| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 60.0, | |
| "global_step": 7440, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 12.2551, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "eval_loss": 4.356050968170166, | |
| "eval_runtime": 197.2044, | |
| "eval_samples_per_second": 9.351, | |
| "eval_wer": 1.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 3.2085, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "eval_loss": 2.925891876220703, | |
| "eval_runtime": 199.5724, | |
| "eval_samples_per_second": 9.24, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 0.0001, | |
| "loss": 2.8642, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "eval_loss": 2.8566176891326904, | |
| "eval_runtime": 201.2361, | |
| "eval_samples_per_second": 9.163, | |
| "eval_wer": 1.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 9.859943977591037e-05, | |
| "loss": 2.7372, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "eval_loss": 2.3264265060424805, | |
| "eval_runtime": 202.5449, | |
| "eval_samples_per_second": 9.104, | |
| "eval_wer": 1.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 9.719887955182073e-05, | |
| "loss": 1.4523, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "eval_loss": 0.6223726868629456, | |
| "eval_runtime": 203.3756, | |
| "eval_samples_per_second": 9.067, | |
| "eval_wer": 0.7100503449561812, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "learning_rate": 9.579831932773111e-05, | |
| "loss": 0.7072, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "eval_loss": 0.4705396890640259, | |
| "eval_runtime": 205.0375, | |
| "eval_samples_per_second": 8.993, | |
| "eval_wer": 0.5925787805332836, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "learning_rate": 9.439775910364146e-05, | |
| "loss": 0.5552, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "eval_loss": 0.39274999499320984, | |
| "eval_runtime": 205.4744, | |
| "eval_samples_per_second": 8.974, | |
| "eval_wer": 0.536826403132575, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "learning_rate": 9.299719887955183e-05, | |
| "loss": 0.4751, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "eval_loss": 0.36413395404815674, | |
| "eval_runtime": 206.0611, | |
| "eval_samples_per_second": 8.949, | |
| "eval_wer": 0.5096960656349059, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 7.26, | |
| "learning_rate": 9.159663865546218e-05, | |
| "loss": 0.4116, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 7.26, | |
| "eval_loss": 0.34078019857406616, | |
| "eval_runtime": 206.851, | |
| "eval_samples_per_second": 8.915, | |
| "eval_wer": 0.4906768599664367, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "learning_rate": 9.019607843137255e-05, | |
| "loss": 0.3737, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "eval_loss": 0.3288322687149048, | |
| "eval_runtime": 207.383, | |
| "eval_samples_per_second": 8.892, | |
| "eval_wer": 0.4697930262912549, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.87, | |
| "learning_rate": 8.879551820728292e-05, | |
| "loss": 0.338, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 8.87, | |
| "eval_loss": 0.3296125531196594, | |
| "eval_runtime": 207.5205, | |
| "eval_samples_per_second": 8.886, | |
| "eval_wer": 0.4619615886630617, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 9.68, | |
| "learning_rate": 8.739495798319329e-05, | |
| "loss": 0.3154, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 9.68, | |
| "eval_loss": 0.34391966462135315, | |
| "eval_runtime": 207.9354, | |
| "eval_samples_per_second": 8.868, | |
| "eval_wer": 0.4543166138355398, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 10.48, | |
| "learning_rate": 8.599439775910365e-05, | |
| "loss": 0.3054, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.48, | |
| "eval_loss": 0.3120010197162628, | |
| "eval_runtime": 206.6673, | |
| "eval_samples_per_second": 8.923, | |
| "eval_wer": 0.44173037479022936, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 11.29, | |
| "learning_rate": 8.4593837535014e-05, | |
| "loss": 0.2766, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 11.29, | |
| "eval_loss": 0.3251303434371948, | |
| "eval_runtime": 208.2654, | |
| "eval_samples_per_second": 8.854, | |
| "eval_wer": 0.439026664180496, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 12.1, | |
| "learning_rate": 8.319327731092437e-05, | |
| "loss": 0.2743, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 12.1, | |
| "eval_loss": 0.31674692034721375, | |
| "eval_runtime": 207.1288, | |
| "eval_samples_per_second": 8.903, | |
| "eval_wer": 0.43361924296102927, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "learning_rate": 8.179271708683474e-05, | |
| "loss": 0.2615, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "eval_loss": 0.31189170479774475, | |
| "eval_runtime": 206.6256, | |
| "eval_samples_per_second": 8.924, | |
| "eval_wer": 0.43492448256572813, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 13.71, | |
| "learning_rate": 8.039215686274511e-05, | |
| "loss": 0.2349, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 13.71, | |
| "eval_loss": 0.3296045660972595, | |
| "eval_runtime": 208.4817, | |
| "eval_samples_per_second": 8.845, | |
| "eval_wer": 0.42988998694760394, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 14.52, | |
| "learning_rate": 7.899159663865546e-05, | |
| "loss": 0.228, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 14.52, | |
| "eval_loss": 0.30033424496650696, | |
| "eval_runtime": 207.6073, | |
| "eval_samples_per_second": 8.882, | |
| "eval_wer": 0.4192616073093418, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 15.32, | |
| "learning_rate": 7.759103641456583e-05, | |
| "loss": 0.2243, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 15.32, | |
| "eval_loss": 0.3134298026561737, | |
| "eval_runtime": 207.5471, | |
| "eval_samples_per_second": 8.885, | |
| "eval_wer": 0.4184225247063211, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 16.13, | |
| "learning_rate": 7.619047619047618e-05, | |
| "loss": 0.2078, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 16.13, | |
| "eval_loss": 0.3405207097530365, | |
| "eval_runtime": 208.3189, | |
| "eval_samples_per_second": 8.852, | |
| "eval_wer": 0.4215923923177326, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 16.94, | |
| "learning_rate": 7.478991596638657e-05, | |
| "loss": 0.2127, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 16.94, | |
| "eval_loss": 0.31670939922332764, | |
| "eval_runtime": 207.0917, | |
| "eval_samples_per_second": 8.904, | |
| "eval_wer": 0.4139474174902107, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 17.74, | |
| "learning_rate": 7.338935574229692e-05, | |
| "loss": 0.2032, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 17.74, | |
| "eval_loss": 0.30317744612693787, | |
| "eval_runtime": 207.7221, | |
| "eval_samples_per_second": 8.877, | |
| "eval_wer": 0.4090993846727578, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 18.55, | |
| "learning_rate": 7.198879551820729e-05, | |
| "loss": 0.1853, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 18.55, | |
| "eval_loss": 0.3155224025249481, | |
| "eval_runtime": 207.5316, | |
| "eval_samples_per_second": 8.885, | |
| "eval_wer": 0.40471750885698304, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 19.35, | |
| "learning_rate": 7.058823529411765e-05, | |
| "loss": 0.1844, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 19.35, | |
| "eval_loss": 0.3215440511703491, | |
| "eval_runtime": 207.3486, | |
| "eval_samples_per_second": 8.893, | |
| "eval_wer": 0.4049972030579899, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 20.16, | |
| "learning_rate": 6.918767507002801e-05, | |
| "loss": 0.1871, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 20.16, | |
| "eval_loss": 0.30251652002334595, | |
| "eval_runtime": 207.2199, | |
| "eval_samples_per_second": 8.899, | |
| "eval_wer": 0.3992168562371807, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 20.97, | |
| "learning_rate": 6.778711484593838e-05, | |
| "loss": 0.1896, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 20.97, | |
| "eval_loss": 0.3048883080482483, | |
| "eval_runtime": 207.9453, | |
| "eval_samples_per_second": 8.868, | |
| "eval_wer": 0.39996270743986573, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 21.77, | |
| "learning_rate": 6.638655462184874e-05, | |
| "loss": 0.1775, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 21.77, | |
| "eval_loss": 0.32283803820610046, | |
| "eval_runtime": 209.0912, | |
| "eval_samples_per_second": 8.819, | |
| "eval_wer": 0.40564982286033935, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 22.58, | |
| "learning_rate": 6.498599439775911e-05, | |
| "loss": 0.1667, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 22.58, | |
| "eval_loss": 0.31883877515792847, | |
| "eval_runtime": 207.5813, | |
| "eval_samples_per_second": 8.883, | |
| "eval_wer": 0.4055565914600037, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 23.39, | |
| "learning_rate": 6.358543417366946e-05, | |
| "loss": 0.1714, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 23.39, | |
| "eval_loss": 0.32032692432403564, | |
| "eval_runtime": 208.3503, | |
| "eval_samples_per_second": 8.85, | |
| "eval_wer": 0.3992168562371807, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 24.19, | |
| "learning_rate": 6.218487394957983e-05, | |
| "loss": 0.1692, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 24.19, | |
| "eval_loss": 0.33072519302368164, | |
| "eval_runtime": 208.6398, | |
| "eval_samples_per_second": 8.838, | |
| "eval_wer": 0.40369196345329106, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 6.078431372549019e-05, | |
| "loss": 0.1759, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_loss": 0.3215318024158478, | |
| "eval_runtime": 207.9891, | |
| "eval_samples_per_second": 8.866, | |
| "eval_wer": 0.39912362483684505, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 25.81, | |
| "learning_rate": 5.938375350140056e-05, | |
| "loss": 0.1561, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 25.81, | |
| "eval_loss": 0.32872405648231506, | |
| "eval_runtime": 208.2303, | |
| "eval_samples_per_second": 8.856, | |
| "eval_wer": 0.40164087264590714, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 26.61, | |
| "learning_rate": 5.7983193277310935e-05, | |
| "loss": 0.1459, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 26.61, | |
| "eval_loss": 0.3299988806247711, | |
| "eval_runtime": 208.2294, | |
| "eval_samples_per_second": 8.856, | |
| "eval_wer": 0.40173410404624277, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 27.42, | |
| "learning_rate": 5.658263305322129e-05, | |
| "loss": 0.1489, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 27.42, | |
| "eval_loss": 0.3266052305698395, | |
| "eval_runtime": 207.8609, | |
| "eval_samples_per_second": 8.871, | |
| "eval_wer": 0.3958605258250979, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 28.23, | |
| "learning_rate": 5.5182072829131656e-05, | |
| "loss": 0.1526, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 28.23, | |
| "eval_loss": 0.35196033120155334, | |
| "eval_runtime": 208.9297, | |
| "eval_samples_per_second": 8.826, | |
| "eval_wer": 0.39800484803281744, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 29.03, | |
| "learning_rate": 5.378151260504202e-05, | |
| "loss": 0.1533, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 29.03, | |
| "eval_loss": 0.3176701068878174, | |
| "eval_runtime": 208.8703, | |
| "eval_samples_per_second": 8.828, | |
| "eval_wer": 0.39567406302442665, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 29.84, | |
| "learning_rate": 5.2380952380952384e-05, | |
| "loss": 0.1396, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 29.84, | |
| "eval_loss": 0.32652196288108826, | |
| "eval_runtime": 208.1553, | |
| "eval_samples_per_second": 8.859, | |
| "eval_wer": 0.3936229722170427, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 30.65, | |
| "learning_rate": 5.0980392156862745e-05, | |
| "loss": 0.1392, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 30.65, | |
| "eval_loss": 0.31360137462615967, | |
| "eval_runtime": 209.7641, | |
| "eval_samples_per_second": 8.791, | |
| "eval_wer": 0.39119895580831626, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 31.45, | |
| "learning_rate": 4.957983193277311e-05, | |
| "loss": 0.1358, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 31.45, | |
| "eval_loss": 0.31799858808517456, | |
| "eval_runtime": 209.0208, | |
| "eval_samples_per_second": 8.822, | |
| "eval_wer": 0.3893343278016036, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 32.26, | |
| "learning_rate": 4.817927170868347e-05, | |
| "loss": 0.1401, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 32.26, | |
| "eval_loss": 0.3084213137626648, | |
| "eval_runtime": 209.0337, | |
| "eval_samples_per_second": 8.822, | |
| "eval_wer": 0.38607122878985645, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 33.06, | |
| "learning_rate": 4.677871148459384e-05, | |
| "loss": 0.1327, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 33.06, | |
| "eval_loss": 0.3314739763736725, | |
| "eval_runtime": 209.2003, | |
| "eval_samples_per_second": 8.815, | |
| "eval_wer": 0.3857915345888495, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 33.87, | |
| "learning_rate": 4.53781512605042e-05, | |
| "loss": 0.1273, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 33.87, | |
| "eval_loss": 0.32433462142944336, | |
| "eval_runtime": 208.6782, | |
| "eval_samples_per_second": 8.837, | |
| "eval_wer": 0.388215550997576, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 34.68, | |
| "learning_rate": 4.397759103641457e-05, | |
| "loss": 2.8674, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 34.68, | |
| "eval_loss": 0.4206547737121582, | |
| "eval_runtime": 255.7774, | |
| "eval_samples_per_second": 7.209, | |
| "eval_wer": 0.2714898377773634, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 35.48, | |
| "learning_rate": 4.257703081232493e-05, | |
| "loss": 0.2443, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 35.48, | |
| "eval_loss": 0.3481670320034027, | |
| "eval_runtime": 258.3191, | |
| "eval_samples_per_second": 7.138, | |
| "eval_wer": 0.2341972776431102, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 36.29, | |
| "learning_rate": 4.11764705882353e-05, | |
| "loss": 0.1918, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 36.29, | |
| "eval_loss": 0.35066932439804077, | |
| "eval_runtime": 260.0692, | |
| "eval_samples_per_second": 7.09, | |
| "eval_wer": 0.23074771583069179, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 37.1, | |
| "learning_rate": 3.977591036414566e-05, | |
| "loss": 0.1678, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 37.1, | |
| "eval_loss": 0.3326978087425232, | |
| "eval_runtime": 260.7521, | |
| "eval_samples_per_second": 7.072, | |
| "eval_wer": 0.22673876561625955, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 37.9, | |
| "learning_rate": 3.8375350140056026e-05, | |
| "loss": 0.1544, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 37.9, | |
| "eval_loss": 0.3412529230117798, | |
| "eval_runtime": 260.8062, | |
| "eval_samples_per_second": 7.07, | |
| "eval_wer": 0.22440798060786873, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 38.71, | |
| "learning_rate": 3.697478991596639e-05, | |
| "loss": 0.1475, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 38.71, | |
| "eval_loss": 0.33288225531578064, | |
| "eval_runtime": 262.5936, | |
| "eval_samples_per_second": 7.022, | |
| "eval_wer": 0.22077195599477903, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 39.52, | |
| "learning_rate": 3.5574229691876754e-05, | |
| "loss": 0.1447, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 39.52, | |
| "eval_loss": 0.3270339071750641, | |
| "eval_runtime": 262.0824, | |
| "eval_samples_per_second": 7.036, | |
| "eval_wer": 0.21909379078873764, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 40.32, | |
| "learning_rate": 3.4173669467787114e-05, | |
| "loss": 0.1402, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 40.32, | |
| "eval_loss": 0.3369254171848297, | |
| "eval_runtime": 263.645, | |
| "eval_samples_per_second": 6.994, | |
| "eval_wer": 0.2214245757971285, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 41.13, | |
| "learning_rate": 3.277310924369748e-05, | |
| "loss": 0.1413, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 41.13, | |
| "eval_loss": 0.3336828649044037, | |
| "eval_runtime": 263.0331, | |
| "eval_samples_per_second": 7.011, | |
| "eval_wer": 0.21946671639008017, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 41.94, | |
| "learning_rate": 3.137254901960784e-05, | |
| "loss": 0.1381, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 41.94, | |
| "eval_loss": 0.323373407125473, | |
| "eval_runtime": 262.7696, | |
| "eval_samples_per_second": 7.018, | |
| "eval_wer": 0.214432220771956, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 42.74, | |
| "learning_rate": 2.9971988795518207e-05, | |
| "loss": 0.1257, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 42.74, | |
| "eval_loss": 0.32813942432403564, | |
| "eval_runtime": 262.9096, | |
| "eval_samples_per_second": 7.014, | |
| "eval_wer": 0.21629684877866864, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 43.55, | |
| "learning_rate": 2.857142857142857e-05, | |
| "loss": 0.1276, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 43.55, | |
| "eval_loss": 0.3204418420791626, | |
| "eval_runtime": 262.6036, | |
| "eval_samples_per_second": 7.022, | |
| "eval_wer": 0.21946671639008017, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 44.35, | |
| "learning_rate": 2.7170868347338935e-05, | |
| "loss": 0.1248, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 44.35, | |
| "eval_loss": 0.3329363465309143, | |
| "eval_runtime": 273.4973, | |
| "eval_samples_per_second": 6.742, | |
| "eval_wer": 0.21722916278202498, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 45.16, | |
| "learning_rate": 2.57703081232493e-05, | |
| "loss": 0.1287, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 45.16, | |
| "eval_loss": 0.31906744837760925, | |
| "eval_runtime": 274.362, | |
| "eval_samples_per_second": 6.721, | |
| "eval_wer": 0.21284728696625024, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 45.97, | |
| "learning_rate": 2.4369747899159663e-05, | |
| "loss": 0.1215, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 45.97, | |
| "eval_loss": 0.3277473747730255, | |
| "eval_runtime": 273.8431, | |
| "eval_samples_per_second": 6.734, | |
| "eval_wer": 0.21517807197464106, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 46.77, | |
| "learning_rate": 2.2969187675070027e-05, | |
| "loss": 0.1219, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 46.77, | |
| "eval_loss": 0.3112068772315979, | |
| "eval_runtime": 272.6473, | |
| "eval_samples_per_second": 6.763, | |
| "eval_wer": 0.20995711355584562, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 47.58, | |
| "learning_rate": 2.1568627450980395e-05, | |
| "loss": 0.1179, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 47.58, | |
| "eval_loss": 0.3236621618270874, | |
| "eval_runtime": 264.1309, | |
| "eval_samples_per_second": 6.981, | |
| "eval_wer": 0.21461868357262726, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 48.39, | |
| "learning_rate": 2.016806722689076e-05, | |
| "loss": 0.1188, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 48.39, | |
| "eval_loss": 0.3213033080101013, | |
| "eval_runtime": 263.9442, | |
| "eval_samples_per_second": 6.986, | |
| "eval_wer": 0.21126235316054448, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 49.19, | |
| "learning_rate": 1.876750700280112e-05, | |
| "loss": 0.1243, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 49.19, | |
| "eval_loss": 0.32063964009284973, | |
| "eval_runtime": 265.6074, | |
| "eval_samples_per_second": 6.943, | |
| "eval_wer": 0.21368636956927092, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 1.7366946778711484e-05, | |
| "loss": 0.1162, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_loss": 0.31902745366096497, | |
| "eval_runtime": 264.7304, | |
| "eval_samples_per_second": 6.966, | |
| "eval_wer": 0.20949095655416744, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 50.81, | |
| "learning_rate": 1.5966386554621848e-05, | |
| "loss": 0.1274, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 50.81, | |
| "eval_loss": 0.3174721896648407, | |
| "eval_runtime": 265.1372, | |
| "eval_samples_per_second": 6.955, | |
| "eval_wer": 0.21070296475853068, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 51.61, | |
| "learning_rate": 1.4565826330532215e-05, | |
| "loss": 0.1179, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 51.61, | |
| "eval_loss": 0.32639437913894653, | |
| "eval_runtime": 264.7128, | |
| "eval_samples_per_second": 6.966, | |
| "eval_wer": 0.2127540555659146, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 52.42, | |
| "learning_rate": 1.3165266106442578e-05, | |
| "loss": 0.1084, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 52.42, | |
| "eval_loss": 0.3261864185333252, | |
| "eval_runtime": 265.2652, | |
| "eval_samples_per_second": 6.952, | |
| "eval_wer": 0.211635278761887, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 53.23, | |
| "learning_rate": 1.1764705882352942e-05, | |
| "loss": 0.123, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 53.23, | |
| "eval_loss": 0.31353211402893066, | |
| "eval_runtime": 265.8012, | |
| "eval_samples_per_second": 6.938, | |
| "eval_wer": 0.2119149729628939, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 54.03, | |
| "learning_rate": 1.0364145658263306e-05, | |
| "loss": 0.1077, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 54.03, | |
| "eval_loss": 0.31938183307647705, | |
| "eval_runtime": 265.3957, | |
| "eval_samples_per_second": 6.948, | |
| "eval_wer": 0.21060973335819505, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 54.84, | |
| "learning_rate": 8.96358543417367e-06, | |
| "loss": 0.1088, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 54.84, | |
| "eval_loss": 0.3212938904762268, | |
| "eval_runtime": 266.6583, | |
| "eval_samples_per_second": 6.915, | |
| "eval_wer": 0.21088942755920195, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 55.65, | |
| "learning_rate": 7.563025210084033e-06, | |
| "loss": 0.1085, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 55.65, | |
| "eval_loss": 0.320252925157547, | |
| "eval_runtime": 266.7391, | |
| "eval_samples_per_second": 6.913, | |
| "eval_wer": 0.21060973335819505, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 56.45, | |
| "learning_rate": 6.162464985994398e-06, | |
| "loss": 0.1084, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 56.45, | |
| "eval_loss": 0.32074451446533203, | |
| "eval_runtime": 265.9571, | |
| "eval_samples_per_second": 6.933, | |
| "eval_wer": 0.2093044937534962, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 57.26, | |
| "learning_rate": 4.7619047619047615e-06, | |
| "loss": 0.1086, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 57.26, | |
| "eval_loss": 0.3232016861438751, | |
| "eval_runtime": 267.1546, | |
| "eval_samples_per_second": 6.902, | |
| "eval_wer": 0.20874510535148239, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 58.06, | |
| "learning_rate": 3.361344537815126e-06, | |
| "loss": 0.106, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 58.06, | |
| "eval_loss": 0.3231372535228729, | |
| "eval_runtime": 266.9142, | |
| "eval_samples_per_second": 6.909, | |
| "eval_wer": 0.2093044937534962, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 58.87, | |
| "learning_rate": 1.96078431372549e-06, | |
| "loss": 0.1097, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 58.87, | |
| "eval_loss": 0.32339319586753845, | |
| "eval_runtime": 270.5457, | |
| "eval_samples_per_second": 6.816, | |
| "eval_wer": 0.208838336751818, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 59.68, | |
| "learning_rate": 5.602240896358544e-07, | |
| "loss": 0.1033, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 59.68, | |
| "eval_loss": 0.32322752475738525, | |
| "eval_runtime": 275.739, | |
| "eval_samples_per_second": 6.687, | |
| "eval_wer": 0.208838336751818, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "step": 7440, | |
| "total_flos": 2.7385342136046084e+19, | |
| "train_runtime": 22157.5951, | |
| "train_samples_per_second": 0.336 | |
| } | |
| ], | |
| "max_steps": 7440, | |
| "num_train_epochs": 60, | |
| "total_flos": 2.7385342136046084e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |