| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 29.999581414817914, | |
| "global_step": 17910, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 4.9000000000000005e-05, | |
| "loss": 5.3488, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 9.900000000000001e-05, | |
| "loss": 3.2572, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.000149, | |
| "loss": 2.4392, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.000199, | |
| "loss": 2.2566, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.000249, | |
| "loss": 2.2416, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "eval_loss": 1.2866647243499756, | |
| "eval_runtime": 434.4225, | |
| "eval_samples_per_second": 23.912, | |
| "eval_steps_per_second": 2.99, | |
| "eval_wer": 0.8874521338587047, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 0.000299, | |
| "loss": 2.2596, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 0.00034899999999999997, | |
| "loss": 2.2575, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 0.00039900000000000005, | |
| "loss": 2.2978, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 0.000449, | |
| "loss": 2.2998, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 0.000499, | |
| "loss": 2.3089, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "eval_loss": 1.8336485624313354, | |
| "eval_runtime": 430.7741, | |
| "eval_samples_per_second": 24.115, | |
| "eval_steps_per_second": 3.016, | |
| "eval_wer": 0.9547514660451005, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 0.000549, | |
| "loss": 2.3156, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 0.000599, | |
| "loss": 2.3298, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 0.0006490000000000001, | |
| "loss": 2.3174, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 0.000699, | |
| "loss": 2.349, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 0.000749, | |
| "loss": 2.3614, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "eval_loss": 1.5936506986618042, | |
| "eval_runtime": 430.1389, | |
| "eval_samples_per_second": 24.15, | |
| "eval_steps_per_second": 3.02, | |
| "eval_wer": 0.9468893945279983, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 0.000799, | |
| "loss": 2.3597, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 0.000849, | |
| "loss": 2.397, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 0.0008990000000000001, | |
| "loss": 2.4454, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "learning_rate": 0.000949, | |
| "loss": 2.4806, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 0.000999, | |
| "loss": 2.5234, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "eval_loss": 1.9764641523361206, | |
| "eval_runtime": 434.6883, | |
| "eval_samples_per_second": 23.898, | |
| "eval_steps_per_second": 2.988, | |
| "eval_wer": 0.9866992248922434, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "learning_rate": 0.0009753148614609572, | |
| "loss": 2.5559, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 0.0009501259445843828, | |
| "loss": 2.5543, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "learning_rate": 0.0009249370277078086, | |
| "loss": 2.5819, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 0.0008997481108312343, | |
| "loss": 2.5837, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "learning_rate": 0.00087455919395466, | |
| "loss": 2.5373, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "eval_loss": 1.9062319993972778, | |
| "eval_runtime": 430.3022, | |
| "eval_samples_per_second": 24.141, | |
| "eval_steps_per_second": 3.019, | |
| "eval_wer": 0.9916014577205542, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "learning_rate": 0.0008493702770780856, | |
| "loss": 2.5617, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 0.0008241813602015113, | |
| "loss": 2.5553, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 4.69, | |
| "learning_rate": 0.0007989924433249371, | |
| "loss": 2.549, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 4.86, | |
| "learning_rate": 0.0007738035264483628, | |
| "loss": 2.5636, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 0.0007486146095717884, | |
| "loss": 2.5703, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "eval_loss": 1.977164387702942, | |
| "eval_runtime": 431.1212, | |
| "eval_samples_per_second": 24.095, | |
| "eval_steps_per_second": 3.013, | |
| "eval_wer": 0.9914719647779197, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.19, | |
| "learning_rate": 0.0007234256926952141, | |
| "loss": 2.5526, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 5.36, | |
| "learning_rate": 0.0006982367758186398, | |
| "loss": 2.5277, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 5.53, | |
| "learning_rate": 0.0006730478589420656, | |
| "loss": 2.4969, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 5.69, | |
| "learning_rate": 0.0006478589420654912, | |
| "loss": 2.4849, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 5.86, | |
| "learning_rate": 0.0006226700251889169, | |
| "loss": 2.4656, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 5.86, | |
| "eval_loss": 1.8083465099334717, | |
| "eval_runtime": 432.7157, | |
| "eval_samples_per_second": 24.007, | |
| "eval_steps_per_second": 3.002, | |
| "eval_wer": 0.9829069315722293, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "learning_rate": 0.0005974811083123426, | |
| "loss": 2.4858, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 6.2, | |
| "learning_rate": 0.0005722921914357682, | |
| "loss": 2.4552, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 6.37, | |
| "learning_rate": 0.0005471032745591939, | |
| "loss": 2.4302, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 6.53, | |
| "learning_rate": 0.0005219143576826196, | |
| "loss": 2.4397, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 6.7, | |
| "learning_rate": 0.0004967254408060454, | |
| "loss": 2.4339, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 6.7, | |
| "eval_loss": 1.754757046699524, | |
| "eval_runtime": 442.6863, | |
| "eval_samples_per_second": 23.466, | |
| "eval_steps_per_second": 2.934, | |
| "eval_wer": 0.9752483489649814, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 6.87, | |
| "learning_rate": 0.00047153652392947104, | |
| "loss": 2.4069, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 0.0004463476070528967, | |
| "loss": 2.3863, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "learning_rate": 0.00042115869017632243, | |
| "loss": 2.3614, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "learning_rate": 0.0003959697732997481, | |
| "loss": 2.3534, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 7.54, | |
| "learning_rate": 0.00037078085642317383, | |
| "loss": 2.344, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 7.54, | |
| "eval_loss": 1.6146422624588013, | |
| "eval_runtime": 432.0313, | |
| "eval_samples_per_second": 24.045, | |
| "eval_steps_per_second": 3.007, | |
| "eval_wer": 0.9638344710213294, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 7.7, | |
| "learning_rate": 0.0003455919395465995, | |
| "loss": 2.3322, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 7.87, | |
| "learning_rate": 0.0003204030226700252, | |
| "loss": 2.315, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 8.04, | |
| "learning_rate": 0.00029521410579345085, | |
| "loss": 2.3035, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 8.21, | |
| "learning_rate": 0.0002700251889168766, | |
| "loss": 2.2715, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 8.38, | |
| "learning_rate": 0.00024483627204030224, | |
| "loss": 2.2677, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 8.38, | |
| "eval_loss": 1.5104962587356567, | |
| "eval_runtime": 431.1839, | |
| "eval_samples_per_second": 24.092, | |
| "eval_steps_per_second": 3.013, | |
| "eval_wer": 0.9499232291840095, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 8.54, | |
| "learning_rate": 0.00021964735516372797, | |
| "loss": 2.266, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 8.71, | |
| "learning_rate": 0.00019445843828715364, | |
| "loss": 2.2473, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 8.88, | |
| "learning_rate": 0.00016926952141057937, | |
| "loss": 2.2419, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 9.05, | |
| "learning_rate": 0.00014408060453400504, | |
| "loss": 2.2305, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 9.21, | |
| "learning_rate": 0.00011889168765743074, | |
| "loss": 2.2074, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 9.21, | |
| "eval_loss": 1.4190884828567505, | |
| "eval_runtime": 440.4999, | |
| "eval_samples_per_second": 23.582, | |
| "eval_steps_per_second": 2.949, | |
| "eval_wer": 0.9356790054942006, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 9.38, | |
| "learning_rate": 0.000773852922690132, | |
| "loss": 2.3213, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 9.55, | |
| "learning_rate": 0.0007675675675675676, | |
| "loss": 2.3412, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 9.71, | |
| "learning_rate": 0.0007612822124450032, | |
| "loss": 2.3508, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 9.88, | |
| "learning_rate": 0.0007549968573224387, | |
| "loss": 2.3767, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 10.05, | |
| "learning_rate": 0.0007487115021998742, | |
| "loss": 2.3768, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 10.05, | |
| "eval_loss": 1.6662662029266357, | |
| "eval_runtime": 424.8561, | |
| "eval_samples_per_second": 24.451, | |
| "eval_steps_per_second": 3.058, | |
| "eval_wer": 0.9664798268494367, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 10.22, | |
| "learning_rate": 0.0007424261470773099, | |
| "loss": 2.3681, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 10.39, | |
| "learning_rate": 0.0007361407919547455, | |
| "loss": 2.389, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 10.55, | |
| "learning_rate": 0.0007299182903834066, | |
| "loss": 2.3671, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 10.72, | |
| "learning_rate": 0.0007236329352608423, | |
| "loss": 2.3784, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 10.89, | |
| "learning_rate": 0.0007173475801382778, | |
| "loss": 2.3804, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 10.89, | |
| "eval_loss": 1.6570764780044556, | |
| "eval_runtime": 413.7975, | |
| "eval_samples_per_second": 25.104, | |
| "eval_steps_per_second": 3.139, | |
| "eval_wer": 0.9719740274155059, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 11.06, | |
| "learning_rate": 0.0007110622250157134, | |
| "loss": 2.3757, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 11.22, | |
| "learning_rate": 0.0007047768698931489, | |
| "loss": 2.3487, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 11.39, | |
| "learning_rate": 0.0006984915147705846, | |
| "loss": 2.3473, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 11.56, | |
| "learning_rate": 0.0006922061596480202, | |
| "loss": 2.3481, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 11.72, | |
| "learning_rate": 0.0006859208045254557, | |
| "loss": 2.3237, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 11.72, | |
| "eval_loss": 1.604884147644043, | |
| "eval_runtime": 412.0866, | |
| "eval_samples_per_second": 25.208, | |
| "eval_steps_per_second": 3.152, | |
| "eval_wer": 0.9637049780786947, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 11.89, | |
| "learning_rate": 0.0006796354494028913, | |
| "loss": 2.3379, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 12.06, | |
| "learning_rate": 0.0006733500942803269, | |
| "loss": 2.3362, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 12.23, | |
| "learning_rate": 0.0006670647391577624, | |
| "loss": 2.3148, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 12.4, | |
| "learning_rate": 0.000660779384035198, | |
| "loss": 2.3242, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 12.56, | |
| "learning_rate": 0.0006544940289126335, | |
| "loss": 2.317, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 12.56, | |
| "eval_loss": 1.5874534845352173, | |
| "eval_runtime": 412.3759, | |
| "eval_samples_per_second": 25.191, | |
| "eval_steps_per_second": 3.15, | |
| "eval_wer": 0.9655363782673845, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 12.73, | |
| "learning_rate": 0.0006482086737900693, | |
| "loss": 2.3342, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "learning_rate": 0.0006419233186675048, | |
| "loss": 2.3229, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 13.07, | |
| "learning_rate": 0.0006356379635449403, | |
| "loss": 2.3128, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 13.23, | |
| "learning_rate": 0.0006293526084223759, | |
| "loss": 2.2992, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 13.4, | |
| "learning_rate": 0.0006230672532998114, | |
| "loss": 2.2988, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 13.4, | |
| "eval_loss": 1.5357071161270142, | |
| "eval_runtime": 411.1707, | |
| "eval_samples_per_second": 25.264, | |
| "eval_steps_per_second": 3.159, | |
| "eval_wer": 0.9603381615701945, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 13.57, | |
| "learning_rate": 0.000616781898177247, | |
| "loss": 2.3041, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 13.74, | |
| "learning_rate": 0.0006104965430546826, | |
| "loss": 2.2905, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 13.9, | |
| "learning_rate": 0.0006043368950345695, | |
| "loss": 2.2946, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 14.07, | |
| "learning_rate": 0.000598051539912005, | |
| "loss": 2.3022, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 14.24, | |
| "learning_rate": 0.0005917661847894407, | |
| "loss": 2.2906, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 14.24, | |
| "eval_loss": 1.5637153387069702, | |
| "eval_runtime": 411.7098, | |
| "eval_samples_per_second": 25.231, | |
| "eval_steps_per_second": 3.155, | |
| "eval_wer": 0.9592097230700927, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 14.41, | |
| "learning_rate": 0.0005854808296668762, | |
| "loss": 2.2918, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 14.57, | |
| "learning_rate": 0.0005791954745443117, | |
| "loss": 2.2805, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 14.74, | |
| "learning_rate": 0.0005729101194217473, | |
| "loss": 2.2951, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 14.91, | |
| "learning_rate": 0.0005666247642991829, | |
| "loss": 2.2876, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 15.08, | |
| "learning_rate": 0.0005603394091766186, | |
| "loss": 2.2848, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 15.08, | |
| "eval_loss": 1.5325744152069092, | |
| "eval_runtime": 411.6076, | |
| "eval_samples_per_second": 25.238, | |
| "eval_steps_per_second": 3.156, | |
| "eval_wer": 0.9537340214958285, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 15.24, | |
| "learning_rate": 0.0005540540540540541, | |
| "loss": 2.2537, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 15.41, | |
| "learning_rate": 0.0005477686989314896, | |
| "loss": 2.2504, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 15.58, | |
| "learning_rate": 0.0005414833438089252, | |
| "loss": 2.2542, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 15.75, | |
| "learning_rate": 0.0005351979886863608, | |
| "loss": 2.2455, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 15.91, | |
| "learning_rate": 0.0005289126335637963, | |
| "loss": 2.2381, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 15.91, | |
| "eval_loss": 1.563069462776184, | |
| "eval_runtime": 410.1326, | |
| "eval_samples_per_second": 25.328, | |
| "eval_steps_per_second": 3.167, | |
| "eval_wer": 0.9508296797824518, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 16.08, | |
| "learning_rate": 0.0005226272784412319, | |
| "loss": 2.2406, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 16.25, | |
| "learning_rate": 0.0005163419233186674, | |
| "loss": 2.2265, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 16.42, | |
| "learning_rate": 0.0005100565681961032, | |
| "loss": 2.2221, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 16.58, | |
| "learning_rate": 0.0005037712130735387, | |
| "loss": 2.2122, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 16.75, | |
| "learning_rate": 0.0004974858579509742, | |
| "loss": 2.2072, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 16.75, | |
| "eval_loss": 1.4565062522888184, | |
| "eval_runtime": 409.2254, | |
| "eval_samples_per_second": 25.385, | |
| "eval_steps_per_second": 3.174, | |
| "eval_wer": 0.9395452947814344, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 16.92, | |
| "learning_rate": 0.0004912005028284098, | |
| "loss": 2.1876, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 17.09, | |
| "learning_rate": 0.00048491514770584537, | |
| "loss": 2.2144, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 17.25, | |
| "learning_rate": 0.00047862979258328096, | |
| "loss": 2.1943, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 17.42, | |
| "learning_rate": 0.00047234443746071655, | |
| "loss": 2.1901, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 17.59, | |
| "learning_rate": 0.00046605908233815214, | |
| "loss": 2.197, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 17.59, | |
| "eval_loss": 1.430406093597412, | |
| "eval_runtime": 410.1605, | |
| "eval_samples_per_second": 25.327, | |
| "eval_steps_per_second": 3.167, | |
| "eval_wer": 0.9405997373143163, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 17.76, | |
| "learning_rate": 0.0004597737272155877, | |
| "loss": 2.1872, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 17.92, | |
| "learning_rate": 0.0004534883720930232, | |
| "loss": 2.2033, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 18.09, | |
| "learning_rate": 0.00044720301697045886, | |
| "loss": 2.1865, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 18.26, | |
| "learning_rate": 0.0004409176618478944, | |
| "loss": 2.194, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 18.43, | |
| "learning_rate": 0.00043463230672533, | |
| "loss": 2.198, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 18.43, | |
| "eval_loss": 1.423040747642517, | |
| "eval_runtime": 411.9246, | |
| "eval_samples_per_second": 25.218, | |
| "eval_steps_per_second": 3.153, | |
| "eval_wer": 0.9382318663632832, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 18.59, | |
| "learning_rate": 0.0004283469516027655, | |
| "loss": 2.1784, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 18.76, | |
| "learning_rate": 0.00042206159648020117, | |
| "loss": 2.1739, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 18.93, | |
| "learning_rate": 0.0004157762413576367, | |
| "loss": 2.1686, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 19.1, | |
| "learning_rate": 0.0004094908862350723, | |
| "loss": 2.1639, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 19.26, | |
| "learning_rate": 0.00040320553111250783, | |
| "loss": 2.1668, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 19.26, | |
| "eval_loss": 1.3998422622680664, | |
| "eval_runtime": 412.8679, | |
| "eval_samples_per_second": 25.161, | |
| "eval_steps_per_second": 3.146, | |
| "eval_wer": 0.9314982333462827, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 19.43, | |
| "learning_rate": 0.00039692017598994347, | |
| "loss": 2.1694, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 19.6, | |
| "learning_rate": 0.000390634820867379, | |
| "loss": 2.1492, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 19.77, | |
| "learning_rate": 0.0003843494657448146, | |
| "loss": 2.1465, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 19.93, | |
| "learning_rate": 0.00037806411062225013, | |
| "loss": 2.1484, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 20.1, | |
| "learning_rate": 0.0003718416090509114, | |
| "loss": 2.1498, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 20.1, | |
| "eval_loss": 1.3919602632522583, | |
| "eval_runtime": 412.6773, | |
| "eval_samples_per_second": 25.172, | |
| "eval_steps_per_second": 3.148, | |
| "eval_wer": 0.9257635458867491, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 20.27, | |
| "learning_rate": 0.00036555625392834694, | |
| "loss": 2.1295, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 20.44, | |
| "learning_rate": 0.00035927089880578253, | |
| "loss": 2.1346, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 20.6, | |
| "learning_rate": 0.0003529855436832181, | |
| "loss": 2.1227, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 20.77, | |
| "learning_rate": 0.00034670018856065366, | |
| "loss": 2.1205, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 20.94, | |
| "learning_rate": 0.00034041483343808925, | |
| "loss": 2.1244, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 20.94, | |
| "eval_loss": 1.3584457635879517, | |
| "eval_runtime": 410.923, | |
| "eval_samples_per_second": 25.28, | |
| "eval_steps_per_second": 3.161, | |
| "eval_wer": 0.9152561185415394, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 21.11, | |
| "learning_rate": 0.00033412947831552484, | |
| "loss": 2.1163, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 21.27, | |
| "learning_rate": 0.00032784412319296043, | |
| "loss": 2.1141, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 21.44, | |
| "learning_rate": 0.00032155876807039597, | |
| "loss": 2.1122, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 21.61, | |
| "learning_rate": 0.00031527341294783156, | |
| "loss": 2.0937, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 21.78, | |
| "learning_rate": 0.00030898805782526715, | |
| "loss": 2.0953, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 21.78, | |
| "eval_loss": 1.327351450920105, | |
| "eval_runtime": 411.8656, | |
| "eval_samples_per_second": 25.222, | |
| "eval_steps_per_second": 3.154, | |
| "eval_wer": 0.905377656917698, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 21.94, | |
| "learning_rate": 0.00030270270270270274, | |
| "loss": 2.096, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 22.11, | |
| "learning_rate": 0.0002964173475801383, | |
| "loss": 2.1102, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 22.28, | |
| "learning_rate": 0.00029013199245757386, | |
| "loss": 2.0892, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 22.45, | |
| "learning_rate": 0.00028384663733500945, | |
| "loss": 2.0805, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 22.61, | |
| "learning_rate": 0.00027756128221244504, | |
| "loss": 2.0762, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 22.61, | |
| "eval_loss": 1.2932939529418945, | |
| "eval_runtime": 410.2802, | |
| "eval_samples_per_second": 25.319, | |
| "eval_steps_per_second": 3.166, | |
| "eval_wer": 0.9073015520654124, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 22.78, | |
| "learning_rate": 0.0002712759270898806, | |
| "loss": 2.0867, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 22.95, | |
| "learning_rate": 0.00026499057196731617, | |
| "loss": 2.0757, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 23.12, | |
| "learning_rate": 0.00025870521684475176, | |
| "loss": 2.0883, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 23.28, | |
| "learning_rate": 0.0002524198617221873, | |
| "loss": 2.0696, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 23.45, | |
| "learning_rate": 0.0002461345065996229, | |
| "loss": 2.0587, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 23.45, | |
| "eval_loss": 1.2515921592712402, | |
| "eval_runtime": 410.7551, | |
| "eval_samples_per_second": 25.29, | |
| "eval_steps_per_second": 3.162, | |
| "eval_wer": 0.8944447527609746, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 23.62, | |
| "learning_rate": 0.00023984915147705848, | |
| "loss": 2.0661, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 23.79, | |
| "learning_rate": 0.00023356379635449404, | |
| "loss": 2.0529, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 23.95, | |
| "learning_rate": 0.00022727844123192963, | |
| "loss": 2.0509, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 24.12, | |
| "learning_rate": 0.0002209930861093652, | |
| "loss": 2.0481, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 24.29, | |
| "learning_rate": 0.00021470773098680078, | |
| "loss": 2.0363, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 24.29, | |
| "eval_loss": 1.2214268445968628, | |
| "eval_runtime": 412.9112, | |
| "eval_samples_per_second": 25.158, | |
| "eval_steps_per_second": 3.146, | |
| "eval_wer": 0.8901529866622269, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 24.46, | |
| "learning_rate": 0.00020842237586423635, | |
| "loss": 2.0412, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 24.62, | |
| "learning_rate": 0.00020213702074167188, | |
| "loss": 2.0264, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 24.79, | |
| "learning_rate": 0.00019585166561910747, | |
| "loss": 2.0373, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 24.96, | |
| "learning_rate": 0.00018956631049654304, | |
| "loss": 2.0373, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 25.13, | |
| "learning_rate": 0.00018328095537397863, | |
| "loss": 2.0302, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 25.13, | |
| "eval_loss": 1.2087428569793701, | |
| "eval_runtime": 412.179, | |
| "eval_samples_per_second": 25.203, | |
| "eval_steps_per_second": 3.152, | |
| "eval_wer": 0.8871191520062157, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 25.29, | |
| "learning_rate": 0.0001769956002514142, | |
| "loss": 2.0109, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 25.46, | |
| "learning_rate": 0.00017071024512884978, | |
| "loss": 2.0215, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 25.63, | |
| "learning_rate": 0.00016442489000628534, | |
| "loss": 2.0137, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 25.8, | |
| "learning_rate": 0.00015813953488372093, | |
| "loss": 2.0084, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 25.96, | |
| "learning_rate": 0.0001518541797611565, | |
| "loss": 2.0071, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 25.96, | |
| "eval_loss": 1.1953096389770508, | |
| "eval_runtime": 413.1745, | |
| "eval_samples_per_second": 25.142, | |
| "eval_steps_per_second": 3.144, | |
| "eval_wer": 0.8785726177923303, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 26.13, | |
| "learning_rate": 0.00014556882463859208, | |
| "loss": 2.0112, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 26.3, | |
| "learning_rate": 0.00013928346951602765, | |
| "loss": 2.0077, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 26.47, | |
| "learning_rate": 0.00013299811439346324, | |
| "loss": 2.0052, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 26.63, | |
| "learning_rate": 0.0001267127592708988, | |
| "loss": 1.9947, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 26.8, | |
| "learning_rate": 0.00012042740414833438, | |
| "loss": 1.9882, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 26.8, | |
| "eval_loss": 1.1737616062164307, | |
| "eval_runtime": 414.4584, | |
| "eval_samples_per_second": 25.064, | |
| "eval_steps_per_second": 3.134, | |
| "eval_wer": 0.8711730210703517, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 26.97, | |
| "learning_rate": 0.0001142049025769956, | |
| "loss": 1.9926, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 27.14, | |
| "learning_rate": 0.00010791954745443117, | |
| "loss": 2.0032, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 27.3, | |
| "learning_rate": 0.00010163419233186675, | |
| "loss": 1.996, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 27.47, | |
| "learning_rate": 9.534883720930233e-05, | |
| "loss": 1.9746, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 27.64, | |
| "learning_rate": 8.906348208673789e-05, | |
| "loss": 1.9772, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 27.64, | |
| "eval_loss": 1.164720892906189, | |
| "eval_runtime": 425.7197, | |
| "eval_samples_per_second": 24.401, | |
| "eval_steps_per_second": 3.051, | |
| "eval_wer": 0.867214236824093, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 27.81, | |
| "learning_rate": 8.277812696417347e-05, | |
| "loss": 1.9759, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 27.97, | |
| "learning_rate": 7.649277184160904e-05, | |
| "loss": 1.9657, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 28.14, | |
| "learning_rate": 7.020741671904462e-05, | |
| "loss": 1.9806, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 28.31, | |
| "learning_rate": 6.39220615964802e-05, | |
| "loss": 1.9802, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 28.48, | |
| "learning_rate": 5.763670647391578e-05, | |
| "loss": 1.9585, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 28.48, | |
| "eval_loss": 1.1459153890609741, | |
| "eval_runtime": 417.3472, | |
| "eval_samples_per_second": 24.891, | |
| "eval_steps_per_second": 3.113, | |
| "eval_wer": 0.8634774404794938, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 28.64, | |
| "learning_rate": 5.135135135135136e-05, | |
| "loss": 1.9573, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 28.81, | |
| "learning_rate": 4.506599622878693e-05, | |
| "loss": 1.962, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 28.98, | |
| "learning_rate": 3.8780641106222504e-05, | |
| "loss": 1.9653, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 29.15, | |
| "learning_rate": 3.249528598365808e-05, | |
| "loss": 1.9631, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 29.31, | |
| "learning_rate": 2.620993086109365e-05, | |
| "loss": 1.944, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 29.31, | |
| "eval_loss": 1.1414194107055664, | |
| "eval_runtime": 414.6007, | |
| "eval_samples_per_second": 25.055, | |
| "eval_steps_per_second": 3.133, | |
| "eval_wer": 0.8616275412989992, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 29.48, | |
| "learning_rate": 1.9924575738529227e-05, | |
| "loss": 1.9547, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 29.65, | |
| "learning_rate": 1.3639220615964803e-05, | |
| "loss": 1.9557, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 29.82, | |
| "learning_rate": 7.353865493400377e-06, | |
| "loss": 1.9464, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 29.98, | |
| "learning_rate": 1.0685103708359522e-06, | |
| "loss": 1.9654, | |
| "step": 17900 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "step": 17910, | |
| "total_flos": 3.015307099908152e+20, | |
| "train_loss": 1.4937853462266097, | |
| "train_runtime": 52137.1608, | |
| "train_samples_per_second": 21.986, | |
| "train_steps_per_second": 0.344 | |
| } | |
| ], | |
| "max_steps": 17910, | |
| "num_train_epochs": 30, | |
| "total_flos": 3.015307099908152e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |