, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.9357336430507162, | |
| "eval_loss": 0.4831894338130951, | |
| "eval_runtime": 162.4305, | |
| "eval_samples_per_second": 34.821, | |
| "eval_steps_per_second": 4.353, | |
| "eval_wer": 0.3711704193481087, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.9550909794812235, | |
| "eval_loss": 0.4890592396259308, | |
| "eval_runtime": 162.3568, | |
| "eval_samples_per_second": 34.837, | |
| "eval_steps_per_second": 4.355, | |
| "eval_wer": 0.37829596700422075, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 1.9744483159117305, | |
| "eval_loss": 0.4786697328090668, | |
| "eval_runtime": 163.1452, | |
| "eval_samples_per_second": 34.669, | |
| "eval_steps_per_second": 4.334, | |
| "eval_wer": 0.3783441125964918, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 1.9938056523422376, | |
| "eval_loss": 0.4726457893848419, | |
| "eval_runtime": 162.2371, | |
| "eval_samples_per_second": 34.863, | |
| "eval_steps_per_second": 4.358, | |
| "eval_wer": 0.37141114730946384, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 2.013162988772745, | |
| "eval_loss": 0.49166908860206604, | |
| "eval_runtime": 162.2356, | |
| "eval_samples_per_second": 34.863, | |
| "eval_steps_per_second": 4.358, | |
| "eval_wer": 0.37320858275424884, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 2.032520325203252, | |
| "grad_norm": 0.9389815926551819, | |
| "learning_rate": 0.00019840677966101694, | |
| "loss": 0.4587, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.032520325203252, | |
| "eval_loss": 0.48015162348747253, | |
| "eval_runtime": 162.7857, | |
| "eval_samples_per_second": 34.745, | |
| "eval_steps_per_second": 4.343, | |
| "eval_wer": 0.37264688417775355, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.051877661633759, | |
| "eval_loss": 0.4883776903152466, | |
| "eval_runtime": 162.7407, | |
| "eval_samples_per_second": 34.755, | |
| "eval_steps_per_second": 4.344, | |
| "eval_wer": 0.3825327791240712, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 2.0712349980642664, | |
| "eval_loss": 0.4841337502002716, | |
| "eval_runtime": 162.7772, | |
| "eval_samples_per_second": 34.747, | |
| "eval_steps_per_second": 4.343, | |
| "eval_wer": 0.37845645231179087, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 2.0905923344947737, | |
| "eval_loss": 0.4809282422065735, | |
| "eval_runtime": 162.8688, | |
| "eval_samples_per_second": 34.727, | |
| "eval_steps_per_second": 4.341, | |
| "eval_wer": 0.3738184269230152, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 2.1099496709252805, | |
| "eval_loss": 0.47966596484184265, | |
| "eval_runtime": 163.5921, | |
| "eval_samples_per_second": 34.574, | |
| "eval_steps_per_second": 4.322, | |
| "eval_wer": 0.3713469531864358, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 2.129307007355788, | |
| "grad_norm": 0.6634272933006287, | |
| "learning_rate": 0.0001933220338983051, | |
| "loss": 0.3967, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.129307007355788, | |
| "eval_loss": 0.4866289794445038, | |
| "eval_runtime": 162.8573, | |
| "eval_samples_per_second": 34.73, | |
| "eval_steps_per_second": 4.341, | |
| "eval_wer": 0.37497392113751987, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.148664343786295, | |
| "eval_loss": 0.4938376843929291, | |
| "eval_runtime": 163.5145, | |
| "eval_samples_per_second": 34.59, | |
| "eval_steps_per_second": 4.324, | |
| "eval_wer": 0.3749097270144918, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 2.168021680216802, | |
| "eval_loss": 0.48603999614715576, | |
| "eval_runtime": 162.8433, | |
| "eval_samples_per_second": 34.733, | |
| "eval_steps_per_second": 4.342, | |
| "eval_wer": 0.36796071319670687, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 2.1873790166473093, | |
| "eval_loss": 0.4849016070365906, | |
| "eval_runtime": 162.5249, | |
| "eval_samples_per_second": 34.801, | |
| "eval_steps_per_second": 4.35, | |
| "eval_wer": 0.369966779541333, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 2.2067363530778166, | |
| "eval_loss": 0.49077799916267395, | |
| "eval_runtime": 162.3783, | |
| "eval_samples_per_second": 34.832, | |
| "eval_steps_per_second": 4.354, | |
| "eval_wer": 0.36377204666912744, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 2.226093689508324, | |
| "grad_norm": 0.9132543206214905, | |
| "learning_rate": 0.0001882372881355932, | |
| "loss": 0.406, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.226093689508324, | |
| "eval_loss": 0.4797042906284332, | |
| "eval_runtime": 162.1749, | |
| "eval_samples_per_second": 34.876, | |
| "eval_steps_per_second": 4.359, | |
| "eval_wer": 0.3678644220121648, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.2454510259388307, | |
| "eval_loss": 0.48121991753578186, | |
| "eval_runtime": 161.7765, | |
| "eval_samples_per_second": 34.962, | |
| "eval_steps_per_second": 4.37, | |
| "eval_wer": 0.37585659032915536, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 2.264808362369338, | |
| "eval_loss": 0.47043049335479736, | |
| "eval_runtime": 163.1227, | |
| "eval_samples_per_second": 34.673, | |
| "eval_steps_per_second": 4.334, | |
| "eval_wer": 0.361268475871034, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 2.2841656987998453, | |
| "eval_loss": 0.4715932607650757, | |
| "eval_runtime": 163.2336, | |
| "eval_samples_per_second": 34.65, | |
| "eval_steps_per_second": 4.331, | |
| "eval_wer": 0.36345107605398724, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 2.303523035230352, | |
| "eval_loss": 0.4676753580570221, | |
| "eval_runtime": 163.378, | |
| "eval_samples_per_second": 34.619, | |
| "eval_steps_per_second": 4.327, | |
| "eval_wer": 0.3635473672385293, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 2.3228803716608595, | |
| "grad_norm": 0.6134137511253357, | |
| "learning_rate": 0.00018315254237288135, | |
| "loss": 0.4088, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.3228803716608595, | |
| "eval_loss": 0.47054949402809143, | |
| "eval_runtime": 162.7912, | |
| "eval_samples_per_second": 34.744, | |
| "eval_steps_per_second": 4.343, | |
| "eval_wer": 0.36399672609972555, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.3422377080913668, | |
| "eval_loss": 0.4782082140445709, | |
| "eval_runtime": 162.8776, | |
| "eval_samples_per_second": 34.725, | |
| "eval_steps_per_second": 4.341, | |
| "eval_wer": 0.35905377862656673, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 2.3615950445218736, | |
| "eval_loss": 0.4795554578304291, | |
| "eval_runtime": 163.357, | |
| "eval_samples_per_second": 34.624, | |
| "eval_steps_per_second": 4.328, | |
| "eval_wer": 0.36128452440179104, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 2.380952380952381, | |
| "eval_loss": 0.47130346298217773, | |
| "eval_runtime": 163.1765, | |
| "eval_samples_per_second": 34.662, | |
| "eval_steps_per_second": 4.333, | |
| "eval_wer": 0.3558119754136509, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 2.4003097173828882, | |
| "eval_loss": 0.47632816433906555, | |
| "eval_runtime": 163.2851, | |
| "eval_samples_per_second": 34.639, | |
| "eval_steps_per_second": 4.33, | |
| "eval_wer": 0.3588772447882396, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 2.419667053813395, | |
| "grad_norm": 0.6301820874214172, | |
| "learning_rate": 0.0001780779661016949, | |
| "loss": 0.407, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.419667053813395, | |
| "eval_loss": 0.46899136900901794, | |
| "eval_runtime": 163.5219, | |
| "eval_samples_per_second": 34.589, | |
| "eval_steps_per_second": 4.324, | |
| "eval_wer": 0.3565181107669593, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.4390243902439024, | |
| "eval_loss": 0.4686334431171417, | |
| "eval_runtime": 163.9109, | |
| "eval_samples_per_second": 34.507, | |
| "eval_steps_per_second": 4.313, | |
| "eval_wer": 0.35767360498146394, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 2.4583817266744097, | |
| "eval_loss": 0.467680424451828, | |
| "eval_runtime": 163.7544, | |
| "eval_samples_per_second": 34.54, | |
| "eval_steps_per_second": 4.317, | |
| "eval_wer": 0.3584278859270434, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 2.4777390631049165, | |
| "eval_loss": 0.46144554018974304, | |
| "eval_runtime": 163.8198, | |
| "eval_samples_per_second": 34.526, | |
| "eval_steps_per_second": 4.316, | |
| "eval_wer": 0.35765755645070696, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 2.497096399535424, | |
| "eval_loss": 0.455834299325943, | |
| "eval_runtime": 163.2394, | |
| "eval_samples_per_second": 34.649, | |
| "eval_steps_per_second": 4.331, | |
| "eval_wer": 0.35992039928744524, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 2.516453735965931, | |
| "grad_norm": 0.41953468322753906, | |
| "learning_rate": 0.00017299322033898304, | |
| "loss": 0.3855, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.516453735965931, | |
| "eval_loss": 0.4555678367614746, | |
| "eval_runtime": 164.8785, | |
| "eval_samples_per_second": 34.304, | |
| "eval_steps_per_second": 4.288, | |
| "eval_wer": 0.3565341592977163, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.535811072396438, | |
| "eval_loss": 0.4600988030433655, | |
| "eval_runtime": 163.6778, | |
| "eval_samples_per_second": 34.556, | |
| "eval_steps_per_second": 4.319, | |
| "eval_wer": 0.3558280239444079, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 2.5551684088269453, | |
| "eval_loss": 0.4650043547153473, | |
| "eval_runtime": 163.58, | |
| "eval_samples_per_second": 34.576, | |
| "eval_steps_per_second": 4.322, | |
| "eval_wer": 0.354303413522492, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 2.5745257452574526, | |
| "eval_loss": 0.4737236201763153, | |
| "eval_runtime": 163.9532, | |
| "eval_samples_per_second": 34.498, | |
| "eval_steps_per_second": 4.312, | |
| "eval_wer": 0.35483301503747333, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 2.59388308168796, | |
| "eval_loss": 0.45056912302970886, | |
| "eval_runtime": 163.6462, | |
| "eval_samples_per_second": 34.562, | |
| "eval_steps_per_second": 4.32, | |
| "eval_wer": 0.3534367928616135, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 2.6132404181184667, | |
| "grad_norm": 1.5978127717971802, | |
| "learning_rate": 0.0001679186440677966, | |
| "loss": 0.3748, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 2.6132404181184667, | |
| "eval_loss": 0.4606887698173523, | |
| "eval_runtime": 163.0481, | |
| "eval_samples_per_second": 34.689, | |
| "eval_steps_per_second": 4.336, | |
| "eval_wer": 0.3589253903805107, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 2.632597754548974, | |
| "eval_loss": 0.45494645833969116, | |
| "eval_runtime": 162.9584, | |
| "eval_samples_per_second": 34.708, | |
| "eval_steps_per_second": 4.339, | |
| "eval_wer": 0.35372566641523967, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 2.6519550909794813, | |
| "eval_loss": 0.4562608301639557, | |
| "eval_runtime": 164.1384, | |
| "eval_samples_per_second": 34.459, | |
| "eval_steps_per_second": 4.307, | |
| "eval_wer": 0.36409301728426763, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 2.6713124274099886, | |
| "eval_loss": 0.44666969776153564, | |
| "eval_runtime": 163.7423, | |
| "eval_samples_per_second": 34.542, | |
| "eval_steps_per_second": 4.318, | |
| "eval_wer": 0.34369533469210894, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 2.6906697638404955, | |
| "eval_loss": 0.4536294639110565, | |
| "eval_runtime": 163.3892, | |
| "eval_samples_per_second": 34.617, | |
| "eval_steps_per_second": 4.327, | |
| "eval_wer": 0.35446389883006213, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 2.710027100271003, | |
| "grad_norm": 0.6591352224349976, | |
| "learning_rate": 0.00016283389830508475, | |
| "loss": 0.3888, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.710027100271003, | |
| "eval_loss": 0.4504217505455017, | |
| "eval_runtime": 163.3199, | |
| "eval_samples_per_second": 34.631, | |
| "eval_steps_per_second": 4.329, | |
| "eval_wer": 0.3509653191250341, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.72938443670151, | |
| "eval_loss": 0.44697660207748413, | |
| "eval_runtime": 163.5943, | |
| "eval_samples_per_second": 34.573, | |
| "eval_steps_per_second": 4.322, | |
| "eval_wer": 0.3602092728410714, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 2.748741773132017, | |
| "eval_loss": 0.45640549063682556, | |
| "eval_runtime": 163.6363, | |
| "eval_samples_per_second": 34.564, | |
| "eval_steps_per_second": 4.321, | |
| "eval_wer": 0.3539022002535668, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 2.7680991095625243, | |
| "eval_loss": 0.45214343070983887, | |
| "eval_runtime": 164.011, | |
| "eval_samples_per_second": 34.485, | |
| "eval_steps_per_second": 4.311, | |
| "eval_wer": 0.3561971401518191, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 2.7874564459930316, | |
| "eval_loss": 0.4452911913394928, | |
| "eval_runtime": 163.9028, | |
| "eval_samples_per_second": 34.508, | |
| "eval_steps_per_second": 4.314, | |
| "eval_wer": 0.35221710452408084, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 2.8068137824235384, | |
| "grad_norm": 0.6879103779792786, | |
| "learning_rate": 0.00015774915254237285, | |
| "loss": 0.376, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.8068137824235384, | |
| "eval_loss": 0.45518526434898376, | |
| "eval_runtime": 164.0636, | |
| "eval_samples_per_second": 34.474, | |
| "eval_steps_per_second": 4.309, | |
| "eval_wer": 0.35170355153985655, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.8261711188540457, | |
| "eval_loss": 0.45344606041908264, | |
| "eval_runtime": 163.5444, | |
| "eval_samples_per_second": 34.584, | |
| "eval_steps_per_second": 4.323, | |
| "eval_wer": 0.3549774518142864, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 2.845528455284553, | |
| "eval_loss": 0.45520055294036865, | |
| "eval_runtime": 163.3824, | |
| "eval_samples_per_second": 34.618, | |
| "eval_steps_per_second": 4.327, | |
| "eval_wer": 0.3405016770714641, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 2.86488579171506, | |
| "eval_loss": 0.45560306310653687, | |
| "eval_runtime": 164.0292, | |
| "eval_samples_per_second": 34.482, | |
| "eval_steps_per_second": 4.31, | |
| "eval_wer": 0.35138258092471636, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 2.884243128145567, | |
| "eval_loss": 0.44232824444770813, | |
| "eval_runtime": 164.3662, | |
| "eval_samples_per_second": 34.411, | |
| "eval_steps_per_second": 4.301, | |
| "eval_wer": 0.3467606040666977, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 2.9036004645760745, | |
| "grad_norm": 0.5280432105064392, | |
| "learning_rate": 0.00015267457627118642, | |
| "loss": 0.379, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.9036004645760745, | |
| "eval_loss": 0.43873003125190735, | |
| "eval_runtime": 163.5232, | |
| "eval_samples_per_second": 34.588, | |
| "eval_steps_per_second": 4.324, | |
| "eval_wer": 0.34268427725441736, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.9229578010065813, | |
| "eval_loss": 0.4372723400592804, | |
| "eval_runtime": 163.4018, | |
| "eval_samples_per_second": 34.614, | |
| "eval_steps_per_second": 4.327, | |
| "eval_wer": 0.34364718909983794, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 2.9423151374370886, | |
| "eval_loss": 0.4399470090866089, | |
| "eval_runtime": 164.335, | |
| "eval_samples_per_second": 34.418, | |
| "eval_steps_per_second": 4.302, | |
| "eval_wer": 0.33870424162667906, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 2.961672473867596, | |
| "eval_loss": 0.44378861784935, | |
| "eval_runtime": 164.0477, | |
| "eval_samples_per_second": 34.478, | |
| "eval_steps_per_second": 4.31, | |
| "eval_wer": 0.3380462518656417, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 2.9810298102981028, | |
| "eval_loss": 0.436974436044693, | |
| "eval_runtime": 163.6525, | |
| "eval_samples_per_second": 34.561, | |
| "eval_steps_per_second": 4.32, | |
| "eval_wer": 0.3430694419925856, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 3.00038714672861, | |
| "grad_norm": 0.660970151424408, | |
| "learning_rate": 0.00014758983050847457, | |
| "loss": 0.3731, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 3.00038714672861, | |
| "eval_loss": 0.43810611963272095, | |
| "eval_runtime": 162.9215, | |
| "eval_samples_per_second": 34.716, | |
| "eval_steps_per_second": 4.34, | |
| "eval_wer": 0.33413041036093144, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 3.0197444831591174, | |
| "eval_loss": 0.45139721035957336, | |
| "eval_runtime": 164.2748, | |
| "eval_samples_per_second": 34.43, | |
| "eval_steps_per_second": 4.304, | |
| "eval_wer": 0.3286418128420343, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 3.0391018195896247, | |
| "eval_loss": 0.43782538175582886, | |
| "eval_runtime": 164.2188, | |
| "eval_samples_per_second": 34.442, | |
| "eval_steps_per_second": 4.305, | |
| "eval_wer": 0.3340180706456324, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 3.0584591560201315, | |
| "eval_loss": 0.44340020418167114, | |
| "eval_runtime": 163.4459, | |
| "eval_samples_per_second": 34.605, | |
| "eval_steps_per_second": 4.326, | |
| "eval_wer": 0.3441446935533052, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 3.077816492450639, | |
| "eval_loss": 0.44192126393318176, | |
| "eval_runtime": 164.2009, | |
| "eval_samples_per_second": 34.446, | |
| "eval_steps_per_second": 4.306, | |
| "eval_wer": 0.3399239299642118, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 3.097173828881146, | |
| "grad_norm": 0.6999391913414001, | |
| "learning_rate": 0.0001425050847457627, | |
| "loss": 0.3176, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 3.097173828881146, | |
| "eval_loss": 0.4407601058483124, | |
| "eval_runtime": 164.7219, | |
| "eval_samples_per_second": 34.337, | |
| "eval_steps_per_second": 4.292, | |
| "eval_wer": 0.3335366147229221, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 3.116531165311653, | |
| "eval_loss": 0.436761736869812, | |
| "eval_runtime": 164.2581, | |
| "eval_samples_per_second": 34.434, | |
| "eval_steps_per_second": 4.304, | |
| "eval_wer": 0.33584760315193146, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 3.1358885017421603, | |
| "eval_loss": 0.4477560520172119, | |
| "eval_runtime": 163.8979, | |
| "eval_samples_per_second": 34.509, | |
| "eval_steps_per_second": 4.314, | |
| "eval_wer": 0.3400523182102678, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 3.1552458381726676, | |
| "eval_loss": 0.4414171576499939, | |
| "eval_runtime": 164.4332, | |
| "eval_samples_per_second": 34.397, | |
| "eval_steps_per_second": 4.3, | |
| "eval_wer": 0.3373882621046043, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 3.1746031746031744, | |
| "eval_loss": 0.4476623833179474, | |
| "eval_runtime": 163.6322, | |
| "eval_samples_per_second": 34.565, | |
| "eval_steps_per_second": 4.321, | |
| "eval_wer": 0.335013079552567, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 3.1939605110336817, | |
| "grad_norm": 0.5408484935760498, | |
| "learning_rate": 0.00013742033898305083, | |
| "loss": 0.3201, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 3.1939605110336817, | |
| "eval_loss": 0.4305751919746399, | |
| "eval_runtime": 163.8253, | |
| "eval_samples_per_second": 34.525, | |
| "eval_steps_per_second": 4.316, | |
| "eval_wer": 0.32917141435701563, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 3.213317847464189, | |
| "eval_loss": 0.4534677267074585, | |
| "eval_runtime": 163.8666, | |
| "eval_samples_per_second": 34.516, | |
| "eval_steps_per_second": 4.314, | |
| "eval_wer": 0.32941214231837074, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 3.2326751838946963, | |
| "eval_loss": 0.4379562437534332, | |
| "eval_runtime": 164.1184, | |
| "eval_samples_per_second": 34.463, | |
| "eval_steps_per_second": 4.308, | |
| "eval_wer": 0.33408226476866043, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 3.252032520325203, | |
| "eval_loss": 0.43677663803100586, | |
| "eval_runtime": 163.6899, | |
| "eval_samples_per_second": 34.553, | |
| "eval_steps_per_second": 4.319, | |
| "eval_wer": 0.33252555728523053, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 3.2713898567557105, | |
| "eval_loss": 0.4359833598136902, | |
| "eval_runtime": 164.1551, | |
| "eval_samples_per_second": 34.455, | |
| "eval_steps_per_second": 4.307, | |
| "eval_wer": 0.33043924828681936, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 3.290747193186218, | |
| "grad_norm": 0.5537161231040955, | |
| "learning_rate": 0.00013233559322033898, | |
| "loss": 0.3101, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 3.290747193186218, | |
| "eval_loss": 0.4347226917743683, | |
| "eval_runtime": 164.2964, | |
| "eval_samples_per_second": 34.426, | |
| "eval_steps_per_second": 4.303, | |
| "eval_wer": 0.32812825985781, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 3.3101045296167246, | |
| "eval_loss": 0.4375491738319397, | |
| "eval_runtime": 162.8938, | |
| "eval_samples_per_second": 34.722, | |
| "eval_steps_per_second": 4.34, | |
| "eval_wer": 0.3284973760652212, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 3.329461866047232, | |
| "eval_loss": 0.4491961896419525, | |
| "eval_runtime": 163.5698, | |
| "eval_samples_per_second": 34.579, | |
| "eval_steps_per_second": 4.322, | |
| "eval_wer": 0.33032690857152025, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 3.3488192024777392, | |
| "eval_loss": 0.4268127977848053, | |
| "eval_runtime": 164.9468, | |
| "eval_samples_per_second": 34.29, | |
| "eval_steps_per_second": 4.286, | |
| "eval_wer": 0.3284652790037072, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 3.368176538908246, | |
| "eval_loss": 0.4377237558364868, | |
| "eval_runtime": 164.0847, | |
| "eval_samples_per_second": 34.47, | |
| "eval_steps_per_second": 4.309, | |
| "eval_wer": 0.3269888141740624, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 3.3875338753387534, | |
| "grad_norm": 0.5330023765563965, | |
| "learning_rate": 0.00012726101694915254, | |
| "loss": 0.2963, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 3.3875338753387534, | |
| "eval_loss": 0.42494186758995056, | |
| "eval_runtime": 163.0698, | |
| "eval_samples_per_second": 34.685, | |
| "eval_steps_per_second": 4.336, | |
| "eval_wer": 0.3322527322623614, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 3.4068912117692607, | |
| "eval_loss": 0.4404699206352234, | |
| "eval_runtime": 164.4408, | |
| "eval_samples_per_second": 34.395, | |
| "eval_steps_per_second": 4.299, | |
| "eval_wer": 0.3338736338688193, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 3.4262485481997675, | |
| "eval_loss": 0.43636277318000793, | |
| "eval_runtime": 163.9679, | |
| "eval_samples_per_second": 34.495, | |
| "eval_steps_per_second": 4.312, | |
| "eval_wer": 0.3285615701882493, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 3.445605884630275, | |
| "eval_loss": 0.4350505769252777, | |
| "eval_runtime": 163.6477, | |
| "eval_samples_per_second": 34.562, | |
| "eval_steps_per_second": 4.32, | |
| "eval_wer": 0.3309207042095296, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 3.464963221060782, | |
| "eval_loss": 0.42997920513153076, | |
| "eval_runtime": 163.7592, | |
| "eval_samples_per_second": 34.539, | |
| "eval_steps_per_second": 4.317, | |
| "eval_wer": 0.322880390300268, | |
| "step": 17900 | |
| }, | |
| { | |
| "epoch": 3.484320557491289, | |
| "grad_norm": 0.31616127490997314, | |
| "learning_rate": 0.00012217627118644067, | |
| "loss": 0.3062, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 3.484320557491289, | |
| "eval_loss": 0.42307358980178833, | |
| "eval_runtime": 164.6469, | |
| "eval_samples_per_second": 34.352, | |
| "eval_steps_per_second": 4.294, | |
| "eval_wer": 0.32523952432154835, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 3.5036778939217963, | |
| "eval_loss": 0.4325993061065674, | |
| "eval_runtime": 164.1143, | |
| "eval_samples_per_second": 34.464, | |
| "eval_steps_per_second": 4.308, | |
| "eval_wer": 0.32326555503843624, | |
| "step": 18100 | |
| }, | |
| { | |
| "epoch": 3.5230352303523036, | |
| "eval_loss": 0.43141353130340576, | |
| "eval_runtime": 163.522, | |
| "eval_samples_per_second": 34.589, | |
| "eval_steps_per_second": 4.324, | |
| "eval_wer": 0.3282405995731091, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 3.5423925667828104, | |
| "eval_loss": 0.4343957006931305, | |
| "eval_runtime": 163.8089, | |
| "eval_samples_per_second": 34.528, | |
| "eval_steps_per_second": 4.316, | |
| "eval_wer": 0.32894673492641746, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 3.5617499032133177, | |
| "eval_loss": 0.42664915323257446, | |
| "eval_runtime": 166.5849, | |
| "eval_samples_per_second": 33.953, | |
| "eval_steps_per_second": 4.244, | |
| "eval_wer": 0.32207796376241754, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 3.581107239643825, | |
| "grad_norm": 0.40817028284072876, | |
| "learning_rate": 0.00011710169491525424, | |
| "loss": 0.2968, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 3.581107239643825, | |
| "eval_loss": 0.4305819571018219, | |
| "eval_runtime": 164.3676, | |
| "eval_samples_per_second": 34.411, | |
| "eval_steps_per_second": 4.301, | |
| "eval_wer": 0.32161255637046426, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 3.600464576074332, | |
| "eval_loss": 0.4318545460700989, | |
| "eval_runtime": 166.4377, | |
| "eval_samples_per_second": 33.983, | |
| "eval_steps_per_second": 4.248, | |
| "eval_wer": 0.3238914477379596, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 3.619821912504839, | |
| "eval_loss": 0.4271145164966583, | |
| "eval_runtime": 166.5812, | |
| "eval_samples_per_second": 33.953, | |
| "eval_steps_per_second": 4.244, | |
| "eval_wer": 0.3232013609154082, | |
| "step": 18700 | |
| }, | |
| { | |
| "epoch": 3.6391792489353465, | |
| "eval_loss": 0.41837719082832336, | |
| "eval_runtime": 164.5624, | |
| "eval_samples_per_second": 34.37, | |
| "eval_steps_per_second": 4.296, | |
| "eval_wer": 0.32641106706681006, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 3.658536585365854, | |
| "eval_loss": 0.4237981140613556, | |
| "eval_runtime": 165.5252, | |
| "eval_samples_per_second": 34.17, | |
| "eval_steps_per_second": 4.271, | |
| "eval_wer": 0.31997560623324933, | |
| "step": 18900 | |
| }, | |
| { | |
| "epoch": 3.6778939217963607, | |
| "grad_norm": 0.9548519253730774, | |
| "learning_rate": 0.00011201694915254236, | |
| "loss": 0.3191, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 3.6778939217963607, | |
| "eval_loss": 0.41389960050582886, | |
| "eval_runtime": 163.7093, | |
| "eval_samples_per_second": 34.549, | |
| "eval_steps_per_second": 4.319, | |
| "eval_wer": 0.3225915167466418, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 3.697251258226868, | |
| "eval_loss": 0.42384064197540283, | |
| "eval_runtime": 164.2841, | |
| "eval_samples_per_second": 34.428, | |
| "eval_steps_per_second": 4.304, | |
| "eval_wer": 0.3159955706055111, | |
| "step": 19100 | |
| }, | |
| { | |
| "epoch": 3.7166085946573753, | |
| "eval_loss": 0.4176156520843506, | |
| "eval_runtime": 163.8287, | |
| "eval_samples_per_second": 34.524, | |
| "eval_steps_per_second": 4.315, | |
| "eval_wer": 0.319301567941455, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 3.7359659310878826, | |
| "eval_loss": 0.4196203351020813, | |
| "eval_runtime": 165.0023, | |
| "eval_samples_per_second": 34.278, | |
| "eval_steps_per_second": 4.285, | |
| "eval_wer": 0.3202644797868755, | |
| "step": 19300 | |
| }, | |
| { | |
| "epoch": 3.7553232675183894, | |
| "eval_loss": 0.409524530172348, | |
| "eval_runtime": 164.4937, | |
| "eval_samples_per_second": 34.384, | |
| "eval_steps_per_second": 4.298, | |
| "eval_wer": 0.3181621222577073, | |
| "step": 19400 | |
| }, | |
| { | |
| "epoch": 3.7746806039488967, | |
| "grad_norm": 0.43373510241508484, | |
| "learning_rate": 0.00010693220338983049, | |
| "loss": 0.2921, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 3.7746806039488967, | |
| "eval_loss": 0.41209807991981506, | |
| "eval_runtime": 164.8329, | |
| "eval_samples_per_second": 34.314, | |
| "eval_steps_per_second": 4.289, | |
| "eval_wer": 0.31665356036654846, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 3.794037940379404, | |
| "eval_loss": 0.4112759530544281, | |
| "eval_runtime": 164.1863, | |
| "eval_samples_per_second": 34.449, | |
| "eval_steps_per_second": 4.306, | |
| "eval_wer": 0.31455120283738025, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 3.813395276809911, | |
| "eval_loss": 0.4094259738922119, | |
| "eval_runtime": 164.4841, | |
| "eval_samples_per_second": 34.386, | |
| "eval_steps_per_second": 4.298, | |
| "eval_wer": 0.3160758132592961, | |
| "step": 19700 | |
| }, | |
| { | |
| "epoch": 3.832752613240418, | |
| "eval_loss": 0.40931811928749084, | |
| "eval_runtime": 164.2911, | |
| "eval_samples_per_second": 34.427, | |
| "eval_steps_per_second": 4.303, | |
| "eval_wer": 0.31386111601482886, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 3.8521099496709255, | |
| "eval_loss": 0.41117748618125916, | |
| "eval_runtime": 165.1732, | |
| "eval_samples_per_second": 34.243, | |
| "eval_steps_per_second": 4.28, | |
| "eval_wer": 0.31731155012758583, | |
| "step": 19900 | |
| }, | |
| { | |
| "epoch": 3.8714672861014323, | |
| "grad_norm": 0.5022397637367249, | |
| "learning_rate": 0.00010184745762711863, | |
| "loss": 0.3007, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 3.8714672861014323, | |
| "eval_loss": 0.4092504680156708, | |
| "eval_runtime": 163.9434, | |
| "eval_samples_per_second": 34.5, | |
| "eval_steps_per_second": 4.312, | |
| "eval_wer": 0.31593137648248304, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 3.8908246225319396, | |
| "eval_loss": 0.4147598147392273, | |
| "eval_runtime": 164.6303, | |
| "eval_samples_per_second": 34.356, | |
| "eval_steps_per_second": 4.294, | |
| "eval_wer": 0.31565855145961386, | |
| "step": 20100 | |
| }, | |
| { | |
| "epoch": 3.910181958962447, | |
| "eval_loss": 0.41137251257896423, | |
| "eval_runtime": 164.6634, | |
| "eval_samples_per_second": 34.349, | |
| "eval_steps_per_second": 4.294, | |
| "eval_wer": 0.3150326587600905, | |
| "step": 20200 | |
| }, | |
| { | |
| "epoch": 3.9295392953929538, | |
| "eval_loss": 0.4155375361442566, | |
| "eval_runtime": 164.3406, | |
| "eval_samples_per_second": 34.416, | |
| "eval_steps_per_second": 4.302, | |
| "eval_wer": 0.31456725136813724, | |
| "step": 20300 | |
| }, | |
| { | |
| "epoch": 3.948896631823461, | |
| "eval_loss": 0.4075925648212433, | |
| "eval_runtime": 164.3692, | |
| "eval_samples_per_second": 34.41, | |
| "eval_steps_per_second": 4.301, | |
| "eval_wer": 0.3135722424612027, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 3.9682539682539684, | |
| "grad_norm": 0.6109060049057007, | |
| "learning_rate": 9.67728813559322e-05, | |
| "loss": 0.296, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 3.9682539682539684, | |
| "eval_loss": 0.4066578149795532, | |
| "eval_runtime": 164.7852, | |
| "eval_samples_per_second": 34.323, | |
| "eval_steps_per_second": 4.29, | |
| "eval_wer": 0.3125611850235111, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 3.987611304684475, | |
| "eval_loss": 0.40839362144470215, | |
| "eval_runtime": 164.8883, | |
| "eval_samples_per_second": 34.302, | |
| "eval_steps_per_second": 4.288, | |
| "eval_wer": 0.3150487072908475, | |
| "step": 20600 | |
| }, | |
| { | |
| "epoch": 4.006968641114983, | |
| "eval_loss": 0.4150494635105133, | |
| "eval_runtime": 164.1525, | |
| "eval_samples_per_second": 34.456, | |
| "eval_steps_per_second": 4.307, | |
| "eval_wer": 0.312432796777455, | |
| "step": 20700 | |
| }, | |
| { | |
| "epoch": 4.02632597754549, | |
| "eval_loss": 0.41322341561317444, | |
| "eval_runtime": 164.6726, | |
| "eval_samples_per_second": 34.347, | |
| "eval_steps_per_second": 4.293, | |
| "eval_wer": 0.3132512718460625, | |
| "step": 20800 | |
| }, | |
| { | |
| "epoch": 4.045683313975997, | |
| "eval_loss": 0.4182606339454651, | |
| "eval_runtime": 164.9667, | |
| "eval_samples_per_second": 34.286, | |
| "eval_steps_per_second": 4.286, | |
| "eval_wer": 0.31464749402192227, | |
| "step": 20900 | |
| }, | |
| { | |
| "epoch": 4.065040650406504, | |
| "grad_norm": 0.9771650433540344, | |
| "learning_rate": 9.168813559322032e-05, | |
| "loss": 0.2611, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 4.065040650406504, | |
| "eval_loss": 0.41840454936027527, | |
| "eval_runtime": 164.8893, | |
| "eval_samples_per_second": 34.302, | |
| "eval_steps_per_second": 4.288, | |
| "eval_wer": 0.30952801271043634, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 4.084397986837011, | |
| "eval_loss": 0.4167742431163788, | |
| "eval_runtime": 165.1089, | |
| "eval_samples_per_second": 34.256, | |
| "eval_steps_per_second": 4.282, | |
| "eval_wer": 0.30845276114971676, | |
| "step": 21100 | |
| }, | |
| { | |
| "epoch": 4.103755323267518, | |
| "eval_loss": 0.42244288325309753, | |
| "eval_runtime": 164.9167, | |
| "eval_samples_per_second": 34.296, | |
| "eval_steps_per_second": 4.287, | |
| "eval_wer": 0.31015390540995974, | |
| "step": 21200 | |
| }, | |
| { | |
| "epoch": 4.123112659698026, | |
| "eval_loss": 0.4187394678592682, | |
| "eval_runtime": 164.7166, | |
| "eval_samples_per_second": 34.338, | |
| "eval_steps_per_second": 4.292, | |
| "eval_wer": 0.30456901670652053, | |
| "step": 21300 | |
| }, | |
| { | |
| "epoch": 4.142469996128533, | |
| "eval_loss": 0.41454723477363586, | |
| "eval_runtime": 164.38, | |
| "eval_samples_per_second": 34.408, | |
| "eval_steps_per_second": 4.301, | |
| "eval_wer": 0.3110044775400812, | |
| "step": 21400 | |
| }, | |
| { | |
| "epoch": 4.16182733255904, | |
| "grad_norm": 0.8976078629493713, | |
| "learning_rate": 8.660338983050847e-05, | |
| "loss": 0.2431, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 4.16182733255904, | |
| "eval_loss": 0.42720434069633484, | |
| "eval_runtime": 165.0533, | |
| "eval_samples_per_second": 34.268, | |
| "eval_steps_per_second": 4.283, | |
| "eval_wer": 0.31071560398645504, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 4.181184668989547, | |
| "eval_loss": 0.41736435890197754, | |
| "eval_runtime": 164.6627, | |
| "eval_samples_per_second": 34.349, | |
| "eval_steps_per_second": 4.294, | |
| "eval_wer": 0.3069923448508289, | |
| "step": 21600 | |
| }, | |
| { | |
| "epoch": 4.200542005420054, | |
| "eval_loss": 0.41904589533805847, | |
| "eval_runtime": 165.6169, | |
| "eval_samples_per_second": 34.151, | |
| "eval_steps_per_second": 4.269, | |
| "eval_wer": 0.3085811493957728, | |
| "step": 21700 | |
| }, | |
| { | |
| "epoch": 4.219899341850561, | |
| "eval_loss": 0.41643446683883667, | |
| "eval_runtime": 165.1417, | |
| "eval_samples_per_second": 34.249, | |
| "eval_steps_per_second": 4.281, | |
| "eval_wer": 0.3050825696907448, | |
| "step": 21800 | |
| }, | |
| { | |
| "epoch": 4.239256678281069, | |
| "eval_loss": 0.41955476999282837, | |
| "eval_runtime": 165.2591, | |
| "eval_samples_per_second": 34.225, | |
| "eval_steps_per_second": 4.278, | |
| "eval_wer": 0.30777872285792235, | |
| "step": 21900 | |
| }, | |
| { | |
| "epoch": 4.258614014711576, | |
| "grad_norm": 1.5854851007461548, | |
| "learning_rate": 8.15186440677966e-05, | |
| "loss": 0.2453, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 4.258614014711576, | |
| "eval_loss": 0.42485129833221436, | |
| "eval_runtime": 164.7312, | |
| "eval_samples_per_second": 34.335, | |
| "eval_steps_per_second": 4.292, | |
| "eval_wer": 0.30915889650302514, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 4.2779713511420825, | |
| "eval_loss": 0.4246067404747009, | |
| "eval_runtime": 164.9099, | |
| "eval_samples_per_second": 34.298, | |
| "eval_steps_per_second": 4.287, | |
| "eval_wer": 0.30736146105824014, | |
| "step": 22100 | |
| }, | |
| { | |
| "epoch": 4.29732868757259, | |
| "eval_loss": 0.4166228771209717, | |
| "eval_runtime": 164.9564, | |
| "eval_samples_per_second": 34.288, | |
| "eval_steps_per_second": 4.286, | |
| "eval_wer": 0.30740960665051115, | |
| "step": 22200 | |
| }, | |
| { | |
| "epoch": 4.316686024003097, | |
| "eval_loss": 0.4192067086696625, | |
| "eval_runtime": 165.6591, | |
| "eval_samples_per_second": 34.142, | |
| "eval_steps_per_second": 4.268, | |
| "eval_wer": 0.3027555327309785, | |
| "step": 22300 | |
| }, | |
| { | |
| "epoch": 4.336043360433604, | |
| "eval_loss": 0.41863906383514404, | |
| "eval_runtime": 164.9558, | |
| "eval_samples_per_second": 34.288, | |
| "eval_steps_per_second": 4.286, | |
| "eval_wer": 0.3020975429699411, | |
| "step": 22400 | |
| }, | |
| { | |
| "epoch": 4.355400696864112, | |
| "grad_norm": 1.1900339126586914, | |
| "learning_rate": 7.645423728813559e-05, | |
| "loss": 0.2336, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 4.355400696864112, | |
| "eval_loss": 0.4268459677696228, | |
| "eval_runtime": 166.7137, | |
| "eval_samples_per_second": 33.926, | |
| "eval_steps_per_second": 4.241, | |
| "eval_wer": 0.3083885670266887, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 4.3747580332946185, | |
| "eval_loss": 0.4346672296524048, | |
| "eval_runtime": 170.3751, | |
| "eval_samples_per_second": 33.197, | |
| "eval_steps_per_second": 4.15, | |
| "eval_wer": 0.307104684566128, | |
| "step": 22600 | |
| }, | |
| { | |
| "epoch": 4.394115369725126, | |
| "eval_loss": 0.47525468468666077, | |
| "eval_runtime": 164.9807, | |
| "eval_samples_per_second": 34.283, | |
| "eval_steps_per_second": 4.285, | |
| "eval_wer": 0.3208582754248849, | |
| "step": 22700 | |
| }, | |
| { | |
| "epoch": 4.413472706155633, | |
| "eval_loss": 0.582381546497345, | |
| "eval_runtime": 165.1397, | |
| "eval_samples_per_second": 34.25, | |
| "eval_steps_per_second": 4.281, | |
| "eval_wer": 0.415416218645183, | |
| "step": 22800 | |
| }, | |
| { | |
| "epoch": 4.43283004258614, | |
| "eval_loss": 0.5073803067207336, | |
| "eval_runtime": 165.3352, | |
| "eval_samples_per_second": 34.209, | |
| "eval_steps_per_second": 4.276, | |
| "eval_wer": 0.3415448315706697, | |
| "step": 22900 | |
| }, | |
| { | |
| "epoch": 4.452187379016648, | |
| "grad_norm": 1.5807456970214844, | |
| "learning_rate": 7.136949152542373e-05, | |
| "loss": 0.3426, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 4.452187379016648, | |
| "eval_loss": 0.6242379546165466, | |
| "eval_runtime": 164.8642, | |
| "eval_samples_per_second": 34.307, | |
| "eval_steps_per_second": 4.288, | |
| "eval_wer": 0.41979746754184655, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 4.471544715447155, | |
| "eval_loss": 0.5862211585044861, | |
| "eval_runtime": 164.8283, | |
| "eval_samples_per_second": 34.314, | |
| "eval_steps_per_second": 4.289, | |
| "eval_wer": 0.4200702925647157, | |
| "step": 23100 | |
| }, | |
| { | |
| "epoch": 4.4909020518776614, | |
| "eval_loss": 0.6151086091995239, | |
| "eval_runtime": 165.0995, | |
| "eval_samples_per_second": 34.258, | |
| "eval_steps_per_second": 4.282, | |
| "eval_wer": 0.39638266116737014, | |
| "step": 23200 | |
| }, | |
| { | |
| "epoch": 4.510259388308169, | |
| "eval_loss": 0.5640283226966858, | |
| "eval_runtime": 164.849, | |
| "eval_samples_per_second": 34.31, | |
| "eval_steps_per_second": 4.289, | |
| "eval_wer": 0.3685705573654732, | |
| "step": 23300 | |
| }, | |
| { | |
| "epoch": 4.529616724738676, | |
| "eval_loss": 0.6589744091033936, | |
| "eval_runtime": 164.8194, | |
| "eval_samples_per_second": 34.316, | |
| "eval_steps_per_second": 4.29, | |
| "eval_wer": 0.4647494021922293, | |
| "step": 23400 | |
| }, | |
| { | |
| "epoch": 4.548974061169183, | |
| "grad_norm": 1.0218427181243896, | |
| "learning_rate": 6.628474576271186e-05, | |
| "loss": 0.4541, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 4.548974061169183, | |
| "eval_loss": 0.6010532975196838, | |
| "eval_runtime": 165.0253, | |
| "eval_samples_per_second": 34.274, | |
| "eval_steps_per_second": 4.284, | |
| "eval_wer": 0.3959974964292019, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 4.568331397599691, | |
| "eval_loss": 0.5802894830703735, | |
| "eval_runtime": 166.0838, | |
| "eval_samples_per_second": 34.055, | |
| "eval_steps_per_second": 4.257, | |
| "eval_wer": 0.39505063311453836, | |
| "step": 23600 | |
| }, | |
| { | |
| "epoch": 4.5876887340301975, | |
| "eval_loss": 0.5762883424758911, | |
| "eval_runtime": 165.1308, | |
| "eval_samples_per_second": 34.252, | |
| "eval_steps_per_second": 4.281, | |
| "eval_wer": 0.3910545489560431, | |
| "step": 23700 | |
| }, | |
| { | |
| "epoch": 4.607046070460704, | |
| "eval_loss": 0.5418487787246704, | |
| "eval_runtime": 165.1407, | |
| "eval_samples_per_second": 34.25, | |
| "eval_steps_per_second": 4.281, | |
| "eval_wer": 0.36550528799088444, | |
| "step": 23800 | |
| }, | |
| { | |
| "epoch": 4.626403406891212, | |
| "eval_loss": 0.5546759366989136, | |
| "eval_runtime": 165.8678, | |
| "eval_samples_per_second": 34.099, | |
| "eval_steps_per_second": 4.262, | |
| "eval_wer": 0.38877565758854776, | |
| "step": 23900 | |
| }, | |
| { | |
| "epoch": 4.645760743321719, | |
| "grad_norm": 32.50680923461914, | |
| "learning_rate": 6.12e-05, | |
| "loss": 0.4145, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 4.645760743321719, | |
| "eval_loss": 0.5300523638725281, | |
| "eval_runtime": 164.9724, | |
| "eval_samples_per_second": 34.285, | |
| "eval_steps_per_second": 4.286, | |
| "eval_wer": 0.3608030684790807, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 4.665118079752226, | |
| "eval_loss": 0.573882520198822, | |
| "eval_runtime": 165.2226, | |
| "eval_samples_per_second": 34.233, | |
| "eval_steps_per_second": 4.279, | |
| "eval_wer": 0.39927139670363176, | |
| "step": 24100 | |
| }, | |
| { | |
| "epoch": 4.6844754161827336, | |
| "eval_loss": 0.5775899887084961, | |
| "eval_runtime": 165.3691, | |
| "eval_samples_per_second": 34.202, | |
| "eval_steps_per_second": 4.275, | |
| "eval_wer": 0.39816404808139816, | |
| "step": 24200 | |
| }, | |
| { | |
| "epoch": 4.70383275261324, | |
| "eval_loss": 0.5412492156028748, | |
| "eval_runtime": 164.9818, | |
| "eval_samples_per_second": 34.283, | |
| "eval_steps_per_second": 4.285, | |
| "eval_wer": 0.37078525460994044, | |
| "step": 24300 | |
| }, | |
| { | |
| "epoch": 4.723190089043747, | |
| "eval_loss": 0.5329325199127197, | |
| "eval_runtime": 165.8065, | |
| "eval_samples_per_second": 34.112, | |
| "eval_steps_per_second": 4.264, | |
| "eval_wer": 0.37044823546404326, | |
| "step": 24400 | |
| }, | |
| { | |
| "epoch": 4.742547425474255, | |
| "grad_norm": 1.8765805959701538, | |
| "learning_rate": 5.611525423728813e-05, | |
| "loss": 0.3834, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 4.742547425474255, | |
| "eval_loss": 0.5299070477485657, | |
| "eval_runtime": 165.3917, | |
| "eval_samples_per_second": 34.198, | |
| "eval_steps_per_second": 4.275, | |
| "eval_wer": 0.3732246312850059, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 4.761904761904762, | |
| "eval_loss": 0.5424681901931763, | |
| "eval_runtime": 165.4071, | |
| "eval_samples_per_second": 34.194, | |
| "eval_steps_per_second": 4.274, | |
| "eval_wer": 0.3928519844008281, | |
| "step": 24600 | |
| }, | |
| { | |
| "epoch": 4.781262098335269, | |
| "eval_loss": 0.5111268758773804, | |
| "eval_runtime": 165.4914, | |
| "eval_samples_per_second": 34.177, | |
| "eval_steps_per_second": 4.272, | |
| "eval_wer": 0.3585241771115854, | |
| "step": 24700 | |
| }, | |
| { | |
| "epoch": 4.8006194347657765, | |
| "eval_loss": 0.5076457858085632, | |
| "eval_runtime": 165.6732, | |
| "eval_samples_per_second": 34.14, | |
| "eval_steps_per_second": 4.267, | |
| "eval_wer": 0.35033942642551075, | |
| "step": 24800 | |
| }, | |
| { | |
| "epoch": 4.819976771196283, | |
| "eval_loss": 0.5261921882629395, | |
| "eval_runtime": 165.2946, | |
| "eval_samples_per_second": 34.218, | |
| "eval_steps_per_second": 4.277, | |
| "eval_wer": 0.3681372470350339, | |
| "step": 24900 | |
| }, | |
| { | |
| "epoch": 4.83933410762679, | |
| "grad_norm": 5.934371471405029, | |
| "learning_rate": 5.1030508474576264e-05, | |
| "loss": 0.3719, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 4.83933410762679, | |
| "eval_loss": 0.547415018081665, | |
| "eval_runtime": 165.6997, | |
| "eval_samples_per_second": 34.134, | |
| "eval_steps_per_second": 4.267, | |
| "eval_wer": 0.3833031086004076, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 4.858691444057298, | |
| "eval_loss": 0.5746738910675049, | |
| "eval_runtime": 165.8407, | |
| "eval_samples_per_second": 34.105, | |
| "eval_steps_per_second": 4.263, | |
| "eval_wer": 0.40389337356165045, | |
| "step": 25100 | |
| }, | |
| { | |
| "epoch": 4.878048780487805, | |
| "eval_loss": 0.5188133120536804, | |
| "eval_runtime": 165.5746, | |
| "eval_samples_per_second": 34.16, | |
| "eval_steps_per_second": 4.27, | |
| "eval_wer": 0.3503073293639967, | |
| "step": 25200 | |
| }, | |
| { | |
| "epoch": 4.897406116918312, | |
| "eval_loss": 0.5522667169570923, | |
| "eval_runtime": 165.1011, | |
| "eval_samples_per_second": 34.258, | |
| "eval_steps_per_second": 4.282, | |
| "eval_wer": 0.3865609603440805, | |
| "step": 25300 | |
| }, | |
| { | |
| "epoch": 4.916763453348819, | |
| "eval_loss": 0.5302358865737915, | |
| "eval_runtime": 165.871, | |
| "eval_samples_per_second": 34.099, | |
| "eval_steps_per_second": 4.262, | |
| "eval_wer": 0.36446213349167883, | |
| "step": 25400 | |
| }, | |
| { | |
| "epoch": 4.936120789779326, | |
| "grad_norm": 1.1752023696899414, | |
| "learning_rate": 4.595593220338983e-05, | |
| "loss": 0.3798, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 4.936120789779326, | |
| "eval_loss": 0.5099266767501831, | |
| "eval_runtime": 165.9652, | |
| "eval_samples_per_second": 34.079, | |
| "eval_steps_per_second": 4.26, | |
| "eval_wer": 0.3499542616873425, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 4.955478126209833, | |
| "eval_loss": 0.4823363125324249, | |
| "eval_runtime": 164.8602, | |
| "eval_samples_per_second": 34.308, | |
| "eval_steps_per_second": 4.288, | |
| "eval_wer": 0.33761294153520244, | |
| "step": 25600 | |
| }, | |
| { | |
| "epoch": 4.974835462640341, | |
| "eval_loss": 0.4805842936038971, | |
| "eval_runtime": 166.8028, | |
| "eval_samples_per_second": 33.908, | |
| "eval_steps_per_second": 4.239, | |
| "eval_wer": 0.3357352634366324, | |
| "step": 25700 | |
| }, | |
| { | |
| "epoch": 4.994192799070848, | |
| "eval_loss": 0.4942820370197296, | |
| "eval_runtime": 165.4552, | |
| "eval_samples_per_second": 34.184, | |
| "eval_steps_per_second": 4.273, | |
| "eval_wer": 0.35093322206352007, | |
| "step": 25800 | |
| }, | |
| { | |
| "epoch": 5.013550135501355, | |
| "eval_loss": 0.49528568983078003, | |
| "eval_runtime": 165.947, | |
| "eval_samples_per_second": 34.083, | |
| "eval_steps_per_second": 4.26, | |
| "eval_wer": 0.35245783248543594, | |
| "step": 25900 | |
| }, | |
| { | |
| "epoch": 5.032907471931862, | |
| "grad_norm": 27.542322158813477, | |
| "learning_rate": 4.087118644067796e-05, | |
| "loss": 0.3158, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 5.032907471931862, | |
| "eval_loss": 0.485315203666687, | |
| "eval_runtime": 165.2844, | |
| "eval_samples_per_second": 34.22, | |
| "eval_steps_per_second": 4.277, | |
| "eval_wer": 0.34703342908956686, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 5.052264808362369, | |
| "eval_loss": 0.5204781293869019, | |
| "eval_runtime": 165.773, | |
| "eval_samples_per_second": 34.119, | |
| "eval_steps_per_second": 4.265, | |
| "eval_wer": 0.36183017444752935, | |
| "step": 26100 | |
| }, | |
| { | |
| "epoch": 5.071622144792877, | |
| "eval_loss": 0.5013459920883179, | |
| "eval_runtime": 165.057, | |
| "eval_samples_per_second": 34.267, | |
| "eval_steps_per_second": 4.283, | |
| "eval_wer": 0.3510455617788191, | |
| "step": 26200 | |
| }, | |
| { | |
| "epoch": 5.090979481223384, | |
| "eval_loss": 0.4863474667072296, | |
| "eval_runtime": 165.6964, | |
| "eval_samples_per_second": 34.135, | |
| "eval_steps_per_second": 4.267, | |
| "eval_wer": 0.3396511049413426, | |
| "step": 26300 | |
| }, | |
| { | |
| "epoch": 5.110336817653891, | |
| "eval_loss": 0.47152572870254517, | |
| "eval_runtime": 166.0563, | |
| "eval_samples_per_second": 34.061, | |
| "eval_steps_per_second": 4.258, | |
| "eval_wer": 0.32851342459597827, | |
| "step": 26400 | |
| }, | |
| { | |
| "epoch": 5.129694154084398, | |
| "grad_norm": 0.8464019894599915, | |
| "learning_rate": 3.5786440677966095e-05, | |
| "loss": 0.2993, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 5.129694154084398, | |
| "eval_loss": 0.4816218912601471, | |
| "eval_runtime": 165.4176, | |
| "eval_samples_per_second": 34.192, | |
| "eval_steps_per_second": 4.274, | |
| "eval_wer": 0.33273418818507167, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 5.149051490514905, | |
| "eval_loss": 0.48058804869651794, | |
| "eval_runtime": 166.4075, | |
| "eval_samples_per_second": 33.989, | |
| "eval_steps_per_second": 4.249, | |
| "eval_wer": 0.33811044598866974, | |
| "step": 26600 | |
| }, | |
| { | |
| "epoch": 5.168408826945412, | |
| "eval_loss": 0.4854019284248352, | |
| "eval_runtime": 165.1934, | |
| "eval_samples_per_second": 34.239, | |
| "eval_steps_per_second": 4.28, | |
| "eval_wer": 0.33416250742244547, | |
| "step": 26700 | |
| }, | |
| { | |
| "epoch": 5.18776616337592, | |
| "eval_loss": 0.49545472860336304, | |
| "eval_runtime": 165.6735, | |
| "eval_samples_per_second": 34.139, | |
| "eval_steps_per_second": 4.267, | |
| "eval_wer": 0.3433422670154547, | |
| "step": 26800 | |
| }, | |
| { | |
| "epoch": 5.207123499806427, | |
| "eval_loss": 0.4862872064113617, | |
| "eval_runtime": 165.5277, | |
| "eval_samples_per_second": 34.17, | |
| "eval_steps_per_second": 4.271, | |
| "eval_wer": 0.34337436407696875, | |
| "step": 26900 | |
| }, | |
| { | |
| "epoch": 5.2264808362369335, | |
| "grad_norm": 10.611580848693848, | |
| "learning_rate": 3.0701694915254236e-05, | |
| "loss": 0.2902, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 5.2264808362369335, | |
| "eval_loss": 0.48670876026153564, | |
| "eval_runtime": 165.4986, | |
| "eval_samples_per_second": 34.176, | |
| "eval_steps_per_second": 4.272, | |
| "eval_wer": 0.3448508289066136, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 5.245838172667441, | |
| "eval_loss": 0.4787338674068451, | |
| "eval_runtime": 165.461, | |
| "eval_samples_per_second": 34.183, | |
| "eval_steps_per_second": 4.273, | |
| "eval_wer": 0.33778947537352955, | |
| "step": 27100 | |
| }, | |
| { | |
| "epoch": 5.265195509097948, | |
| "eval_loss": 0.4861724376678467, | |
| "eval_runtime": 165.3459, | |
| "eval_samples_per_second": 34.207, | |
| "eval_steps_per_second": 4.276, | |
| "eval_wer": 0.33793391215034263, | |
| "step": 27200 | |
| }, | |
| { | |
| "epoch": 5.284552845528455, | |
| "eval_loss": 0.4954308271408081, | |
| "eval_runtime": 165.6637, | |
| "eval_samples_per_second": 34.141, | |
| "eval_steps_per_second": 4.268, | |
| "eval_wer": 0.3467927011282117, | |
| "step": 27300 | |
| }, | |
| { | |
| "epoch": 5.303910181958963, | |
| "eval_loss": 0.572640061378479, | |
| "eval_runtime": 165.6804, | |
| "eval_samples_per_second": 34.138, | |
| "eval_steps_per_second": 4.267, | |
| "eval_wer": 0.41416443324613633, | |
| "step": 27400 | |
| }, | |
| { | |
| "epoch": 5.3232675183894695, | |
| "grad_norm": 1.2211335897445679, | |
| "learning_rate": 2.5627118644067793e-05, | |
| "loss": 0.305, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 5.3232675183894695, | |
| "eval_loss": 0.5179979205131531, | |
| "eval_runtime": 165.6208, | |
| "eval_samples_per_second": 34.15, | |
| "eval_steps_per_second": 4.269, | |
| "eval_wer": 0.35735263436632375, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 5.342624854819976, | |
| "eval_loss": 0.4996646046638489, | |
| "eval_runtime": 164.9613, | |
| "eval_samples_per_second": 34.287, | |
| "eval_steps_per_second": 4.286, | |
| "eval_wer": 0.3452038965832678, | |
| "step": 27600 | |
| }, | |
| { | |
| "epoch": 5.361982191250484, | |
| "eval_loss": 0.4949517846107483, | |
| "eval_runtime": 165.643, | |
| "eval_samples_per_second": 34.146, | |
| "eval_steps_per_second": 4.268, | |
| "eval_wer": 0.34130410360931457, | |
| "step": 27700 | |
| }, | |
| { | |
| "epoch": 5.381339527680991, | |
| "eval_loss": 0.5071349143981934, | |
| "eval_runtime": 166.3001, | |
| "eval_samples_per_second": 34.011, | |
| "eval_steps_per_second": 4.251, | |
| "eval_wer": 0.3491999807417631, | |
| "step": 27800 | |
| }, | |
| { | |
| "epoch": 5.400696864111498, | |
| "eval_loss": 0.5095939040184021, | |
| "eval_runtime": 165.4785, | |
| "eval_samples_per_second": 34.18, | |
| "eval_steps_per_second": 4.272, | |
| "eval_wer": 0.3544799473608191, | |
| "step": 27900 | |
| }, | |
| { | |
| "epoch": 5.420054200542006, | |
| "grad_norm": 15.024033546447754, | |
| "learning_rate": 2.054237288135593e-05, | |
| "loss": 0.3163, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 5.420054200542006, | |
| "eval_loss": 0.5129156112670898, | |
| "eval_runtime": 166.1505, | |
| "eval_samples_per_second": 34.041, | |
| "eval_steps_per_second": 4.255, | |
| "eval_wer": 0.3565983534207443, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 5.4394115369725125, | |
| "eval_loss": 0.5067318677902222, | |
| "eval_runtime": 165.9899, | |
| "eval_samples_per_second": 34.074, | |
| "eval_steps_per_second": 4.259, | |
| "eval_wer": 0.3506122514483799, | |
| "step": 28100 | |
| }, | |
| { | |
| "epoch": 5.45876887340302, | |
| "eval_loss": 0.5053198337554932, | |
| "eval_runtime": 165.3351, | |
| "eval_samples_per_second": 34.209, | |
| "eval_steps_per_second": 4.276, | |
| "eval_wer": 0.35000240727961357, | |
| "step": 28200 | |
| }, | |
| { | |
| "epoch": 5.478126209833527, | |
| "eval_loss": 0.5077947974205017, | |
| "eval_runtime": 165.3012, | |
| "eval_samples_per_second": 34.216, | |
| "eval_steps_per_second": 4.277, | |
| "eval_wer": 0.3518640368474266, | |
| "step": 28300 | |
| }, | |
| { | |
| "epoch": 5.497483546264034, | |
| "eval_loss": 0.48453789949417114, | |
| "eval_runtime": 165.2767, | |
| "eval_samples_per_second": 34.221, | |
| "eval_steps_per_second": 4.278, | |
| "eval_wer": 0.3375166503506604, | |
| "step": 28400 | |
| }, | |
| { | |
| "epoch": 5.516840882694542, | |
| "grad_norm": 0.43120303750038147, | |
| "learning_rate": 1.5457627118644067e-05, | |
| "loss": 0.3136, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 5.516840882694542, | |
| "eval_loss": 0.4930485486984253, | |
| "eval_runtime": 165.9777, | |
| "eval_samples_per_second": 34.077, | |
| "eval_steps_per_second": 4.26, | |
| "eval_wer": 0.3439842082457351, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 5.5361982191250485, | |
| "eval_loss": 0.5025920271873474, | |
| "eval_runtime": 165.742, | |
| "eval_samples_per_second": 34.125, | |
| "eval_steps_per_second": 4.266, | |
| "eval_wer": 0.35122209561714623, | |
| "step": 28600 | |
| }, | |
| { | |
| "epoch": 5.555555555555555, | |
| "eval_loss": 0.5056036710739136, | |
| "eval_runtime": 165.717, | |
| "eval_samples_per_second": 34.13, | |
| "eval_steps_per_second": 4.266, | |
| "eval_wer": 0.3518800853781836, | |
| "step": 28700 | |
| }, | |
| { | |
| "epoch": 5.574912891986063, | |
| "eval_loss": 0.5090658068656921, | |
| "eval_runtime": 165.5604, | |
| "eval_samples_per_second": 34.163, | |
| "eval_steps_per_second": 4.27, | |
| "eval_wer": 0.3546404326683892, | |
| "step": 28800 | |
| }, | |
| { | |
| "epoch": 5.59427022841657, | |
| "eval_loss": 0.5027741193771362, | |
| "eval_runtime": 170.2845, | |
| "eval_samples_per_second": 33.215, | |
| "eval_steps_per_second": 4.152, | |
| "eval_wer": 0.34952095135690325, | |
| "step": 28900 | |
| }, | |
| { | |
| "epoch": 5.613627564847077, | |
| "grad_norm": 1.859834909439087, | |
| "learning_rate": 1.0372881355932203e-05, | |
| "loss": 0.3092, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 5.613627564847077, | |
| "eval_loss": 0.505651593208313, | |
| "eval_runtime": 164.9869, | |
| "eval_samples_per_second": 34.282, | |
| "eval_steps_per_second": 4.285, | |
| "eval_wer": 0.3509974161865481, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 5.6329849012775846, | |
| "eval_loss": 0.5085631608963013, | |
| "eval_runtime": 165.6325, | |
| "eval_samples_per_second": 34.148, | |
| "eval_steps_per_second": 4.268, | |
| "eval_wer": 0.3532923560848004, | |
| "step": 29100 | |
| }, | |
| { | |
| "epoch": 5.652342237708091, | |
| "eval_loss": 0.5055486559867859, | |
| "eval_runtime": 165.6348, | |
| "eval_samples_per_second": 34.147, | |
| "eval_steps_per_second": 4.268, | |
| "eval_wer": 0.35144677504774435, | |
| "step": 29200 | |
| }, | |
| { | |
| "epoch": 5.671699574138598, | |
| "eval_loss": 0.5133376717567444, | |
| "eval_runtime": 166.8503, | |
| "eval_samples_per_second": 33.899, | |
| "eval_steps_per_second": 4.237, | |
| "eval_wer": 0.35765755645070696, | |
| "step": 29300 | |
| }, | |
| { | |
| "epoch": 5.691056910569106, | |
| "eval_loss": 0.5129527449607849, | |
| "eval_runtime": 165.063, | |
| "eval_samples_per_second": 34.266, | |
| "eval_steps_per_second": 4.283, | |
| "eval_wer": 0.35703166375118356, | |
| "step": 29400 | |
| }, | |
| { | |
| "epoch": 5.710414246999613, | |
| "grad_norm": 1.5260862112045288, | |
| "learning_rate": 5.288135593220339e-06, | |
| "loss": 0.3152, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 5.710414246999613, | |
| "eval_loss": 0.5147610902786255, | |
| "eval_runtime": 165.1582, | |
| "eval_samples_per_second": 34.246, | |
| "eval_steps_per_second": 4.281, | |
| "eval_wer": 0.3581390123734172, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 5.72977158343012, | |
| "eval_loss": 0.5114809274673462, | |
| "eval_runtime": 165.7617, | |
| "eval_samples_per_second": 34.121, | |
| "eval_steps_per_second": 4.265, | |
| "eval_wer": 0.3554589077369967, | |
| "step": 29600 | |
| }, | |
| { | |
| "epoch": 5.7491289198606275, | |
| "eval_loss": 0.5053985714912415, | |
| "eval_runtime": 165.7121, | |
| "eval_samples_per_second": 34.131, | |
| "eval_steps_per_second": 4.266, | |
| "eval_wer": 0.35263436632376305, | |
| "step": 29700 | |
| }, | |
| { | |
| "epoch": 5.768486256291134, | |
| "eval_loss": 0.5080947279930115, | |
| "eval_runtime": 165.6502, | |
| "eval_samples_per_second": 34.144, | |
| "eval_steps_per_second": 4.268, | |
| "eval_wer": 0.3535651811076696, | |
| "step": 29800 | |
| }, | |
| { | |
| "epoch": 5.787843592721641, | |
| "eval_loss": 0.5076740384101868, | |
| "eval_runtime": 164.8589, | |
| "eval_samples_per_second": 34.308, | |
| "eval_steps_per_second": 4.289, | |
| "eval_wer": 0.35348493845388457, | |
| "step": 29900 | |
| }, | |
| { | |
| "epoch": 5.807200929152149, | |
| "grad_norm": 24.957311630249023, | |
| "learning_rate": 2.0338983050847458e-07, | |
| "loss": 0.3085, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 5.807200929152149, | |
| "eval_loss": 0.5066753029823303, | |
| "eval_runtime": 165.8811, | |
| "eval_samples_per_second": 34.097, | |
| "eval_steps_per_second": 4.262, | |
| "eval_wer": 0.35224920158559486, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 5.807200929152149, | |
| "step": 30000, | |
| "total_flos": 3.3745707679449666e+19, | |
| "train_loss": 0.49102539647420246, | |
| "train_runtime": 61359.0363, | |
| "train_samples_per_second": 3.911, | |
| "train_steps_per_second": 0.489 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 30000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 6, | |
| "save_steps": 400, | |
| "total_flos": 3.3745707679449666e+19, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |