| { | |
| "best_metric": 15.749067778021498, | |
| "best_model_checkpoint": "./checkpoint-4000", | |
| "epoch": 8.0048, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 4.6e-09, | |
| "loss": 0.9435, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 9.6e-09, | |
| "loss": 0.9281, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 1.4599999999999998e-08, | |
| "loss": 0.9519, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.94e-08, | |
| "loss": 0.9436, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 2.4399999999999997e-08, | |
| "loss": 0.9253, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 2.94e-08, | |
| "loss": 0.9338, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 3.44e-08, | |
| "loss": 0.9108, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 3.94e-08, | |
| "loss": 0.8899, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 4.44e-08, | |
| "loss": 0.8772, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 4.92e-08, | |
| "loss": 0.8194, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 5.42e-08, | |
| "loss": 0.7858, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 5.9199999999999994e-08, | |
| "loss": 0.772, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 6.419999999999999e-08, | |
| "loss": 0.7039, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 6.92e-08, | |
| "loss": 0.6403, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 7.419999999999999e-08, | |
| "loss": 0.5761, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 7.92e-08, | |
| "loss": 0.4753, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 8.42e-08, | |
| "loss": 0.4406, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 8.919999999999999e-08, | |
| "loss": 0.4276, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.42e-08, | |
| "loss": 0.3747, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.919999999999999e-08, | |
| "loss": 0.3983, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_cer": 10.639092728485657, | |
| "eval_loss": 0.5338271260261536, | |
| "eval_runtime": 1141.2363, | |
| "eval_samples_per_second": 1.084, | |
| "eval_steps_per_second": 0.272, | |
| "eval_wer": 19.587628865979383, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.953333333333332e-08, | |
| "loss": 0.3437, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 9.897777777777777e-08, | |
| "loss": 0.3388, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 9.842222222222222e-08, | |
| "loss": 0.32, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 9.786666666666667e-08, | |
| "loss": 0.2985, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 9.73111111111111e-08, | |
| "loss": 0.3006, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 9.675555555555554e-08, | |
| "loss": 0.2939, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 9.619999999999999e-08, | |
| "loss": 0.2768, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 9.564444444444444e-08, | |
| "loss": 0.2703, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 9.511111111111111e-08, | |
| "loss": 0.2733, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 9.455555555555556e-08, | |
| "loss": 0.2623, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 9.4e-08, | |
| "loss": 0.2679, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 9.344444444444444e-08, | |
| "loss": 0.2431, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 9.288888888888888e-08, | |
| "loss": 0.2415, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 9.233333333333333e-08, | |
| "loss": 0.2584, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 9.177777777777777e-08, | |
| "loss": 0.229, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 9.122222222222222e-08, | |
| "loss": 0.2333, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 9.066666666666666e-08, | |
| "loss": 0.2333, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 9.01111111111111e-08, | |
| "loss": 0.2233, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 8.955555555555555e-08, | |
| "loss": 0.2306, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 8.9e-08, | |
| "loss": 0.2277, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "eval_cer": 8.266844563042028, | |
| "eval_loss": 0.4134095311164856, | |
| "eval_runtime": 1163.4198, | |
| "eval_samples_per_second": 1.063, | |
| "eval_steps_per_second": 0.266, | |
| "eval_wer": 16.582583899978065, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 8.844444444444444e-08, | |
| "loss": 0.213, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 8.788888888888889e-08, | |
| "loss": 0.2195, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 8.733333333333332e-08, | |
| "loss": 0.2242, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 8.677777777777777e-08, | |
| "loss": 0.2075, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 8.622222222222222e-08, | |
| "loss": 0.2077, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 8.566666666666667e-08, | |
| "loss": 0.2004, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 8.511111111111111e-08, | |
| "loss": 0.209, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 8.457777777777778e-08, | |
| "loss": 0.2083, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 8.402222222222221e-08, | |
| "loss": 0.1933, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 8.346666666666666e-08, | |
| "loss": 0.2018, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 8.291111111111111e-08, | |
| "loss": 0.1949, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 8.235555555555556e-08, | |
| "loss": 0.2019, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 8.179999999999999e-08, | |
| "loss": 0.1877, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 8.124444444444444e-08, | |
| "loss": 0.192, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 8.068888888888889e-08, | |
| "loss": 0.204, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 8.013333333333334e-08, | |
| "loss": 0.1995, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 7.957777777777777e-08, | |
| "loss": 0.1851, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 7.902222222222222e-08, | |
| "loss": 0.187, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 7.846666666666665e-08, | |
| "loss": 0.1929, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 7.79111111111111e-08, | |
| "loss": 0.172, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "eval_cer": 7.978652434956637, | |
| "eval_loss": 0.3968185782432556, | |
| "eval_runtime": 1160.0567, | |
| "eval_samples_per_second": 1.066, | |
| "eval_steps_per_second": 0.267, | |
| "eval_wer": 16.30840096512393, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 7.735555555555555e-08, | |
| "loss": 0.1847, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 7.68e-08, | |
| "loss": 0.1803, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 7.624444444444445e-08, | |
| "loss": 0.1808, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 7.56888888888889e-08, | |
| "loss": 0.1782, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 7.513333333333333e-08, | |
| "loss": 0.1894, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 7.457777777777777e-08, | |
| "loss": 0.1754, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 7.402222222222221e-08, | |
| "loss": 0.1811, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 7.346666666666666e-08, | |
| "loss": 0.1733, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 7.291111111111111e-08, | |
| "loss": 0.1717, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 7.235555555555555e-08, | |
| "loss": 0.1814, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 7.18e-08, | |
| "loss": 0.1753, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 7.124444444444444e-08, | |
| "loss": 0.1671, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 7.068888888888889e-08, | |
| "loss": 0.1668, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 7.013333333333334e-08, | |
| "loss": 0.163, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 6.957777777777777e-08, | |
| "loss": 0.1589, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 6.902222222222221e-08, | |
| "loss": 0.1689, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 6.846666666666666e-08, | |
| "loss": 0.1641, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 6.79111111111111e-08, | |
| "loss": 0.1598, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 6.735555555555555e-08, | |
| "loss": 0.1692, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 6.68e-08, | |
| "loss": 0.1823, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "eval_cer": 7.815877251501001, | |
| "eval_loss": 0.39564967155456543, | |
| "eval_runtime": 1167.325, | |
| "eval_samples_per_second": 1.06, | |
| "eval_steps_per_second": 0.266, | |
| "eval_wer": 16.176793156393945, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 6.624444444444444e-08, | |
| "loss": 0.1666, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 6.568888888888889e-08, | |
| "loss": 0.1582, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 6.513333333333333e-08, | |
| "loss": 0.1562, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 6.457777777777777e-08, | |
| "loss": 0.1703, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 6.402222222222222e-08, | |
| "loss": 0.1494, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 6.346666666666667e-08, | |
| "loss": 0.1726, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 6.29111111111111e-08, | |
| "loss": 0.1482, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 6.235555555555555e-08, | |
| "loss": 0.1645, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 6.18e-08, | |
| "loss": 0.1549, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 6.124444444444445e-08, | |
| "loss": 0.1496, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 6.06888888888889e-08, | |
| "loss": 0.1558, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "learning_rate": 6.013333333333334e-08, | |
| "loss": 0.1611, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "learning_rate": 5.957777777777777e-08, | |
| "loss": 0.1535, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 5.902222222222222e-08, | |
| "loss": 0.1482, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 5.846666666666667e-08, | |
| "loss": 0.156, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "learning_rate": 5.791111111111111e-08, | |
| "loss": 0.1583, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "learning_rate": 5.735555555555556e-08, | |
| "loss": 0.1505, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 5.679999999999999e-08, | |
| "loss": 0.1421, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 5.624444444444444e-08, | |
| "loss": 0.1498, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 5.568888888888888e-08, | |
| "loss": 0.1445, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_cer": 7.743829219479653, | |
| "eval_loss": 0.39546218514442444, | |
| "eval_runtime": 1163.0632, | |
| "eval_samples_per_second": 1.064, | |
| "eval_steps_per_second": 0.267, | |
| "eval_wer": 16.034218030269795, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 5.513333333333333e-08, | |
| "loss": 0.1497, | |
| "step": 2525 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 5.457777777777778e-08, | |
| "loss": 0.1455, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 5.402222222222223e-08, | |
| "loss": 0.1453, | |
| "step": 2575 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 5.346666666666666e-08, | |
| "loss": 0.1441, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 5.2911111111111104e-08, | |
| "loss": 0.1611, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 5.235555555555555e-08, | |
| "loss": 0.1483, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 5.18e-08, | |
| "loss": 0.1444, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 5.124444444444445e-08, | |
| "loss": 0.1414, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 5.068888888888889e-08, | |
| "loss": 0.148, | |
| "step": 2725 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 5.0133333333333326e-08, | |
| "loss": 0.1401, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 4.9577777777777774e-08, | |
| "loss": 0.1486, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 4.902222222222222e-08, | |
| "loss": 0.1329, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 4.8466666666666664e-08, | |
| "loss": 0.1514, | |
| "step": 2825 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 4.7911111111111106e-08, | |
| "loss": 0.1392, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "learning_rate": 4.7355555555555554e-08, | |
| "loss": 0.1362, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "learning_rate": 4.68e-08, | |
| "loss": 0.1417, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "learning_rate": 4.6244444444444444e-08, | |
| "loss": 0.1514, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "learning_rate": 4.5688888888888886e-08, | |
| "loss": 0.1461, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "learning_rate": 4.513333333333333e-08, | |
| "loss": 0.1355, | |
| "step": 2975 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "learning_rate": 4.4577777777777775e-08, | |
| "loss": 0.147, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "eval_cer": 7.71447631754503, | |
| "eval_loss": 0.3965264558792114, | |
| "eval_runtime": 1162.9158, | |
| "eval_samples_per_second": 1.064, | |
| "eval_steps_per_second": 0.267, | |
| "eval_wer": 15.88067558675148, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "learning_rate": 4.4022222222222224e-08, | |
| "loss": 0.135, | |
| "step": 3025 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "learning_rate": 4.3466666666666665e-08, | |
| "loss": 0.133, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "learning_rate": 4.291111111111111e-08, | |
| "loss": 0.141, | |
| "step": 3075 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "learning_rate": 4.2355555555555555e-08, | |
| "loss": 0.1312, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.18e-08, | |
| "loss": 0.1332, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 4.1244444444444445e-08, | |
| "loss": 0.1314, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 4.068888888888889e-08, | |
| "loss": 0.139, | |
| "step": 3175 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "learning_rate": 4.013333333333333e-08, | |
| "loss": 0.1368, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "learning_rate": 3.957777777777778e-08, | |
| "loss": 0.1367, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 3.9022222222222225e-08, | |
| "loss": 0.1448, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 3.846666666666667e-08, | |
| "loss": 0.1383, | |
| "step": 3275 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "learning_rate": 3.791111111111111e-08, | |
| "loss": 0.1374, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "learning_rate": 3.735555555555555e-08, | |
| "loss": 0.1333, | |
| "step": 3325 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "learning_rate": 3.68e-08, | |
| "loss": 0.1325, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "learning_rate": 3.624444444444445e-08, | |
| "loss": 0.1315, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 3.571111111111111e-08, | |
| "loss": 0.1311, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 3.515555555555556e-08, | |
| "loss": 0.1401, | |
| "step": 3425 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "learning_rate": 3.46e-08, | |
| "loss": 0.1297, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "learning_rate": 3.404444444444444e-08, | |
| "loss": 0.1339, | |
| "step": 3475 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "learning_rate": 3.348888888888889e-08, | |
| "loss": 0.1292, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "eval_cer": 7.706470980653769, | |
| "eval_loss": 0.3999946713447571, | |
| "eval_runtime": 1162.3373, | |
| "eval_samples_per_second": 1.064, | |
| "eval_steps_per_second": 0.267, | |
| "eval_wer": 15.85874095196315, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "learning_rate": 3.293333333333333e-08, | |
| "loss": 0.132, | |
| "step": 3525 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "learning_rate": 3.237777777777778e-08, | |
| "loss": 0.1399, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "learning_rate": 3.182222222222222e-08, | |
| "loss": 0.1248, | |
| "step": 3575 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "learning_rate": 3.126666666666666e-08, | |
| "loss": 0.1364, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "learning_rate": 3.071111111111111e-08, | |
| "loss": 0.1318, | |
| "step": 3625 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "learning_rate": 3.015555555555556e-08, | |
| "loss": 0.1319, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "learning_rate": 2.9599999999999997e-08, | |
| "loss": 0.1284, | |
| "step": 3675 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "learning_rate": 2.9044444444444442e-08, | |
| "loss": 0.1231, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "learning_rate": 2.848888888888889e-08, | |
| "loss": 0.1305, | |
| "step": 3725 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 2.7933333333333332e-08, | |
| "loss": 0.1231, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "learning_rate": 2.7377777777777777e-08, | |
| "loss": 0.1236, | |
| "step": 3775 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "learning_rate": 2.682222222222222e-08, | |
| "loss": 0.1296, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 2.6266666666666664e-08, | |
| "loss": 0.129, | |
| "step": 3825 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 2.5711111111111112e-08, | |
| "loss": 0.1317, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "learning_rate": 2.5155555555555554e-08, | |
| "loss": 0.1415, | |
| "step": 3875 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "learning_rate": 2.46e-08, | |
| "loss": 0.1189, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 2.4044444444444444e-08, | |
| "loss": 0.1356, | |
| "step": 3925 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 2.348888888888889e-08, | |
| "loss": 0.1232, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "learning_rate": 2.2933333333333334e-08, | |
| "loss": 0.1315, | |
| "step": 3975 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "learning_rate": 2.2377777777777775e-08, | |
| "loss": 0.1187, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "eval_cer": 7.639759839893262, | |
| "eval_loss": 0.40291938185691833, | |
| "eval_runtime": 1161.4251, | |
| "eval_samples_per_second": 1.065, | |
| "eval_steps_per_second": 0.267, | |
| "eval_wer": 15.749067778021498, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "learning_rate": 2.1822222222222224e-08, | |
| "loss": 0.127, | |
| "step": 4025 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "learning_rate": 2.1266666666666665e-08, | |
| "loss": 0.1269, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "learning_rate": 2.071111111111111e-08, | |
| "loss": 0.1273, | |
| "step": 4075 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "learning_rate": 2.0155555555555552e-08, | |
| "loss": 0.1253, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "learning_rate": 1.96e-08, | |
| "loss": 0.1218, | |
| "step": 4125 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "learning_rate": 1.906666666666667e-08, | |
| "loss": 0.1284, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 6.09, | |
| "learning_rate": 1.851111111111111e-08, | |
| "loss": 0.1379, | |
| "step": 4175 | |
| }, | |
| { | |
| "epoch": 6.09, | |
| "learning_rate": 1.7955555555555555e-08, | |
| "loss": 0.1184, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 6.1, | |
| "learning_rate": 1.7399999999999997e-08, | |
| "loss": 0.133, | |
| "step": 4225 | |
| }, | |
| { | |
| "epoch": 6.1, | |
| "learning_rate": 1.6844444444444445e-08, | |
| "loss": 0.1334, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 6.11, | |
| "learning_rate": 1.6288888888888887e-08, | |
| "loss": 0.1184, | |
| "step": 4275 | |
| }, | |
| { | |
| "epoch": 6.11, | |
| "learning_rate": 1.5733333333333332e-08, | |
| "loss": 0.1254, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "learning_rate": 1.5177777777777777e-08, | |
| "loss": 0.1212, | |
| "step": 4325 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "learning_rate": 1.4622222222222222e-08, | |
| "loss": 0.118, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 1.4066666666666665e-08, | |
| "loss": 0.124, | |
| "step": 4375 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "learning_rate": 1.3511111111111109e-08, | |
| "loss": 0.12, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "learning_rate": 1.2955555555555555e-08, | |
| "loss": 0.1266, | |
| "step": 4425 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "learning_rate": 1.2399999999999999e-08, | |
| "loss": 0.1211, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "learning_rate": 1.1844444444444445e-08, | |
| "loss": 0.1311, | |
| "step": 4475 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "learning_rate": 1.1288888888888889e-08, | |
| "loss": 0.1368, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "eval_cer": 7.655770513675784, | |
| "eval_loss": 0.40414080023765564, | |
| "eval_runtime": 1163.7244, | |
| "eval_samples_per_second": 1.063, | |
| "eval_steps_per_second": 0.266, | |
| "eval_wer": 15.760035095415661, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "learning_rate": 1.0733333333333333e-08, | |
| "loss": 0.1263, | |
| "step": 4525 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 1.0177777777777777e-08, | |
| "loss": 0.1122, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 9.622222222222222e-09, | |
| "loss": 0.1285, | |
| "step": 4575 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "learning_rate": 9.066666666666667e-09, | |
| "loss": 0.1236, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "learning_rate": 8.51111111111111e-09, | |
| "loss": 0.1237, | |
| "step": 4625 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "learning_rate": 7.955555555555555e-09, | |
| "loss": 0.1276, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "learning_rate": 7.399999999999999e-09, | |
| "loss": 0.1252, | |
| "step": 4675 | |
| }, | |
| { | |
| "epoch": 7.07, | |
| "learning_rate": 6.844444444444444e-09, | |
| "loss": 0.1223, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 7.07, | |
| "learning_rate": 6.288888888888888e-09, | |
| "loss": 0.1232, | |
| "step": 4725 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "learning_rate": 5.733333333333333e-09, | |
| "loss": 0.1205, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "learning_rate": 5.1777777777777775e-09, | |
| "loss": 0.1253, | |
| "step": 4775 | |
| }, | |
| { | |
| "epoch": 7.09, | |
| "learning_rate": 4.622222222222222e-09, | |
| "loss": 0.1359, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 7.09, | |
| "learning_rate": 4.066666666666666e-09, | |
| "loss": 0.1182, | |
| "step": 4825 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "learning_rate": 3.5111111111111113e-09, | |
| "loss": 0.1315, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "learning_rate": 2.9555555555555554e-09, | |
| "loss": 0.1221, | |
| "step": 4875 | |
| }, | |
| { | |
| "epoch": 7.11, | |
| "learning_rate": 2.4e-09, | |
| "loss": 0.1175, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 7.11, | |
| "learning_rate": 1.8444444444444443e-09, | |
| "loss": 0.1226, | |
| "step": 4925 | |
| }, | |
| { | |
| "epoch": 7.12, | |
| "learning_rate": 1.288888888888889e-09, | |
| "loss": 0.1228, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 7.12, | |
| "learning_rate": 7.333333333333333e-10, | |
| "loss": 0.1236, | |
| "step": 4975 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 1.7777777777777778e-10, | |
| "loss": 0.1231, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_cer": 7.6691127418278855, | |
| "eval_loss": 0.40414172410964966, | |
| "eval_runtime": 1171.5804, | |
| "eval_samples_per_second": 1.056, | |
| "eval_steps_per_second": 0.265, | |
| "eval_wer": 15.771002412809827, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "step": 5000, | |
| "total_flos": 2.546786102870016e+20, | |
| "train_loss": 0.2192393777370453, | |
| "train_runtime": 54122.9791, | |
| "train_samples_per_second": 2.217, | |
| "train_steps_per_second": 0.092 | |
| } | |
| ], | |
| "max_steps": 5000, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 2.546786102870016e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |