| { | |
| "best_global_step": 286140, | |
| "best_metric": 0.1884266993162269, | |
| "best_model_checkpoint": "wav2vec2-asr-africa-base-fintuned-luganda-400hrs-v0.1/checkpoint-286140", | |
| "epoch": 99.0, | |
| "eval_steps": 500, | |
| "global_step": 298188, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 7.066422462463379, | |
| "learning_rate": 6.98140770252324e-05, | |
| "loss": 3.2575, | |
| "step": 3012 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_cer": 0.10930492502565166, | |
| "eval_loss": 0.37255266308784485, | |
| "eval_runtime": 151.5408, | |
| "eval_samples_per_second": 239.869, | |
| "eval_steps_per_second": 7.496, | |
| "eval_wer": 0.534090083733671, | |
| "step": 3012 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 15.221166610717773, | |
| "learning_rate": 6.929480730277542e-05, | |
| "loss": 0.8396, | |
| "step": 6024 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_cer": 0.07810481279107405, | |
| "eval_loss": 0.2538328468799591, | |
| "eval_runtime": 152.0166, | |
| "eval_samples_per_second": 239.119, | |
| "eval_steps_per_second": 7.473, | |
| "eval_wer": 0.3983557626127949, | |
| "step": 6024 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 7.060754299163818, | |
| "learning_rate": 6.858773659570472e-05, | |
| "loss": 0.7487, | |
| "step": 9036 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_cer": 0.0682448584368034, | |
| "eval_loss": 0.22769133746623993, | |
| "eval_runtime": 153.0969, | |
| "eval_samples_per_second": 237.431, | |
| "eval_steps_per_second": 7.42, | |
| "eval_wer": 0.35308562554689743, | |
| "step": 9036 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 9.472733497619629, | |
| "learning_rate": 6.788090063986478e-05, | |
| "loss": 0.7226, | |
| "step": 12048 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_cer": 0.06425062788093966, | |
| "eval_loss": 0.21326717734336853, | |
| "eval_runtime": 171.3308, | |
| "eval_samples_per_second": 212.163, | |
| "eval_steps_per_second": 6.63, | |
| "eval_wer": 0.33642189872410777, | |
| "step": 12048 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 23.48148536682129, | |
| "learning_rate": 6.717429943525561e-05, | |
| "loss": 0.7096, | |
| "step": 15060 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_cer": 0.06122744648913919, | |
| "eval_loss": 0.20842401683330536, | |
| "eval_runtime": 174.4012, | |
| "eval_samples_per_second": 208.427, | |
| "eval_steps_per_second": 6.514, | |
| "eval_wer": 0.3211001241651708, | |
| "step": 15060 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 7.9525041580200195, | |
| "learning_rate": 6.646746347941567e-05, | |
| "loss": 0.6979, | |
| "step": 18072 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_cer": 0.06395399482657282, | |
| "eval_loss": 0.20914477109909058, | |
| "eval_runtime": 153.4786, | |
| "eval_samples_per_second": 236.841, | |
| "eval_steps_per_second": 7.402, | |
| "eval_wer": 0.3287402596055075, | |
| "step": 18072 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 43.115169525146484, | |
| "learning_rate": 6.57608622748065e-05, | |
| "loss": 0.6899, | |
| "step": 21084 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_cer": 0.06076026089037598, | |
| "eval_loss": 0.20185638964176178, | |
| "eval_runtime": 173.3256, | |
| "eval_samples_per_second": 209.721, | |
| "eval_steps_per_second": 6.554, | |
| "eval_wer": 0.31623727652863237, | |
| "step": 21084 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 4.6848602294921875, | |
| "learning_rate": 6.505426107019733e-05, | |
| "loss": 0.6765, | |
| "step": 24096 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_cer": 0.060137193933831115, | |
| "eval_loss": 0.19728189706802368, | |
| "eval_runtime": 154.1162, | |
| "eval_samples_per_second": 235.861, | |
| "eval_steps_per_second": 7.371, | |
| "eval_wer": 0.31059969356450884, | |
| "step": 24096 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 5.830124378204346, | |
| "learning_rate": 6.434719036312661e-05, | |
| "loss": 0.6701, | |
| "step": 27108 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_cer": 0.05820976679231927, | |
| "eval_loss": 0.19281432032585144, | |
| "eval_runtime": 156.5256, | |
| "eval_samples_per_second": 232.23, | |
| "eval_steps_per_second": 7.258, | |
| "eval_wer": 0.30471654717065966, | |
| "step": 27108 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": Infinity, | |
| "learning_rate": 6.364058915851744e-05, | |
| "loss": 0.6621, | |
| "step": 30120 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_cer": 0.058245986284274416, | |
| "eval_loss": 0.19237777590751648, | |
| "eval_runtime": 155.0055, | |
| "eval_samples_per_second": 234.508, | |
| "eval_steps_per_second": 7.329, | |
| "eval_wer": 0.3038691804061135, | |
| "step": 30120 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 5.588762283325195, | |
| "learning_rate": 6.29337532026775e-05, | |
| "loss": 0.6554, | |
| "step": 33132 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_cer": 0.05662298626501848, | |
| "eval_loss": 0.18665704131126404, | |
| "eval_runtime": 153.969, | |
| "eval_samples_per_second": 236.087, | |
| "eval_steps_per_second": 7.378, | |
| "eval_wer": 0.2982627252006876, | |
| "step": 33132 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 5.955714702606201, | |
| "learning_rate": 6.222691724683756e-05, | |
| "loss": 0.6475, | |
| "step": 36144 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_cer": 0.05515495065438077, | |
| "eval_loss": 0.1829417496919632, | |
| "eval_runtime": 155.1287, | |
| "eval_samples_per_second": 234.322, | |
| "eval_steps_per_second": 7.323, | |
| "eval_wer": 0.2873610923768119, | |
| "step": 36144 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 31.985713958740234, | |
| "learning_rate": 6.152008129099762e-05, | |
| "loss": 0.6429, | |
| "step": 39156 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_cer": 0.05419032013446143, | |
| "eval_loss": 0.1801947057247162, | |
| "eval_runtime": 168.8898, | |
| "eval_samples_per_second": 215.229, | |
| "eval_steps_per_second": 6.726, | |
| "eval_wer": 0.28527553254407173, | |
| "step": 39156 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 6.557770252227783, | |
| "learning_rate": 6.081324533515768e-05, | |
| "loss": 0.6351, | |
| "step": 42168 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_cer": 0.055327337097104, | |
| "eval_loss": 0.18261073529720306, | |
| "eval_runtime": 159.2183, | |
| "eval_samples_per_second": 228.303, | |
| "eval_steps_per_second": 7.135, | |
| "eval_wer": 0.2872746263804296, | |
| "step": 42168 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 8.820505142211914, | |
| "learning_rate": 6.010640937931774e-05, | |
| "loss": 0.6319, | |
| "step": 45180 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_cer": 0.05439250742499585, | |
| "eval_loss": 0.17926117777824402, | |
| "eval_runtime": 152.0308, | |
| "eval_samples_per_second": 239.096, | |
| "eval_steps_per_second": 7.472, | |
| "eval_wer": 0.28315884495263394, | |
| "step": 45180 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 8.058792114257812, | |
| "learning_rate": 5.93995734234778e-05, | |
| "loss": 0.6251, | |
| "step": 48192 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_cer": 0.054798715904391546, | |
| "eval_loss": 0.1785019189119339, | |
| "eval_runtime": 154.575, | |
| "eval_samples_per_second": 235.161, | |
| "eval_steps_per_second": 7.349, | |
| "eval_wer": 0.283826362444705, | |
| "step": 48192 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 7.038857936859131, | |
| "learning_rate": 5.86925027164071e-05, | |
| "loss": 0.6172, | |
| "step": 51204 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_cer": 0.051710431121988164, | |
| "eval_loss": 0.17091116309165955, | |
| "eval_runtime": 154.6152, | |
| "eval_samples_per_second": 235.1, | |
| "eval_steps_per_second": 7.347, | |
| "eval_wer": 0.27192518270265037, | |
| "step": 51204 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.7985901511797926e-05, | |
| "loss": 0.6122, | |
| "step": 54216 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_cer": 0.05208454638066411, | |
| "eval_loss": 0.1720370054244995, | |
| "eval_runtime": 154.3969, | |
| "eval_samples_per_second": 235.432, | |
| "eval_steps_per_second": 7.358, | |
| "eval_wer": 0.27160698783596365, | |
| "step": 54216 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 4.123114109039307, | |
| "learning_rate": 5.727930030718875e-05, | |
| "loss": 0.6068, | |
| "step": 57228 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_cer": 0.0505266497520111, | |
| "eval_loss": 0.16939722001552582, | |
| "eval_runtime": 154.2835, | |
| "eval_samples_per_second": 235.605, | |
| "eval_steps_per_second": 7.363, | |
| "eval_wer": 0.26646744901100194, | |
| "step": 57228 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 11.014168739318848, | |
| "learning_rate": 5.657222960011804e-05, | |
| "loss": 0.6035, | |
| "step": 60240 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_cer": 0.049698644657441546, | |
| "eval_loss": 0.1669510304927826, | |
| "eval_runtime": 155.9471, | |
| "eval_samples_per_second": 233.092, | |
| "eval_steps_per_second": 7.285, | |
| "eval_wer": 0.26278053892526226, | |
| "step": 60240 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "grad_norm": 7.567544937133789, | |
| "learning_rate": 5.5865158893047335e-05, | |
| "loss": 0.5957, | |
| "step": 63252 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_cer": 0.050415698903237105, | |
| "eval_loss": 0.1704263538122177, | |
| "eval_runtime": 155.0829, | |
| "eval_samples_per_second": 234.391, | |
| "eval_steps_per_second": 7.325, | |
| "eval_wer": 0.2643818891782618, | |
| "step": 63252 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "grad_norm": 3.353114366531372, | |
| "learning_rate": 5.5158557688438164e-05, | |
| "loss": 0.5909, | |
| "step": 66264 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_cer": 0.049318569229203364, | |
| "eval_loss": 0.16528591513633728, | |
| "eval_runtime": 155.149, | |
| "eval_samples_per_second": 234.291, | |
| "eval_steps_per_second": 7.322, | |
| "eval_wer": 0.25990640920551583, | |
| "step": 66264 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "grad_norm": 5.273142337799072, | |
| "learning_rate": 5.445172173259822e-05, | |
| "loss": 0.5879, | |
| "step": 69276 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_cer": 0.048735389561267335, | |
| "eval_loss": 0.16745983064174652, | |
| "eval_runtime": 155.8132, | |
| "eval_samples_per_second": 233.292, | |
| "eval_steps_per_second": 7.291, | |
| "eval_wer": 0.2573400984328903, | |
| "step": 69276 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "grad_norm": 12.098519325256348, | |
| "learning_rate": 5.374512052798905e-05, | |
| "loss": 0.5966, | |
| "step": 72288 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_cer": 0.05103463958854657, | |
| "eval_loss": 0.19431033730506897, | |
| "eval_runtime": 154.0761, | |
| "eval_samples_per_second": 235.922, | |
| "eval_steps_per_second": 7.373, | |
| "eval_wer": 0.2738551037419025, | |
| "step": 72288 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "grad_norm": 12.023294448852539, | |
| "learning_rate": 5.3038519323379875e-05, | |
| "loss": 0.6444, | |
| "step": 75300 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_cer": 0.05154996501838942, | |
| "eval_loss": 0.1868334412574768, | |
| "eval_runtime": 152.5369, | |
| "eval_samples_per_second": 238.303, | |
| "eval_steps_per_second": 7.447, | |
| "eval_wer": 0.27229871580702175, | |
| "step": 75300 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "grad_norm": 9.066435813903809, | |
| "learning_rate": 5.2331448616309165e-05, | |
| "loss": 0.5999, | |
| "step": 78312 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_cer": 0.04910904634536157, | |
| "eval_loss": 0.16771361231803894, | |
| "eval_runtime": 154.4851, | |
| "eval_samples_per_second": 235.298, | |
| "eval_steps_per_second": 7.353, | |
| "eval_wer": 0.25782430801263095, | |
| "step": 78312 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "grad_norm": 9.274683952331543, | |
| "learning_rate": 5.1624847411699994e-05, | |
| "loss": 0.5911, | |
| "step": 81324 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_cer": 0.04738747429103783, | |
| "eval_loss": 0.16794191300868988, | |
| "eval_runtime": 154.6471, | |
| "eval_samples_per_second": 235.051, | |
| "eval_steps_per_second": 7.346, | |
| "eval_wer": 0.25102462205712983, | |
| "step": 81324 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "grad_norm": 5.650504112243652, | |
| "learning_rate": 5.091777670462929e-05, | |
| "loss": 0.586, | |
| "step": 84336 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_cer": 0.04840666328618075, | |
| "eval_loss": 0.1722731739282608, | |
| "eval_runtime": 153.0438, | |
| "eval_samples_per_second": 237.514, | |
| "eval_steps_per_second": 7.423, | |
| "eval_wer": 0.25386416537832335, | |
| "step": 84336 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "grad_norm": 18.539613723754883, | |
| "learning_rate": 5.021070599755859e-05, | |
| "loss": 0.5816, | |
| "step": 87348 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_cer": 0.04769969548118283, | |
| "eval_loss": 0.16775010526180267, | |
| "eval_runtime": 156.0962, | |
| "eval_samples_per_second": 232.869, | |
| "eval_steps_per_second": 7.278, | |
| "eval_wer": 0.25264326550940575, | |
| "step": 87348 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 4.343358039855957, | |
| "learning_rate": 4.950457429541094e-05, | |
| "loss": 0.5886, | |
| "step": 90360 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_cer": 0.04993246669411401, | |
| "eval_loss": 0.18236766755580902, | |
| "eval_runtime": 151.7294, | |
| "eval_samples_per_second": 239.571, | |
| "eval_steps_per_second": 7.487, | |
| "eval_wer": 0.2629396363586056, | |
| "step": 90360 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "grad_norm": 13.675621032714844, | |
| "learning_rate": 4.879773833957101e-05, | |
| "loss": 0.5978, | |
| "step": 93372 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_cer": 0.04701886072734242, | |
| "eval_loss": 0.16201142966747284, | |
| "eval_runtime": 152.4808, | |
| "eval_samples_per_second": 238.391, | |
| "eval_steps_per_second": 7.45, | |
| "eval_wer": 0.24908778373816712, | |
| "step": 93372 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "grad_norm": 5.842775821685791, | |
| "learning_rate": 4.809066763250029e-05, | |
| "loss": 0.5722, | |
| "step": 96384 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_cer": 0.04652920987407537, | |
| "eval_loss": 0.15837915241718292, | |
| "eval_runtime": 153.9577, | |
| "eval_samples_per_second": 236.104, | |
| "eval_steps_per_second": 7.379, | |
| "eval_wer": 0.24719590773732322, | |
| "step": 96384 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "grad_norm": 12.179231643676758, | |
| "learning_rate": 4.738359692542959e-05, | |
| "loss": 0.5615, | |
| "step": 99396 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_cer": 0.046122542920097966, | |
| "eval_loss": 0.15639054775238037, | |
| "eval_runtime": 165.4748, | |
| "eval_samples_per_second": 219.671, | |
| "eval_steps_per_second": 6.865, | |
| "eval_wer": 0.2421047898703356, | |
| "step": 99396 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "grad_norm": 11.811565399169922, | |
| "learning_rate": 4.6676760969589654e-05, | |
| "loss": 0.5566, | |
| "step": 102408 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_cer": 0.04475858103950859, | |
| "eval_loss": 0.15303878486156464, | |
| "eval_runtime": 151.973, | |
| "eval_samples_per_second": 239.187, | |
| "eval_steps_per_second": 7.475, | |
| "eval_wer": 0.23680615361203053, | |
| "step": 102408 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "grad_norm": 27.314680099487305, | |
| "learning_rate": 4.5970394516211246e-05, | |
| "loss": 0.5514, | |
| "step": 105420 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_cer": 0.04322590051284967, | |
| "eval_loss": 0.14988180994987488, | |
| "eval_runtime": 152.0916, | |
| "eval_samples_per_second": 239.001, | |
| "eval_steps_per_second": 7.469, | |
| "eval_wer": 0.2308780449000626, | |
| "step": 105420 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "grad_norm": 6.874896049499512, | |
| "learning_rate": 4.526332380914054e-05, | |
| "loss": 0.5485, | |
| "step": 108432 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_cer": 0.043550500516700855, | |
| "eval_loss": 0.15108104050159454, | |
| "eval_runtime": 156.7097, | |
| "eval_samples_per_second": 231.958, | |
| "eval_steps_per_second": 7.249, | |
| "eval_wer": 0.23083308258194382, | |
| "step": 108432 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "grad_norm": 18.988279342651367, | |
| "learning_rate": 4.4556722604531365e-05, | |
| "loss": 0.5451, | |
| "step": 111444 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_cer": 0.04385080136772137, | |
| "eval_loss": 0.15071320533752441, | |
| "eval_runtime": 172.1252, | |
| "eval_samples_per_second": 211.183, | |
| "eval_steps_per_second": 6.6, | |
| "eval_wer": 0.23185338133925454, | |
| "step": 111444 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "grad_norm": 28.172042846679688, | |
| "learning_rate": 4.384988664869143e-05, | |
| "loss": 0.5433, | |
| "step": 114456 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_cer": 0.0433771971248142, | |
| "eval_loss": 0.14824804663658142, | |
| "eval_runtime": 152.6042, | |
| "eval_samples_per_second": 238.198, | |
| "eval_steps_per_second": 7.444, | |
| "eval_wer": 0.23122390888559166, | |
| "step": 114456 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "grad_norm": 4.653916835784912, | |
| "learning_rate": 4.3143050692851484e-05, | |
| "loss": 0.5391, | |
| "step": 117468 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_cer": 0.04353903865215809, | |
| "eval_loss": 0.14684619009494781, | |
| "eval_runtime": 156.6046, | |
| "eval_samples_per_second": 232.113, | |
| "eval_steps_per_second": 7.254, | |
| "eval_wer": 0.22910722129415387, | |
| "step": 117468 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 5.8400702476501465, | |
| "learning_rate": 4.243621473701154e-05, | |
| "loss": 0.5347, | |
| "step": 120480 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_cer": 0.042970530170836796, | |
| "eval_loss": 0.1462726891040802, | |
| "eval_runtime": 178.4468, | |
| "eval_samples_per_second": 203.702, | |
| "eval_steps_per_second": 6.366, | |
| "eval_wer": 0.22744707416361443, | |
| "step": 120480 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "grad_norm": 16.060422897338867, | |
| "learning_rate": 4.172914402994084e-05, | |
| "loss": 0.5313, | |
| "step": 123492 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_cer": 0.04216269795786252, | |
| "eval_loss": 0.14503081142902374, | |
| "eval_runtime": 152.5692, | |
| "eval_samples_per_second": 238.253, | |
| "eval_steps_per_second": 7.446, | |
| "eval_wer": 0.22400226886774507, | |
| "step": 123492 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "grad_norm": 8.755998611450195, | |
| "learning_rate": 4.102254282533167e-05, | |
| "loss": 0.5291, | |
| "step": 126504 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_cer": 0.04194263015864137, | |
| "eval_loss": 0.1446152627468109, | |
| "eval_runtime": 152.4116, | |
| "eval_samples_per_second": 238.499, | |
| "eval_steps_per_second": 7.454, | |
| "eval_wer": 0.224061065745285, | |
| "step": 126504 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "grad_norm": 9.50841999053955, | |
| "learning_rate": 4.031570686949173e-05, | |
| "loss": 0.5269, | |
| "step": 129516 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_cer": 0.04270048864220919, | |
| "eval_loss": 0.14530107378959656, | |
| "eval_runtime": 153.731, | |
| "eval_samples_per_second": 236.452, | |
| "eval_steps_per_second": 7.39, | |
| "eval_wer": 0.22547219080624353, | |
| "step": 129516 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "grad_norm": 28.027828216552734, | |
| "learning_rate": 3.960887091365178e-05, | |
| "loss": 0.5253, | |
| "step": 132528 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_cer": 0.042549192030244654, | |
| "eval_loss": 0.14459766447544098, | |
| "eval_runtime": 179.3972, | |
| "eval_samples_per_second": 202.623, | |
| "eval_steps_per_second": 6.332, | |
| "eval_wer": 0.22531309337290018, | |
| "step": 132528 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "grad_norm": 9.71485710144043, | |
| "learning_rate": 3.890226970904261e-05, | |
| "loss": 0.523, | |
| "step": 135540 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_cer": 0.041189356420890666, | |
| "eval_loss": 0.1429988592863083, | |
| "eval_runtime": 151.6718, | |
| "eval_samples_per_second": 239.662, | |
| "eval_steps_per_second": 7.49, | |
| "eval_wer": 0.22018738910735963, | |
| "step": 135540 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "grad_norm": 13.113300323486328, | |
| "learning_rate": 3.819519900197191e-05, | |
| "loss": 0.5192, | |
| "step": 138552 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_cer": 0.040866590315366325, | |
| "eval_loss": 0.14137160778045654, | |
| "eval_runtime": 152.9576, | |
| "eval_samples_per_second": 237.648, | |
| "eval_steps_per_second": 7.427, | |
| "eval_wer": 0.21718528971296747, | |
| "step": 138552 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "grad_norm": 15.17225456237793, | |
| "learning_rate": 3.7488597797362736e-05, | |
| "loss": 0.518, | |
| "step": 141564 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_cer": 0.040502561497488015, | |
| "eval_loss": 0.14037571847438812, | |
| "eval_runtime": 151.9659, | |
| "eval_samples_per_second": 239.198, | |
| "eval_steps_per_second": 7.475, | |
| "eval_wer": 0.21598168304332638, | |
| "step": 141564 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "grad_norm": 5.159524917602539, | |
| "learning_rate": 3.678152709029203e-05, | |
| "loss": 0.5139, | |
| "step": 144576 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_cer": 0.04006930301777139, | |
| "eval_loss": 0.13999390602111816, | |
| "eval_runtime": 161.9319, | |
| "eval_samples_per_second": 224.477, | |
| "eval_steps_per_second": 7.015, | |
| "eval_wer": 0.2143319118323528, | |
| "step": 144576 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "grad_norm": 5.24137020111084, | |
| "learning_rate": 3.6074925885682855e-05, | |
| "loss": 0.5133, | |
| "step": 147588 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_cer": 0.04118523014965527, | |
| "eval_loss": 0.14138683676719666, | |
| "eval_runtime": 154.3411, | |
| "eval_samples_per_second": 235.517, | |
| "eval_steps_per_second": 7.36, | |
| "eval_wer": 0.21796694232026315, | |
| "step": 147588 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "grad_norm": 4.531148910522461, | |
| "learning_rate": 3.5367855178612144e-05, | |
| "loss": 0.5114, | |
| "step": 150600 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_cer": 0.04037235471628217, | |
| "eval_loss": 0.14019279181957245, | |
| "eval_runtime": 152.1041, | |
| "eval_samples_per_second": 238.981, | |
| "eval_steps_per_second": 7.469, | |
| "eval_wer": 0.21485070781064639, | |
| "step": 150600 | |
| }, | |
| { | |
| "epoch": 51.0, | |
| "grad_norm": 6.761490821838379, | |
| "learning_rate": 3.4661019222772204e-05, | |
| "loss": 0.5087, | |
| "step": 153612 | |
| }, | |
| { | |
| "epoch": 51.0, | |
| "eval_cer": 0.04063826997367439, | |
| "eval_loss": 0.14041763544082642, | |
| "eval_runtime": 154.2132, | |
| "eval_samples_per_second": 235.713, | |
| "eval_steps_per_second": 7.366, | |
| "eval_wer": 0.21654889997959403, | |
| "step": 153612 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "grad_norm": 4.379857540130615, | |
| "learning_rate": 3.395441801816303e-05, | |
| "loss": 0.5066, | |
| "step": 156624 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_cer": 0.04042691319150575, | |
| "eval_loss": 0.13891662657260895, | |
| "eval_runtime": 159.6511, | |
| "eval_samples_per_second": 227.684, | |
| "eval_steps_per_second": 7.116, | |
| "eval_wer": 0.215715367774469, | |
| "step": 156624 | |
| }, | |
| { | |
| "epoch": 53.0, | |
| "grad_norm": 19.332077026367188, | |
| "learning_rate": 3.324734731109233e-05, | |
| "loss": 0.5037, | |
| "step": 159636 | |
| }, | |
| { | |
| "epoch": 53.0, | |
| "eval_cer": 0.03977129453965943, | |
| "eval_loss": 0.1375364065170288, | |
| "eval_runtime": 164.6123, | |
| "eval_samples_per_second": 220.822, | |
| "eval_steps_per_second": 6.901, | |
| "eval_wer": 0.2132078538793834, | |
| "step": 159636 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "grad_norm": 11.953184127807617, | |
| "learning_rate": 3.254098085771392e-05, | |
| "loss": 0.5024, | |
| "step": 162648 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "eval_cer": 0.039842816574406296, | |
| "eval_loss": 0.13721118867397308, | |
| "eval_runtime": 156.3278, | |
| "eval_samples_per_second": 232.524, | |
| "eval_steps_per_second": 7.267, | |
| "eval_wer": 0.21213221688438805, | |
| "step": 162648 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "grad_norm": 7.574815273284912, | |
| "learning_rate": 3.183391015064322e-05, | |
| "loss": 0.5, | |
| "step": 165660 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "eval_cer": 0.04010139623849114, | |
| "eval_loss": 0.13785392045974731, | |
| "eval_runtime": 152.1653, | |
| "eval_samples_per_second": 238.885, | |
| "eval_steps_per_second": 7.466, | |
| "eval_wer": 0.2131732674808305, | |
| "step": 165660 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "grad_norm": 11.841280937194824, | |
| "learning_rate": 3.112707419480328e-05, | |
| "loss": 0.4976, | |
| "step": 168672 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_cer": 0.038647114865304755, | |
| "eval_loss": 0.13485907018184662, | |
| "eval_runtime": 153.6401, | |
| "eval_samples_per_second": 236.592, | |
| "eval_steps_per_second": 7.394, | |
| "eval_wer": 0.20721403101016495, | |
| "step": 168672 | |
| }, | |
| { | |
| "epoch": 57.0, | |
| "grad_norm": 14.78765869140625, | |
| "learning_rate": 3.0420238238963334e-05, | |
| "loss": 0.4948, | |
| "step": 171684 | |
| }, | |
| { | |
| "epoch": 57.0, | |
| "eval_cer": 0.0392779758897387, | |
| "eval_loss": 0.13624149560928345, | |
| "eval_runtime": 152.0309, | |
| "eval_samples_per_second": 239.096, | |
| "eval_steps_per_second": 7.472, | |
| "eval_wer": 0.2102472581632547, | |
| "step": 171684 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "grad_norm": 7.023338794708252, | |
| "learning_rate": 2.9713402283123397e-05, | |
| "loss": 0.4933, | |
| "step": 174696 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "eval_cer": 0.038942372495926456, | |
| "eval_loss": 0.13551433384418488, | |
| "eval_runtime": 151.9877, | |
| "eval_samples_per_second": 239.164, | |
| "eval_steps_per_second": 7.474, | |
| "eval_wer": 0.20676094918912188, | |
| "step": 174696 | |
| }, | |
| { | |
| "epoch": 59.0, | |
| "grad_norm": 6.568221092224121, | |
| "learning_rate": 2.9006801078514222e-05, | |
| "loss": 0.4924, | |
| "step": 177708 | |
| }, | |
| { | |
| "epoch": 59.0, | |
| "eval_cer": 0.03848756571086942, | |
| "eval_loss": 0.13611619174480438, | |
| "eval_runtime": 177.5315, | |
| "eval_samples_per_second": 204.752, | |
| "eval_steps_per_second": 6.399, | |
| "eval_wer": 0.20549508700208555, | |
| "step": 177708 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "grad_norm": 23.931304931640625, | |
| "learning_rate": 2.8300199873905052e-05, | |
| "loss": 0.4901, | |
| "step": 180720 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_cer": 0.03840274791325294, | |
| "eval_loss": 0.13464532792568207, | |
| "eval_runtime": 153.1916, | |
| "eval_samples_per_second": 237.285, | |
| "eval_steps_per_second": 7.416, | |
| "eval_wer": 0.2053671173274398, | |
| "step": 180720 | |
| }, | |
| { | |
| "epoch": 61.0, | |
| "grad_norm": 18.084495544433594, | |
| "learning_rate": 2.759312916683434e-05, | |
| "loss": 0.4898, | |
| "step": 183732 | |
| }, | |
| { | |
| "epoch": 61.0, | |
| "eval_cer": 0.038370654692533195, | |
| "eval_loss": 0.13341517746448517, | |
| "eval_runtime": 151.6501, | |
| "eval_samples_per_second": 239.696, | |
| "eval_steps_per_second": 7.491, | |
| "eval_wer": 0.2050074187824896, | |
| "step": 183732 | |
| }, | |
| { | |
| "epoch": 62.0, | |
| "grad_norm": 2.890596866607666, | |
| "learning_rate": 2.6886293210994404e-05, | |
| "loss": 0.4873, | |
| "step": 186744 | |
| }, | |
| { | |
| "epoch": 62.0, | |
| "eval_cer": 0.038351857234683054, | |
| "eval_loss": 0.1341981142759323, | |
| "eval_runtime": 150.7747, | |
| "eval_samples_per_second": 241.088, | |
| "eval_steps_per_second": 7.534, | |
| "eval_wer": 0.20600696570066857, | |
| "step": 186744 | |
| }, | |
| { | |
| "epoch": 63.0, | |
| "grad_norm": 11.759881973266602, | |
| "learning_rate": 2.617969200638523e-05, | |
| "loss": 0.4865, | |
| "step": 189756 | |
| }, | |
| { | |
| "epoch": 63.0, | |
| "eval_cer": 0.03869296232347583, | |
| "eval_loss": 0.13458400964736938, | |
| "eval_runtime": 152.6729, | |
| "eval_samples_per_second": 238.091, | |
| "eval_steps_per_second": 7.441, | |
| "eval_wer": 0.20699613669928163, | |
| "step": 189756 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "grad_norm": 13.245360374450684, | |
| "learning_rate": 2.547309080177606e-05, | |
| "loss": 0.4842, | |
| "step": 192768 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_cer": 0.03874110215455545, | |
| "eval_loss": 0.13456492125988007, | |
| "eval_runtime": 153.3987, | |
| "eval_samples_per_second": 236.964, | |
| "eval_steps_per_second": 7.406, | |
| "eval_wer": 0.2072278655695861, | |
| "step": 192768 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "grad_norm": 11.684355735778809, | |
| "learning_rate": 2.4766020094705352e-05, | |
| "loss": 0.4822, | |
| "step": 195780 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "eval_cer": 0.03811941062175572, | |
| "eval_loss": 0.13252592086791992, | |
| "eval_runtime": 156.7414, | |
| "eval_samples_per_second": 231.911, | |
| "eval_steps_per_second": 7.248, | |
| "eval_wer": 0.20395599226648128, | |
| "step": 195780 | |
| }, | |
| { | |
| "epoch": 66.0, | |
| "grad_norm": 25.19974708557129, | |
| "learning_rate": 2.405918413886541e-05, | |
| "loss": 0.4814, | |
| "step": 198792 | |
| }, | |
| { | |
| "epoch": 66.0, | |
| "eval_cer": 0.037090135185815165, | |
| "eval_loss": 0.13119570910930634, | |
| "eval_runtime": 154.348, | |
| "eval_samples_per_second": 235.507, | |
| "eval_steps_per_second": 7.36, | |
| "eval_wer": 0.19890983671761242, | |
| "step": 198792 | |
| }, | |
| { | |
| "epoch": 67.0, | |
| "grad_norm": 12.580814361572266, | |
| "learning_rate": 2.335234818302547e-05, | |
| "loss": 0.4796, | |
| "step": 201804 | |
| }, | |
| { | |
| "epoch": 67.0, | |
| "eval_cer": 0.03740739959635898, | |
| "eval_loss": 0.13117973506450653, | |
| "eval_runtime": 162.9804, | |
| "eval_samples_per_second": 223.033, | |
| "eval_steps_per_second": 6.97, | |
| "eval_wer": 0.1999750977930419, | |
| "step": 201804 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "grad_norm": 11.110360145568848, | |
| "learning_rate": 2.2645746978416297e-05, | |
| "loss": 0.4771, | |
| "step": 204816 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_cer": 0.037213006373713636, | |
| "eval_loss": 0.1303921490907669, | |
| "eval_runtime": 152.708, | |
| "eval_samples_per_second": 238.036, | |
| "eval_steps_per_second": 7.439, | |
| "eval_wer": 0.1997191584437504, | |
| "step": 204816 | |
| }, | |
| { | |
| "epoch": 69.0, | |
| "grad_norm": 4.782271385192871, | |
| "learning_rate": 2.193891102257636e-05, | |
| "loss": 0.4756, | |
| "step": 207828 | |
| }, | |
| { | |
| "epoch": 69.0, | |
| "eval_cer": 0.037708617396542916, | |
| "eval_loss": 0.13083402812480927, | |
| "eval_runtime": 152.8061, | |
| "eval_samples_per_second": 237.883, | |
| "eval_steps_per_second": 7.434, | |
| "eval_wer": 0.20086396823585156, | |
| "step": 207828 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "grad_norm": 7.983453750610352, | |
| "learning_rate": 2.1232544569197956e-05, | |
| "loss": 0.4745, | |
| "step": 210840 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "eval_cer": 0.0370488724734612, | |
| "eval_loss": 0.13116249442100525, | |
| "eval_runtime": 151.9447, | |
| "eval_samples_per_second": 239.232, | |
| "eval_steps_per_second": 7.476, | |
| "eval_wer": 0.19823886058568607, | |
| "step": 210840 | |
| }, | |
| { | |
| "epoch": 71.0, | |
| "grad_norm": 23.63794708251953, | |
| "learning_rate": 2.052547386212725e-05, | |
| "loss": 0.4738, | |
| "step": 213852 | |
| }, | |
| { | |
| "epoch": 71.0, | |
| "eval_cer": 0.037366136884005016, | |
| "eval_loss": 0.1306936889886856, | |
| "eval_runtime": 154.0224, | |
| "eval_samples_per_second": 236.005, | |
| "eval_steps_per_second": 7.376, | |
| "eval_wer": 0.20006848106913475, | |
| "step": 213852 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "grad_norm": 10.838956832885742, | |
| "learning_rate": 1.9818637906287305e-05, | |
| "loss": 0.473, | |
| "step": 216864 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_cer": 0.0372285945094918, | |
| "eval_loss": 0.13071005046367645, | |
| "eval_runtime": 154.8642, | |
| "eval_samples_per_second": 234.722, | |
| "eval_steps_per_second": 7.335, | |
| "eval_wer": 0.19911043782921928, | |
| "step": 216864 | |
| }, | |
| { | |
| "epoch": 73.0, | |
| "grad_norm": 11.969744682312012, | |
| "learning_rate": 1.9111801950447367e-05, | |
| "loss": 0.472, | |
| "step": 219876 | |
| }, | |
| { | |
| "epoch": 73.0, | |
| "eval_cer": 0.03662890975661418, | |
| "eval_loss": 0.12924158573150635, | |
| "eval_runtime": 154.3055, | |
| "eval_samples_per_second": 235.572, | |
| "eval_steps_per_second": 7.362, | |
| "eval_wer": 0.19607375203627422, | |
| "step": 219876 | |
| }, | |
| { | |
| "epoch": 74.0, | |
| "grad_norm": 6.115599155426025, | |
| "learning_rate": 1.840496599460743e-05, | |
| "loss": 0.4693, | |
| "step": 222888 | |
| }, | |
| { | |
| "epoch": 74.0, | |
| "eval_cer": 0.036412051279465014, | |
| "eval_loss": 0.12866230309009552, | |
| "eval_runtime": 157.4725, | |
| "eval_samples_per_second": 230.834, | |
| "eval_steps_per_second": 7.214, | |
| "eval_wer": 0.19521600935216216, | |
| "step": 222888 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "grad_norm": 44.51272964477539, | |
| "learning_rate": 1.7698130038767486e-05, | |
| "loss": 0.4693, | |
| "step": 225900 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "eval_cer": 0.03628138602367746, | |
| "eval_loss": 0.12844808399677277, | |
| "eval_runtime": 164.6927, | |
| "eval_samples_per_second": 220.714, | |
| "eval_steps_per_second": 6.898, | |
| "eval_wer": 0.1944724017832747, | |
| "step": 225900 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "grad_norm": 9.162590026855469, | |
| "learning_rate": 1.6991528834158312e-05, | |
| "loss": 0.4664, | |
| "step": 228912 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_cer": 0.03683797416587427, | |
| "eval_loss": 0.12876588106155396, | |
| "eval_runtime": 152.2551, | |
| "eval_samples_per_second": 238.744, | |
| "eval_steps_per_second": 7.461, | |
| "eval_wer": 0.19688999104212276, | |
| "step": 228912 | |
| }, | |
| { | |
| "epoch": 77.0, | |
| "grad_norm": 3.896597146987915, | |
| "learning_rate": 1.6284692878318375e-05, | |
| "loss": 0.4651, | |
| "step": 231924 | |
| }, | |
| { | |
| "epoch": 77.0, | |
| "eval_cer": 0.03683384789463887, | |
| "eval_loss": 0.12869854271411896, | |
| "eval_runtime": 168.0203, | |
| "eval_samples_per_second": 216.343, | |
| "eval_steps_per_second": 6.761, | |
| "eval_wer": 0.1971009680732955, | |
| "step": 231924 | |
| }, | |
| { | |
| "epoch": 78.0, | |
| "grad_norm": 15.388148307800293, | |
| "learning_rate": 1.5577856922478434e-05, | |
| "loss": 0.4641, | |
| "step": 234936 | |
| }, | |
| { | |
| "epoch": 78.0, | |
| "eval_cer": 0.03656747416266495, | |
| "eval_loss": 0.1286703646183014, | |
| "eval_runtime": 154.3849, | |
| "eval_samples_per_second": 235.45, | |
| "eval_steps_per_second": 7.358, | |
| "eval_wer": 0.1952090920724516, | |
| "step": 234936 | |
| }, | |
| { | |
| "epoch": 79.0, | |
| "grad_norm": 2.635507345199585, | |
| "learning_rate": 1.4871020966638496e-05, | |
| "loss": 0.462, | |
| "step": 237948 | |
| }, | |
| { | |
| "epoch": 79.0, | |
| "eval_cer": 0.03641342670321015, | |
| "eval_loss": 0.12868022918701172, | |
| "eval_runtime": 152.2309, | |
| "eval_samples_per_second": 238.782, | |
| "eval_steps_per_second": 7.462, | |
| "eval_wer": 0.19447586042313, | |
| "step": 237948 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "grad_norm": 5.287237167358398, | |
| "learning_rate": 1.4163950259567787e-05, | |
| "loss": 0.4608, | |
| "step": 240960 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_cer": 0.036332276702247354, | |
| "eval_loss": 0.12745150923728943, | |
| "eval_runtime": 154.8487, | |
| "eval_samples_per_second": 234.745, | |
| "eval_steps_per_second": 7.336, | |
| "eval_wer": 0.19517796431375398, | |
| "step": 240960 | |
| }, | |
| { | |
| "epoch": 81.0, | |
| "grad_norm": 11.394750595092773, | |
| "learning_rate": 1.3457349054958616e-05, | |
| "loss": 0.4594, | |
| "step": 243972 | |
| }, | |
| { | |
| "epoch": 81.0, | |
| "eval_cer": 0.0360989131401566, | |
| "eval_loss": 0.12770119309425354, | |
| "eval_runtime": 152.6766, | |
| "eval_samples_per_second": 238.085, | |
| "eval_steps_per_second": 7.441, | |
| "eval_wer": 0.19389480892744118, | |
| "step": 243972 | |
| }, | |
| { | |
| "epoch": 82.0, | |
| "grad_norm": 9.252601623535156, | |
| "learning_rate": 1.2750278347887909e-05, | |
| "loss": 0.4595, | |
| "step": 246984 | |
| }, | |
| { | |
| "epoch": 82.0, | |
| "eval_cer": 0.03594715805361035, | |
| "eval_loss": 0.12681059539318085, | |
| "eval_runtime": 152.0339, | |
| "eval_samples_per_second": 239.091, | |
| "eval_steps_per_second": 7.472, | |
| "eval_wer": 0.19371841829482137, | |
| "step": 246984 | |
| }, | |
| { | |
| "epoch": 83.0, | |
| "grad_norm": 16.776588439941406, | |
| "learning_rate": 1.2043677143278735e-05, | |
| "loss": 0.4575, | |
| "step": 249996 | |
| }, | |
| { | |
| "epoch": 83.0, | |
| "eval_cer": 0.03615393008996188, | |
| "eval_loss": 0.12722131609916687, | |
| "eval_runtime": 152.0731, | |
| "eval_samples_per_second": 239.03, | |
| "eval_steps_per_second": 7.47, | |
| "eval_wer": 0.1942475901926808, | |
| "step": 249996 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "grad_norm": 8.937053680419922, | |
| "learning_rate": 1.1336606436208028e-05, | |
| "loss": 0.4569, | |
| "step": 253008 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "eval_cer": 0.03605260720740382, | |
| "eval_loss": 0.12680456042289734, | |
| "eval_runtime": 152.9563, | |
| "eval_samples_per_second": 237.65, | |
| "eval_steps_per_second": 7.427, | |
| "eval_wer": 0.19341059934770052, | |
| "step": 253008 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "grad_norm": 13.364398956298828, | |
| "learning_rate": 1.0630005231598857e-05, | |
| "loss": 0.4552, | |
| "step": 256020 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "eval_cer": 0.035670697880838785, | |
| "eval_loss": 0.12619073688983917, | |
| "eval_runtime": 156.3815, | |
| "eval_samples_per_second": 232.444, | |
| "eval_steps_per_second": 7.264, | |
| "eval_wer": 0.1916293998222259, | |
| "step": 256020 | |
| }, | |
| { | |
| "epoch": 86.0, | |
| "grad_norm": 9.33234691619873, | |
| "learning_rate": 9.923404026989685e-06, | |
| "loss": 0.4538, | |
| "step": 259032 | |
| }, | |
| { | |
| "epoch": 86.0, | |
| "eval_cer": 0.03549051737022648, | |
| "eval_loss": 0.12592804431915283, | |
| "eval_runtime": 156.5096, | |
| "eval_samples_per_second": 232.254, | |
| "eval_steps_per_second": 7.258, | |
| "eval_wer": 0.19070940162071864, | |
| "step": 259032 | |
| }, | |
| { | |
| "epoch": 87.0, | |
| "grad_norm": 4.286988735198975, | |
| "learning_rate": 9.216568071149744e-06, | |
| "loss": 0.4532, | |
| "step": 262044 | |
| }, | |
| { | |
| "epoch": 87.0, | |
| "eval_cer": 0.03551573347222057, | |
| "eval_loss": 0.12575581669807434, | |
| "eval_runtime": 155.6329, | |
| "eval_samples_per_second": 233.562, | |
| "eval_steps_per_second": 7.299, | |
| "eval_wer": 0.19122473895915693, | |
| "step": 262044 | |
| }, | |
| { | |
| "epoch": 88.0, | |
| "grad_norm": 7.920403957366943, | |
| "learning_rate": 8.509732115309804e-06, | |
| "loss": 0.4524, | |
| "step": 265056 | |
| }, | |
| { | |
| "epoch": 88.0, | |
| "eval_cer": 0.03555103601501229, | |
| "eval_loss": 0.1259673833847046, | |
| "eval_runtime": 166.3365, | |
| "eval_samples_per_second": 218.533, | |
| "eval_steps_per_second": 6.83, | |
| "eval_wer": 0.19095150641058897, | |
| "step": 265056 | |
| }, | |
| { | |
| "epoch": 89.0, | |
| "grad_norm": 9.81010913848877, | |
| "learning_rate": 7.802896159469865e-06, | |
| "loss": 0.4501, | |
| "step": 268068 | |
| }, | |
| { | |
| "epoch": 89.0, | |
| "eval_cer": 0.03596458008771536, | |
| "eval_loss": 0.12655647099018097, | |
| "eval_runtime": 155.3246, | |
| "eval_samples_per_second": 234.026, | |
| "eval_steps_per_second": 7.314, | |
| "eval_wer": 0.19276729233461648, | |
| "step": 268068 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "grad_norm": 10.394911766052246, | |
| "learning_rate": 7.096060203629924e-06, | |
| "loss": 0.4491, | |
| "step": 271080 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "eval_cer": 0.03546667669197752, | |
| "eval_loss": 0.12519720196723938, | |
| "eval_runtime": 177.0207, | |
| "eval_samples_per_second": 205.343, | |
| "eval_steps_per_second": 6.417, | |
| "eval_wer": 0.19042579315258482, | |
| "step": 271080 | |
| }, | |
| { | |
| "epoch": 91.0, | |
| "grad_norm": 9.381885528564453, | |
| "learning_rate": 6.3892242477899846e-06, | |
| "loss": 0.4486, | |
| "step": 274092 | |
| }, | |
| { | |
| "epoch": 91.0, | |
| "eval_cer": 0.03518288092589859, | |
| "eval_loss": 0.12525735795497894, | |
| "eval_runtime": 173.742, | |
| "eval_samples_per_second": 209.218, | |
| "eval_steps_per_second": 6.538, | |
| "eval_wer": 0.18893166073509932, | |
| "step": 274092 | |
| }, | |
| { | |
| "epoch": 92.0, | |
| "grad_norm": 26.816091537475586, | |
| "learning_rate": 5.682623043180811e-06, | |
| "loss": 0.4487, | |
| "step": 277104 | |
| }, | |
| { | |
| "epoch": 92.0, | |
| "eval_cer": 0.035369480080654846, | |
| "eval_loss": 0.12525933980941772, | |
| "eval_runtime": 153.4339, | |
| "eval_samples_per_second": 236.91, | |
| "eval_steps_per_second": 7.404, | |
| "eval_wer": 0.190249402519965, | |
| "step": 277104 | |
| }, | |
| { | |
| "epoch": 93.0, | |
| "grad_norm": 12.131733894348145, | |
| "learning_rate": 4.975552336110105e-06, | |
| "loss": 0.4471, | |
| "step": 280116 | |
| }, | |
| { | |
| "epoch": 93.0, | |
| "eval_cer": 0.03521634957036347, | |
| "eval_loss": 0.1251526027917862, | |
| "eval_runtime": 176.0621, | |
| "eval_samples_per_second": 206.461, | |
| "eval_steps_per_second": 6.452, | |
| "eval_wer": 0.1893743666365765, | |
| "step": 280116 | |
| }, | |
| { | |
| "epoch": 94.0, | |
| "grad_norm": 14.601805686950684, | |
| "learning_rate": 4.2687163802701646e-06, | |
| "loss": 0.4458, | |
| "step": 283128 | |
| }, | |
| { | |
| "epoch": 94.0, | |
| "eval_cer": 0.0351705021121924, | |
| "eval_loss": 0.1253127008676529, | |
| "eval_runtime": 177.0148, | |
| "eval_samples_per_second": 205.35, | |
| "eval_steps_per_second": 6.418, | |
| "eval_wer": 0.18914263776627205, | |
| "step": 283128 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "grad_norm": 3.8078722953796387, | |
| "learning_rate": 3.562115175660992e-06, | |
| "loss": 0.4449, | |
| "step": 286140 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "eval_cer": 0.035079265670431965, | |
| "eval_loss": 0.12475291639566422, | |
| "eval_runtime": 168.7285, | |
| "eval_samples_per_second": 215.435, | |
| "eval_steps_per_second": 6.733, | |
| "eval_wer": 0.1884266993162269, | |
| "step": 286140 | |
| }, | |
| { | |
| "epoch": 96.0, | |
| "grad_norm": 4.562527656555176, | |
| "learning_rate": 2.855044468590285e-06, | |
| "loss": 0.4434, | |
| "step": 289152 | |
| }, | |
| { | |
| "epoch": 96.0, | |
| "eval_cer": 0.035126030077766456, | |
| "eval_loss": 0.1246921494603157, | |
| "eval_runtime": 179.6539, | |
| "eval_samples_per_second": 202.333, | |
| "eval_steps_per_second": 6.323, | |
| "eval_wer": 0.18909421680829797, | |
| "step": 289152 | |
| }, | |
| { | |
| "epoch": 97.0, | |
| "grad_norm": 8.47780990600586, | |
| "learning_rate": 2.1482085127503454e-06, | |
| "loss": 0.4435, | |
| "step": 292164 | |
| }, | |
| { | |
| "epoch": 97.0, | |
| "eval_cer": 0.035188841095460825, | |
| "eval_loss": 0.12471602112054825, | |
| "eval_runtime": 152.6845, | |
| "eval_samples_per_second": 238.073, | |
| "eval_steps_per_second": 7.44, | |
| "eval_wer": 0.18912880320685088, | |
| "step": 292164 | |
| }, | |
| { | |
| "epoch": 98.0, | |
| "grad_norm": 16.65981101989746, | |
| "learning_rate": 1.4413725569104053e-06, | |
| "loss": 0.4444, | |
| "step": 295176 | |
| }, | |
| { | |
| "epoch": 98.0, | |
| "eval_cer": 0.03511594363696882, | |
| "eval_loss": 0.12448572367429733, | |
| "eval_runtime": 159.0007, | |
| "eval_samples_per_second": 228.615, | |
| "eval_steps_per_second": 7.145, | |
| "eval_wer": 0.1887518114626242, | |
| "step": 295176 | |
| }, | |
| { | |
| "epoch": 99.0, | |
| "grad_norm": 10.248809814453125, | |
| "learning_rate": 7.345366010704655e-07, | |
| "loss": 0.4429, | |
| "step": 298188 | |
| }, | |
| { | |
| "epoch": 99.0, | |
| "eval_cer": 0.03512648855234817, | |
| "eval_loss": 0.12444119900465012, | |
| "eval_runtime": 155.6325, | |
| "eval_samples_per_second": 233.563, | |
| "eval_steps_per_second": 7.299, | |
| "eval_wer": 0.18870684914450545, | |
| "step": 298188 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 301200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 100, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 10, | |
| "early_stopping_threshold": 0.001 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 8 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.0954167922548843e+21, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |