| { | |
| "best_metric": 6.3671568743912, | |
| "best_model_checkpoint": "./checkpoint-12000", | |
| "epoch": 2.095833333333333, | |
| "global_step": 12000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 0.0, | |
| "loss": 1.9643, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 9.4e-06, | |
| "loss": 1.184, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 1.94e-05, | |
| "loss": 0.4478, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 2.94e-05, | |
| "loss": 0.307, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 3.94e-05, | |
| "loss": 0.2939, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 4.94e-05, | |
| "loss": 0.2464, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 5.94e-05, | |
| "loss": 0.2403, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 6.939999999999999e-05, | |
| "loss": 0.2404, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 7.94e-05, | |
| "loss": 0.2563, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 8.94e-05, | |
| "loss": 0.2366, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 9.94e-05, | |
| "loss": 0.2237, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 9.959130434782609e-05, | |
| "loss": 0.234, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 9.915652173913044e-05, | |
| "loss": 0.2301, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 9.872173913043479e-05, | |
| "loss": 0.2404, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 9.828695652173913e-05, | |
| "loss": 0.2155, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 9.785217391304348e-05, | |
| "loss": 0.1606, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 9.741739130434783e-05, | |
| "loss": 0.1744, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 9.698260869565218e-05, | |
| "loss": 0.2005, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 9.654782608695653e-05, | |
| "loss": 0.2168, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 9.611304347826087e-05, | |
| "loss": 0.1448, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 9.567826086956522e-05, | |
| "loss": 0.1907, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_loss": 0.2546018362045288, | |
| "eval_runtime": 6940.5327, | |
| "eval_samples_per_second": 2.287, | |
| "eval_steps_per_second": 0.036, | |
| "eval_wer": 25.463875887957045, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 9.524347826086957e-05, | |
| "loss": 0.1314, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 9.480869565217392e-05, | |
| "loss": 0.1387, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.437391304347827e-05, | |
| "loss": 0.139, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.393913043478261e-05, | |
| "loss": 0.1395, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.350434782608696e-05, | |
| "loss": 0.1268, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 9.306956521739131e-05, | |
| "loss": 0.177, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 9.263478260869566e-05, | |
| "loss": 0.1791, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 9.22e-05, | |
| "loss": 0.0979, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 9.176521739130435e-05, | |
| "loss": 0.09, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 9.13304347826087e-05, | |
| "loss": 0.138, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 9.089565217391305e-05, | |
| "loss": 0.1316, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 9.04608695652174e-05, | |
| "loss": 0.104, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 9.002608695652174e-05, | |
| "loss": 0.1265, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 8.959130434782609e-05, | |
| "loss": 0.1421, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 8.915652173913044e-05, | |
| "loss": 0.1799, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 8.872173913043479e-05, | |
| "loss": 0.1733, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 8.828695652173914e-05, | |
| "loss": 0.1525, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 8.785217391304348e-05, | |
| "loss": 0.1638, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 8.741739130434783e-05, | |
| "loss": 0.1642, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 8.698260869565218e-05, | |
| "loss": 0.1482, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "eval_loss": 0.1641348898410797, | |
| "eval_runtime": 6931.8394, | |
| "eval_samples_per_second": 2.29, | |
| "eval_steps_per_second": 0.036, | |
| "eval_wer": 17.16756551281747, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 8.654782608695653e-05, | |
| "loss": 0.1376, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 8.611304347826088e-05, | |
| "loss": 0.1323, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 8.567826086956522e-05, | |
| "loss": 0.1637, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 8.524347826086957e-05, | |
| "loss": 0.1518, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 8.480869565217392e-05, | |
| "loss": 0.1357, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 8.437391304347827e-05, | |
| "loss": 0.1373, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 8.393913043478261e-05, | |
| "loss": 0.1257, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 8.350434782608696e-05, | |
| "loss": 0.1336, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 8.306956521739131e-05, | |
| "loss": 0.1351, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 8.263478260869566e-05, | |
| "loss": 0.1159, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 8.22e-05, | |
| "loss": 0.1198, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 8.176521739130435e-05, | |
| "loss": 0.1179, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 8.13304347826087e-05, | |
| "loss": 0.128, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 8.089565217391305e-05, | |
| "loss": 0.1037, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 8.04608695652174e-05, | |
| "loss": 0.1172, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 8.002608695652175e-05, | |
| "loss": 0.1169, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 7.959130434782609e-05, | |
| "loss": 0.1165, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 7.915652173913044e-05, | |
| "loss": 0.117, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 7.872173913043479e-05, | |
| "loss": 0.1225, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 7.828695652173914e-05, | |
| "loss": 0.1175, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 0.1453797072172165, | |
| "eval_runtime": 6986.1082, | |
| "eval_samples_per_second": 2.272, | |
| "eval_steps_per_second": 0.035, | |
| "eval_wer": 15.593990797716058, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 7.785217391304348e-05, | |
| "loss": 0.1011, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 7.741739130434783e-05, | |
| "loss": 0.1185, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 7.698260869565218e-05, | |
| "loss": 0.1094, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 7.654782608695651e-05, | |
| "loss": 0.1078, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 7.611304347826088e-05, | |
| "loss": 0.1171, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 7.567826086956522e-05, | |
| "loss": 0.1098, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 7.524347826086957e-05, | |
| "loss": 0.0955, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 7.480869565217392e-05, | |
| "loss": 0.1081, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 7.437391304347825e-05, | |
| "loss": 0.124, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 7.393913043478262e-05, | |
| "loss": 0.0987, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 7.350434782608696e-05, | |
| "loss": 0.0937, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 7.306956521739131e-05, | |
| "loss": 0.1307, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 7.263478260869565e-05, | |
| "loss": 0.0974, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 7.22e-05, | |
| "loss": 0.0971, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 7.176521739130435e-05, | |
| "loss": 0.1063, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 7.13304347826087e-05, | |
| "loss": 0.134, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 7.089565217391305e-05, | |
| "loss": 0.1118, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 7.046086956521738e-05, | |
| "loss": 0.109, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 7.002608695652175e-05, | |
| "loss": 0.1063, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 6.95913043478261e-05, | |
| "loss": 0.0958, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_loss": 0.12609383463859558, | |
| "eval_runtime": 6883.3335, | |
| "eval_samples_per_second": 2.306, | |
| "eval_steps_per_second": 0.036, | |
| "eval_wer": 13.262534350177791, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 6.915652173913044e-05, | |
| "loss": 0.0891, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 6.872173913043478e-05, | |
| "loss": 0.1144, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 6.828695652173914e-05, | |
| "loss": 0.096, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 6.785217391304349e-05, | |
| "loss": 0.1078, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 6.741739130434783e-05, | |
| "loss": 0.0791, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 6.698260869565218e-05, | |
| "loss": 0.0946, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 6.654782608695652e-05, | |
| "loss": 0.0767, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 6.611304347826088e-05, | |
| "loss": 0.0753, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 6.567826086956522e-05, | |
| "loss": 0.0857, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 6.524347826086957e-05, | |
| "loss": 0.0887, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 6.480869565217391e-05, | |
| "loss": 0.0723, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 6.437391304347827e-05, | |
| "loss": 0.0952, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 6.393913043478262e-05, | |
| "loss": 0.0687, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 6.350434782608696e-05, | |
| "loss": 0.0967, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 6.30695652173913e-05, | |
| "loss": 0.0804, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 6.263478260869565e-05, | |
| "loss": 0.118, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 6.220000000000001e-05, | |
| "loss": 0.1065, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 6.176521739130436e-05, | |
| "loss": 0.1005, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 6.13304347826087e-05, | |
| "loss": 0.1045, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 6.0895652173913044e-05, | |
| "loss": 0.099, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "eval_loss": 0.10123514384031296, | |
| "eval_runtime": 6887.4112, | |
| "eval_samples_per_second": 2.304, | |
| "eval_steps_per_second": 0.036, | |
| "eval_wer": 10.61430392878921, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 6.046086956521739e-05, | |
| "loss": 0.1035, | |
| "step": 5050 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 6.002608695652174e-05, | |
| "loss": 0.1015, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 5.9591304347826095e-05, | |
| "loss": 0.0909, | |
| "step": 5150 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 5.9156521739130436e-05, | |
| "loss": 0.1056, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 5.8721739130434784e-05, | |
| "loss": 0.1031, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 5.828695652173913e-05, | |
| "loss": 0.0842, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 5.7852173913043486e-05, | |
| "loss": 0.0866, | |
| "step": 5350 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 5.7417391304347834e-05, | |
| "loss": 0.0897, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 5.6982608695652175e-05, | |
| "loss": 0.067, | |
| "step": 5450 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 5.654782608695652e-05, | |
| "loss": 0.0373, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 5.611304347826087e-05, | |
| "loss": 0.0368, | |
| "step": 5550 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 5.5678260869565226e-05, | |
| "loss": 0.0372, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 5.524347826086957e-05, | |
| "loss": 0.0335, | |
| "step": 5650 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 5.4808695652173914e-05, | |
| "loss": 0.0361, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 5.437391304347826e-05, | |
| "loss": 0.0345, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 5.393913043478262e-05, | |
| "loss": 0.0357, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 5.3504347826086965e-05, | |
| "loss": 0.0353, | |
| "step": 5850 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 5.3069565217391306e-05, | |
| "loss": 0.0302, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 5.2634782608695654e-05, | |
| "loss": 0.0291, | |
| "step": 5950 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 5.22e-05, | |
| "loss": 0.028, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "eval_loss": 0.1052699089050293, | |
| "eval_runtime": 6861.1264, | |
| "eval_samples_per_second": 2.313, | |
| "eval_steps_per_second": 0.036, | |
| "eval_wer": 9.879388309456495, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 5.1765217391304356e-05, | |
| "loss": 0.0341, | |
| "step": 6050 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 5.133043478260869e-05, | |
| "loss": 0.058, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 5.0895652173913045e-05, | |
| "loss": 0.0644, | |
| "step": 6150 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 5.046086956521739e-05, | |
| "loss": 0.0605, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 5.002608695652175e-05, | |
| "loss": 0.0509, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 4.959130434782609e-05, | |
| "loss": 0.0499, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 4.915652173913044e-05, | |
| "loss": 0.0505, | |
| "step": 6350 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 4.8721739130434785e-05, | |
| "loss": 0.0574, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 4.828695652173913e-05, | |
| "loss": 0.0461, | |
| "step": 6450 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 4.785217391304348e-05, | |
| "loss": 0.0473, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 4.741739130434783e-05, | |
| "loss": 0.049, | |
| "step": 6550 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 4.6982608695652176e-05, | |
| "loss": 0.0507, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 4.6547826086956524e-05, | |
| "loss": 0.0481, | |
| "step": 6650 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 4.611304347826087e-05, | |
| "loss": 0.0708, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 4.567826086956522e-05, | |
| "loss": 0.0758, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 4.524347826086957e-05, | |
| "loss": 0.051, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 4.4808695652173915e-05, | |
| "loss": 0.0557, | |
| "step": 6850 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 4.437391304347826e-05, | |
| "loss": 0.0659, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 4.393913043478261e-05, | |
| "loss": 0.0795, | |
| "step": 6950 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 4.350434782608696e-05, | |
| "loss": 0.0473, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "eval_loss": 0.10288114845752716, | |
| "eval_runtime": 6910.216, | |
| "eval_samples_per_second": 2.297, | |
| "eval_steps_per_second": 0.036, | |
| "eval_wer": 10.307825109089038, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 4.306956521739131e-05, | |
| "loss": 0.0691, | |
| "step": 7050 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 4.2634782608695655e-05, | |
| "loss": 0.0436, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 4.22e-05, | |
| "loss": 0.0457, | |
| "step": 7150 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 4.176521739130435e-05, | |
| "loss": 0.0455, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 4.13304347826087e-05, | |
| "loss": 0.0437, | |
| "step": 7250 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 4.0895652173913046e-05, | |
| "loss": 0.0398, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 4.0460869565217394e-05, | |
| "loss": 0.0602, | |
| "step": 7350 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 4.002608695652174e-05, | |
| "loss": 0.0653, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 3.959130434782609e-05, | |
| "loss": 0.0318, | |
| "step": 7450 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 3.915652173913044e-05, | |
| "loss": 0.0295, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 3.8721739130434785e-05, | |
| "loss": 0.0442, | |
| "step": 7550 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 3.828695652173913e-05, | |
| "loss": 0.0443, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 3.785217391304348e-05, | |
| "loss": 0.0351, | |
| "step": 7650 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 3.741739130434783e-05, | |
| "loss": 0.0447, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 3.698260869565218e-05, | |
| "loss": 0.0494, | |
| "step": 7750 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 3.6547826086956525e-05, | |
| "loss": 0.0481, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 3.611304347826087e-05, | |
| "loss": 0.0404, | |
| "step": 7850 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 3.5678260869565214e-05, | |
| "loss": 0.0482, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 3.524347826086957e-05, | |
| "loss": 0.0416, | |
| "step": 7950 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 3.4808695652173916e-05, | |
| "loss": 0.0391, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "eval_loss": 0.09238624572753906, | |
| "eval_runtime": 6861.6911, | |
| "eval_samples_per_second": 2.313, | |
| "eval_steps_per_second": 0.036, | |
| "eval_wer": 9.24188068708275, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 3.4373913043478264e-05, | |
| "loss": 0.0473, | |
| "step": 8050 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 3.393913043478261e-05, | |
| "loss": 0.0423, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 3.350434782608696e-05, | |
| "loss": 0.0365, | |
| "step": 8150 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 3.306956521739131e-05, | |
| "loss": 0.0428, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 3.263478260869565e-05, | |
| "loss": 0.0479, | |
| "step": 8250 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 3.2200000000000003e-05, | |
| "loss": 0.0365, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 3.1765217391304345e-05, | |
| "loss": 0.035, | |
| "step": 8350 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 3.13304347826087e-05, | |
| "loss": 0.0523, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 3.089565217391304e-05, | |
| "loss": 0.0602, | |
| "step": 8450 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 3.046086956521739e-05, | |
| "loss": 0.0504, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 3.0026086956521743e-05, | |
| "loss": 0.05, | |
| "step": 8550 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 2.9591304347826087e-05, | |
| "loss": 0.0579, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 2.915652173913044e-05, | |
| "loss": 0.0509, | |
| "step": 8650 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 2.8721739130434783e-05, | |
| "loss": 0.0439, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 2.8286956521739134e-05, | |
| "loss": 0.0435, | |
| "step": 8750 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 2.785217391304348e-05, | |
| "loss": 0.0419, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 2.7417391304347827e-05, | |
| "loss": 0.0615, | |
| "step": 8850 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 2.6982608695652174e-05, | |
| "loss": 0.0404, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 2.6547826086956522e-05, | |
| "loss": 0.045, | |
| "step": 8950 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 2.6113043478260874e-05, | |
| "loss": 0.0423, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "eval_loss": 0.0796533077955246, | |
| "eval_runtime": 6857.0536, | |
| "eval_samples_per_second": 2.315, | |
| "eval_steps_per_second": 0.036, | |
| "eval_wer": 7.924892890800092, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 2.5678260869565218e-05, | |
| "loss": 0.031, | |
| "step": 9050 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 2.524347826086957e-05, | |
| "loss": 0.0355, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 2.4808695652173914e-05, | |
| "loss": 0.0278, | |
| "step": 9150 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 2.437391304347826e-05, | |
| "loss": 0.0295, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 2.3939130434782613e-05, | |
| "loss": 0.0323, | |
| "step": 9250 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 2.3504347826086957e-05, | |
| "loss": 0.0313, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 2.3069565217391305e-05, | |
| "loss": 0.029, | |
| "step": 9350 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 2.2634782608695653e-05, | |
| "loss": 0.0327, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 2.22e-05, | |
| "loss": 0.0246, | |
| "step": 9450 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 2.176521739130435e-05, | |
| "loss": 0.0333, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 2.1330434782608697e-05, | |
| "loss": 0.031, | |
| "step": 9550 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 2.0895652173913044e-05, | |
| "loss": 0.0471, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 2.0460869565217392e-05, | |
| "loss": 0.0725, | |
| "step": 9650 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 2.002608695652174e-05, | |
| "loss": 0.0773, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 1.9591304347826088e-05, | |
| "loss": 0.0616, | |
| "step": 9750 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 1.9156521739130436e-05, | |
| "loss": 0.0698, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 1.8721739130434784e-05, | |
| "loss": 0.0631, | |
| "step": 9850 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 1.828695652173913e-05, | |
| "loss": 0.0592, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 1.785217391304348e-05, | |
| "loss": 0.0583, | |
| "step": 9950 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 1.7417391304347827e-05, | |
| "loss": 0.0604, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_loss": 0.06884036958217621, | |
| "eval_runtime": 6836.4142, | |
| "eval_samples_per_second": 2.322, | |
| "eval_steps_per_second": 0.036, | |
| "eval_wer": 7.014959650915081, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 1.6982608695652175e-05, | |
| "loss": 0.0542, | |
| "step": 10050 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 1.6556521739130434e-05, | |
| "loss": 0.0485, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 1.612173913043478e-05, | |
| "loss": 0.0483, | |
| "step": 10150 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 1.5686956521739133e-05, | |
| "loss": 0.0484, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 1.5252173913043479e-05, | |
| "loss": 0.0484, | |
| "step": 10250 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 1.4817391304347827e-05, | |
| "loss": 0.0385, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 1.4382608695652175e-05, | |
| "loss": 0.0295, | |
| "step": 10350 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 1.3947826086956523e-05, | |
| "loss": 0.035, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 1.351304347826087e-05, | |
| "loss": 0.028, | |
| "step": 10450 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 1.3078260869565218e-05, | |
| "loss": 0.0436, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 1.2643478260869565e-05, | |
| "loss": 0.0334, | |
| "step": 10550 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 1.2208695652173914e-05, | |
| "loss": 0.032, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 1.1773913043478262e-05, | |
| "loss": 0.0324, | |
| "step": 10650 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 1.133913043478261e-05, | |
| "loss": 0.0309, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 1.0904347826086956e-05, | |
| "loss": 0.0269, | |
| "step": 10750 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 1.0469565217391304e-05, | |
| "loss": 0.0341, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 1.0043478260869566e-05, | |
| "loss": 0.0326, | |
| "step": 10850 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 9.608695652173914e-06, | |
| "loss": 0.0175, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 9.173913043478261e-06, | |
| "loss": 0.0117, | |
| "step": 10950 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 8.73913043478261e-06, | |
| "loss": 0.0121, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "eval_loss": 0.06960244476795197, | |
| "eval_runtime": 6836.8996, | |
| "eval_samples_per_second": 2.322, | |
| "eval_steps_per_second": 0.036, | |
| "eval_wer": 6.463772936415545, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 8.304347826086957e-06, | |
| "loss": 0.0119, | |
| "step": 11050 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 7.869565217391305e-06, | |
| "loss": 0.0163, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 7.434782608695652e-06, | |
| "loss": 0.0136, | |
| "step": 11150 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 7.000000000000001e-06, | |
| "loss": 0.0124, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 6.565217391304349e-06, | |
| "loss": 0.0124, | |
| "step": 11250 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 6.1304347826086965e-06, | |
| "loss": 0.0135, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 5.6956521739130435e-06, | |
| "loss": 0.0111, | |
| "step": 11350 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 5.260869565217392e-06, | |
| "loss": 0.0128, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 4.826086956521739e-06, | |
| "loss": 0.0142, | |
| "step": 11450 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 4.391304347826087e-06, | |
| "loss": 0.0136, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 3.956521739130435e-06, | |
| "loss": 0.0157, | |
| "step": 11550 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 3.521739130434783e-06, | |
| "loss": 0.0154, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 3.0869565217391307e-06, | |
| "loss": 0.016, | |
| "step": 11650 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 2.6521739130434785e-06, | |
| "loss": 0.0146, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 2.217391304347826e-06, | |
| "loss": 0.0133, | |
| "step": 11750 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 1.7826086956521738e-06, | |
| "loss": 0.0142, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 1.3478260869565217e-06, | |
| "loss": 0.0155, | |
| "step": 11850 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 9.130434782608697e-07, | |
| "loss": 0.012, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 4.782608695652174e-07, | |
| "loss": 0.0149, | |
| "step": 11950 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 4.347826086956522e-08, | |
| "loss": 0.0155, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "eval_loss": 0.07060841470956802, | |
| "eval_runtime": 6855.2466, | |
| "eval_samples_per_second": 2.315, | |
| "eval_steps_per_second": 0.036, | |
| "eval_wer": 6.3671568743912, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "step": 12000, | |
| "total_flos": 2.2160934627213312e+20, | |
| "train_loss": 0.08807742363711198, | |
| "train_runtime": 202754.3941, | |
| "train_samples_per_second": 3.788, | |
| "train_steps_per_second": 0.059 | |
| } | |
| ], | |
| "max_steps": 12000, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 2.2160934627213312e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |