| { | |
| "best_metric": 0.24944954621126686, | |
| "best_model_checkpoint": "/workspace/output_models/hi-mr/wav2vec2-large-xlsr-hindi-marathi-3/checkpoint-10000", | |
| "epoch": 16.0, | |
| "global_step": 10000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 6.666666666666666e-05, | |
| "loss": 16.623, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.0001333333333333333, | |
| "loss": 3.8395, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "eval_loss": 3.426492214202881, | |
| "eval_runtime": 80.6897, | |
| "eval_samples_per_second": 24.786, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00019999999999999998, | |
| "loss": 3.4548, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.0002666666666666666, | |
| "loss": 3.3471, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "eval_loss": 2.9866673946380615, | |
| "eval_runtime": 81.4064, | |
| "eval_samples_per_second": 24.568, | |
| "eval_wer": 1.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0002991803278688524, | |
| "loss": 1.9911, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 0.00029754098360655737, | |
| "loss": 0.9874, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_loss": 0.6858933568000793, | |
| "eval_runtime": 82.4721, | |
| "eval_samples_per_second": 24.251, | |
| "eval_wer": 0.5436335320337253, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 0.00029590163934426226, | |
| "loss": 0.7698, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 0.00029426229508196716, | |
| "loss": 0.6471, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "eval_loss": 0.5174924731254578, | |
| "eval_runtime": 82.661, | |
| "eval_samples_per_second": 24.195, | |
| "eval_wer": 0.43928897481338275, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 0.0002926229508196721, | |
| "loss": 0.568, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.000290983606557377, | |
| "loss": 0.5393, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 0.4453641474246979, | |
| "eval_runtime": 83.0998, | |
| "eval_samples_per_second": 24.067, | |
| "eval_wer": 0.39015090489232584, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 0.00028934426229508195, | |
| "loss": 0.5009, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 0.00028770491803278684, | |
| "loss": 0.4932, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "eval_loss": 0.41159647703170776, | |
| "eval_runtime": 83.9345, | |
| "eval_samples_per_second": 23.828, | |
| "eval_wer": 0.37189194994898234, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 0.00028606557377049174, | |
| "loss": 0.4407, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 0.00028442622950819674, | |
| "loss": 0.3891, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "eval_loss": 0.4060937166213989, | |
| "eval_runtime": 82.5906, | |
| "eval_samples_per_second": 24.216, | |
| "eval_wer": 0.3500886096342839, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 0.00028278688524590163, | |
| "loss": 0.393, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 0.00028114754098360653, | |
| "loss": 0.3723, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "eval_loss": 0.37875330448150635, | |
| "eval_runtime": 83.1067, | |
| "eval_samples_per_second": 24.065, | |
| "eval_wer": 0.3370388271306589, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 0.0002795081967213115, | |
| "loss": 0.3732, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 0.00027786885245901637, | |
| "loss": 0.3536, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "eval_loss": 0.3747707009315491, | |
| "eval_runtime": 83.4226, | |
| "eval_samples_per_second": 23.974, | |
| "eval_wer": 0.3268889962945062, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 0.0002762295081967213, | |
| "loss": 0.3394, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 0.0002745901639344262, | |
| "loss": 0.2952, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "eval_loss": 0.3621469736099243, | |
| "eval_runtime": 82.21, | |
| "eval_samples_per_second": 24.328, | |
| "eval_wer": 0.31018742280221256, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "learning_rate": 0.0002729508196721311, | |
| "loss": 0.3014, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "learning_rate": 0.00027131147540983606, | |
| "loss": 0.2984, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "eval_loss": 0.35865798592567444, | |
| "eval_runtime": 83.2434, | |
| "eval_samples_per_second": 24.026, | |
| "eval_wer": 0.2970302346812738, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 0.00026967213114754095, | |
| "loss": 0.2826, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "learning_rate": 0.0002680327868852459, | |
| "loss": 0.2821, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "eval_loss": 0.3647235035896301, | |
| "eval_runtime": 83.2284, | |
| "eval_samples_per_second": 24.03, | |
| "eval_wer": 0.30664303743085763, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 0.0002663934426229508, | |
| "loss": 0.27, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "learning_rate": 0.0002647540983606557, | |
| "loss": 0.2433, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "eval_loss": 0.3836052119731903, | |
| "eval_runtime": 82.8815, | |
| "eval_samples_per_second": 24.131, | |
| "eval_wer": 0.3012190537565115, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "learning_rate": 0.00026311475409836063, | |
| "loss": 0.2373, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "learning_rate": 0.00026147540983606553, | |
| "loss": 0.2304, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "eval_loss": 0.3565887212753296, | |
| "eval_runtime": 83.2398, | |
| "eval_samples_per_second": 24.027, | |
| "eval_wer": 0.2935395521185758, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 4.64, | |
| "learning_rate": 0.0002598360655737705, | |
| "loss": 0.2262, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 0.00025819672131147537, | |
| "loss": 0.2249, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "eval_loss": 0.371224969625473, | |
| "eval_runtime": 83.7191, | |
| "eval_samples_per_second": 23.889, | |
| "eval_wer": 0.285913753289297, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "learning_rate": 0.00025655737704918027, | |
| "loss": 0.2232, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "learning_rate": 0.0002549180327868852, | |
| "loss": 0.2055, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "eval_loss": 0.36438417434692383, | |
| "eval_runtime": 82.4039, | |
| "eval_samples_per_second": 24.271, | |
| "eval_wer": 0.28048976961495087, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "learning_rate": 0.00025327868852459016, | |
| "loss": 0.1999, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 5.44, | |
| "learning_rate": 0.00025163934426229506, | |
| "loss": 0.1969, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 5.44, | |
| "eval_loss": 0.371999055147171, | |
| "eval_runtime": 83.0961, | |
| "eval_samples_per_second": 24.069, | |
| "eval_wer": 0.28204715106600076, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "learning_rate": 0.00025, | |
| "loss": 0.1949, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "learning_rate": 0.0002483606557377049, | |
| "loss": 0.1945, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "eval_loss": 0.3662562370300293, | |
| "eval_runtime": 82.7035, | |
| "eval_samples_per_second": 24.183, | |
| "eval_wer": 0.28435637183824714, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 5.92, | |
| "learning_rate": 0.00024672131147540985, | |
| "loss": 0.191, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "learning_rate": 0.00024508196721311474, | |
| "loss": 0.1689, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "eval_loss": 0.3731316328048706, | |
| "eval_runtime": 83.1862, | |
| "eval_samples_per_second": 24.042, | |
| "eval_wer": 0.2787712797379303, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "learning_rate": 0.00024344262295081966, | |
| "loss": 0.1631, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 0.00024180327868852458, | |
| "loss": 0.1568, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "eval_loss": 0.37625598907470703, | |
| "eval_runtime": 83.6104, | |
| "eval_samples_per_second": 23.92, | |
| "eval_wer": 0.2781805488427045, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "learning_rate": 0.00024016393442622948, | |
| "loss": 0.1665, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "learning_rate": 0.0002385245901639344, | |
| "loss": 0.1581, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "eval_loss": 0.3642520606517792, | |
| "eval_runtime": 83.3706, | |
| "eval_samples_per_second": 23.989, | |
| "eval_wer": 0.2725954567423876, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 6.88, | |
| "learning_rate": 0.00023688524590163932, | |
| "loss": 0.1517, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 0.00023524590163934424, | |
| "loss": 0.1529, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "eval_loss": 0.3819567859172821, | |
| "eval_runtime": 83.7194, | |
| "eval_samples_per_second": 23.889, | |
| "eval_wer": 0.27238064550776003, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "learning_rate": 0.00023360655737704916, | |
| "loss": 0.1344, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 7.36, | |
| "learning_rate": 0.00023196721311475406, | |
| "loss": 0.1363, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 7.36, | |
| "eval_loss": 0.38544762134552, | |
| "eval_runtime": 83.3319, | |
| "eval_samples_per_second": 24.0, | |
| "eval_wer": 0.2733472960635841, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 7.52, | |
| "learning_rate": 0.00023032786885245898, | |
| "loss": 0.1276, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 7.68, | |
| "learning_rate": 0.0002286885245901639, | |
| "loss": 0.1252, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 7.68, | |
| "eval_loss": 0.37892764806747437, | |
| "eval_runtime": 83.6054, | |
| "eval_samples_per_second": 23.922, | |
| "eval_wer": 0.2722732398904463, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 7.84, | |
| "learning_rate": 0.00022704918032786882, | |
| "loss": 0.1328, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 0.00022540983606557374, | |
| "loss": 0.1327, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.3974834084510803, | |
| "eval_runtime": 83.8203, | |
| "eval_samples_per_second": 23.861, | |
| "eval_wer": 0.2739380269588099, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 8.16, | |
| "learning_rate": 0.0002237704918032787, | |
| "loss": 0.1176, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 8.32, | |
| "learning_rate": 0.0002221311475409836, | |
| "loss": 0.1157, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 8.32, | |
| "eval_loss": 0.3960728347301483, | |
| "eval_runtime": 84.4842, | |
| "eval_samples_per_second": 23.673, | |
| "eval_wer": 0.2711991837173084, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 8.48, | |
| "learning_rate": 0.0002204918032786885, | |
| "loss": 0.1151, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 8.64, | |
| "learning_rate": 0.00021885245901639343, | |
| "loss": 0.1173, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 8.64, | |
| "eval_loss": 0.40709516406059265, | |
| "eval_runtime": 82.9277, | |
| "eval_samples_per_second": 24.117, | |
| "eval_wer": 0.2727565651683583, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 8.8, | |
| "learning_rate": 0.00021721311475409835, | |
| "loss": 0.1143, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 8.96, | |
| "learning_rate": 0.00021557377049180327, | |
| "loss": 0.1166, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 8.96, | |
| "eval_loss": 0.41475972533226013, | |
| "eval_runtime": 84.4573, | |
| "eval_samples_per_second": 23.681, | |
| "eval_wer": 0.2736695129155255, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 9.12, | |
| "learning_rate": 0.0002139344262295082, | |
| "loss": 0.1028, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 9.28, | |
| "learning_rate": 0.00021229508196721309, | |
| "loss": 0.1026, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 9.28, | |
| "eval_loss": 0.4114053547382355, | |
| "eval_runtime": 83.5505, | |
| "eval_samples_per_second": 23.938, | |
| "eval_wer": 0.2681381236238655, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 9.44, | |
| "learning_rate": 0.000210655737704918, | |
| "loss": 0.1004, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "learning_rate": 0.00020901639344262293, | |
| "loss": 0.1121, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "eval_loss": 0.4229777753353119, | |
| "eval_runtime": 83.5437, | |
| "eval_samples_per_second": 23.94, | |
| "eval_wer": 0.27060845282208257, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 9.76, | |
| "learning_rate": 0.00020737704918032785, | |
| "loss": 0.1026, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 9.92, | |
| "learning_rate": 0.00020573770491803277, | |
| "loss": 0.1041, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 9.92, | |
| "eval_loss": 0.40711304545402527, | |
| "eval_runtime": 83.5856, | |
| "eval_samples_per_second": 23.928, | |
| "eval_wer": 0.27017883035282747, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 10.08, | |
| "learning_rate": 0.0002040983606557377, | |
| "loss": 0.0877, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 10.24, | |
| "learning_rate": 0.0002024590163934426, | |
| "loss": 0.088, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 10.24, | |
| "eval_loss": 0.437641978263855, | |
| "eval_runtime": 83.2287, | |
| "eval_samples_per_second": 24.03, | |
| "eval_wer": 0.26797701519789485, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 10.4, | |
| "learning_rate": 0.0002008196721311475, | |
| "loss": 0.0941, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 10.56, | |
| "learning_rate": 0.00019918032786885243, | |
| "loss": 0.0884, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 10.56, | |
| "eval_loss": 0.4451253414154053, | |
| "eval_runtime": 84.0696, | |
| "eval_samples_per_second": 23.79, | |
| "eval_wer": 0.26529187476505023, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 10.72, | |
| "learning_rate": 0.00019754098360655735, | |
| "loss": 0.0867, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 10.88, | |
| "learning_rate": 0.00019590163934426227, | |
| "loss": 0.0978, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 10.88, | |
| "eval_loss": 0.4322107136249542, | |
| "eval_runtime": 83.5762, | |
| "eval_samples_per_second": 23.93, | |
| "eval_wer": 0.26475484667848126, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 11.04, | |
| "learning_rate": 0.00019426229508196722, | |
| "loss": 0.0834, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 11.2, | |
| "learning_rate": 0.00019262295081967214, | |
| "loss": 0.0785, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 11.2, | |
| "eval_loss": 0.4244215190410614, | |
| "eval_runtime": 83.288, | |
| "eval_samples_per_second": 24.013, | |
| "eval_wer": 0.25970678266473335, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 11.36, | |
| "learning_rate": 0.00019098360655737704, | |
| "loss": 0.0723, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 11.52, | |
| "learning_rate": 0.00018934426229508196, | |
| "loss": 0.0771, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 11.52, | |
| "eval_loss": 0.43486374616622925, | |
| "eval_runtime": 83.8141, | |
| "eval_samples_per_second": 23.862, | |
| "eval_wer": 0.25755867031845764, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 11.68, | |
| "learning_rate": 0.00018770491803278688, | |
| "loss": 0.0787, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 11.84, | |
| "learning_rate": 0.0001860655737704918, | |
| "loss": 0.0809, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 11.84, | |
| "eval_loss": 0.44311779737472534, | |
| "eval_runtime": 85.6363, | |
| "eval_samples_per_second": 23.355, | |
| "eval_wer": 0.2628215455668331, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 0.00018442622950819672, | |
| "loss": 0.0778, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 12.16, | |
| "learning_rate": 0.00018278688524590162, | |
| "loss": 0.071, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 12.16, | |
| "eval_loss": 0.47199195623397827, | |
| "eval_runtime": 82.9761, | |
| "eval_samples_per_second": 24.103, | |
| "eval_wer": 0.2588475377262231, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 12.32, | |
| "learning_rate": 0.00018114754098360654, | |
| "loss": 0.074, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 12.48, | |
| "learning_rate": 0.00017950819672131146, | |
| "loss": 0.0677, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 12.48, | |
| "eval_loss": 0.4387381672859192, | |
| "eval_runtime": 83.0897, | |
| "eval_samples_per_second": 24.07, | |
| "eval_wer": 0.259867891090704, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 12.64, | |
| "learning_rate": 0.00017786885245901638, | |
| "loss": 0.0703, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 12.8, | |
| "learning_rate": 0.0001762295081967213, | |
| "loss": 0.0765, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 12.8, | |
| "eval_loss": 0.4592679738998413, | |
| "eval_runtime": 84.0398, | |
| "eval_samples_per_second": 23.798, | |
| "eval_wer": 0.2572364534665163, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 12.96, | |
| "learning_rate": 0.0001745901639344262, | |
| "loss": 0.0683, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 13.12, | |
| "learning_rate": 0.00017295081967213112, | |
| "loss": 0.0621, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 13.12, | |
| "eval_loss": 0.4588267505168915, | |
| "eval_runtime": 82.8224, | |
| "eval_samples_per_second": 24.148, | |
| "eval_wer": 0.2633585736534021, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 13.28, | |
| "learning_rate": 0.00017131147540983604, | |
| "loss": 0.065, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 13.44, | |
| "learning_rate": 0.00016967213114754096, | |
| "loss": 0.067, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 13.44, | |
| "eval_loss": 0.4433707296848297, | |
| "eval_runtime": 84.8017, | |
| "eval_samples_per_second": 23.584, | |
| "eval_wer": 0.25922345738682134, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 13.6, | |
| "learning_rate": 0.00016803278688524588, | |
| "loss": 0.067, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 13.76, | |
| "learning_rate": 0.0001663934426229508, | |
| "loss": 0.0633, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 13.76, | |
| "eval_loss": 0.4525003433227539, | |
| "eval_runtime": 82.3869, | |
| "eval_samples_per_second": 24.276, | |
| "eval_wer": 0.2647011438698244, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 13.92, | |
| "learning_rate": 0.00016475409836065575, | |
| "loss": 0.0704, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 14.08, | |
| "learning_rate": 0.00016311475409836064, | |
| "loss": 0.0635, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 14.08, | |
| "eval_loss": 0.46746838092803955, | |
| "eval_runtime": 82.2147, | |
| "eval_samples_per_second": 24.327, | |
| "eval_wer": 0.2580419955963697, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 14.24, | |
| "learning_rate": 0.00016147540983606556, | |
| "loss": 0.0568, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 14.4, | |
| "learning_rate": 0.00015983606557377049, | |
| "loss": 0.06, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 14.4, | |
| "eval_loss": 0.46421578526496887, | |
| "eval_runtime": 84.6457, | |
| "eval_samples_per_second": 23.628, | |
| "eval_wer": 0.25841791525696794, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 14.56, | |
| "learning_rate": 0.0001581967213114754, | |
| "loss": 0.0588, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 14.72, | |
| "learning_rate": 0.00015655737704918033, | |
| "loss": 0.0618, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 14.72, | |
| "eval_loss": 0.465658038854599, | |
| "eval_runtime": 83.6497, | |
| "eval_samples_per_second": 23.909, | |
| "eval_wer": 0.2606734332205574, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 14.88, | |
| "learning_rate": 0.00015491803278688525, | |
| "loss": 0.0578, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 15.04, | |
| "learning_rate": 0.00015327868852459014, | |
| "loss": 0.0562, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 15.04, | |
| "eval_loss": 0.4753575325012207, | |
| "eval_runtime": 82.0241, | |
| "eval_samples_per_second": 24.383, | |
| "eval_wer": 0.25831050963965413, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 15.2, | |
| "learning_rate": 0.00015163934426229507, | |
| "loss": 0.0489, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 15.36, | |
| "learning_rate": 0.00015, | |
| "loss": 0.0528, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 15.36, | |
| "eval_loss": 0.48737889528274536, | |
| "eval_runtime": 85.0733, | |
| "eval_samples_per_second": 23.509, | |
| "eval_wer": 0.2623382202889211, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 15.52, | |
| "learning_rate": 0.0001483606557377049, | |
| "loss": 0.0522, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 15.68, | |
| "learning_rate": 0.00014672131147540983, | |
| "loss": 0.0528, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 15.68, | |
| "eval_loss": 0.47928667068481445, | |
| "eval_runtime": 84.0501, | |
| "eval_samples_per_second": 23.795, | |
| "eval_wer": 0.26206970624563664, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 15.84, | |
| "learning_rate": 0.00014508196721311472, | |
| "loss": 0.0503, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 0.00014344262295081964, | |
| "loss": 0.0507, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.46980950236320496, | |
| "eval_runtime": 82.8709, | |
| "eval_samples_per_second": 24.134, | |
| "eval_wer": 0.24944954621126686, | |
| "step": 10000 | |
| } | |
| ], | |
| "max_steps": 18750, | |
| "num_train_epochs": 30, | |
| "total_flos": 7.756303740936756e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |