| { | |
| "best_metric": 9.891459959995533, | |
| "best_model_checkpoint": "./checkpoint-4000", | |
| "epoch": 5.0375, | |
| "global_step": 4000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 4.2000000000000006e-07, | |
| "loss": 2.3321, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 9.200000000000001e-07, | |
| "loss": 1.2935, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.42e-06, | |
| "loss": 0.9663, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 1.9200000000000003e-06, | |
| "loss": 0.4759, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 2.42e-06, | |
| "loss": 0.3237, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 2.92e-06, | |
| "loss": 0.2198, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 3.4200000000000007e-06, | |
| "loss": 0.2898, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 3.920000000000001e-06, | |
| "loss": 0.4063, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 4.42e-06, | |
| "loss": 0.3432, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 4.92e-06, | |
| "loss": 0.5207, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 5.420000000000001e-06, | |
| "loss": 0.5485, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 5.92e-06, | |
| "loss": 0.3309, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 6.42e-06, | |
| "loss": 0.3645, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 6.92e-06, | |
| "loss": 0.5326, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 7.420000000000001e-06, | |
| "loss": 0.2653, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 7.92e-06, | |
| "loss": 0.2754, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 8.42e-06, | |
| "loss": 0.2309, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 8.920000000000001e-06, | |
| "loss": 0.2041, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 9.42e-06, | |
| "loss": 0.2372, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 9.920000000000002e-06, | |
| "loss": 0.201, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_loss": 0.23912189900875092, | |
| "eval_runtime": 1324.0512, | |
| "eval_samples_per_second": 3.384, | |
| "eval_steps_per_second": 0.106, | |
| "eval_wer": 10.83877387322442, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 9.940000000000001e-06, | |
| "loss": 0.2184, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 9.86857142857143e-06, | |
| "loss": 0.1775, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 9.797142857142858e-06, | |
| "loss": 0.3417, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 9.725714285714287e-06, | |
| "loss": 0.3732, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 9.654285714285716e-06, | |
| "loss": 0.2438, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 9.582857142857143e-06, | |
| "loss": 0.2139, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 9.511428571428572e-06, | |
| "loss": 0.2395, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 9.440000000000001e-06, | |
| "loss": 0.1988, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 9.368571428571428e-06, | |
| "loss": 0.2073, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 9.297142857142857e-06, | |
| "loss": 0.2149, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 9.225714285714286e-06, | |
| "loss": 0.2417, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 9.154285714285715e-06, | |
| "loss": 0.137, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 9.082857142857143e-06, | |
| "loss": 0.2751, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 9.011428571428572e-06, | |
| "loss": 0.3086, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 8.94e-06, | |
| "loss": 0.1372, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 8.86857142857143e-06, | |
| "loss": 0.054, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 8.797142857142857e-06, | |
| "loss": 0.0775, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 8.725714285714286e-06, | |
| "loss": 0.1201, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 8.654285714285715e-06, | |
| "loss": 0.1379, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 8.582857142857144e-06, | |
| "loss": 0.1726, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "eval_loss": 0.20546844601631165, | |
| "eval_runtime": 1256.3177, | |
| "eval_samples_per_second": 3.566, | |
| "eval_steps_per_second": 0.111, | |
| "eval_wer": 11.159621886708159, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 8.511428571428571e-06, | |
| "loss": 0.1883, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 8.44e-06, | |
| "loss": 0.2804, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 8.36857142857143e-06, | |
| "loss": 0.2031, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 8.297142857142859e-06, | |
| "loss": 0.1453, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 8.225714285714288e-06, | |
| "loss": 0.125, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 8.154285714285715e-06, | |
| "loss": 0.0838, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 8.082857142857144e-06, | |
| "loss": 0.1691, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 8.011428571428573e-06, | |
| "loss": 0.2264, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 7.94e-06, | |
| "loss": 0.2588, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 7.86857142857143e-06, | |
| "loss": 0.1729, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 7.797142857142858e-06, | |
| "loss": 0.168, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 7.725714285714286e-06, | |
| "loss": 0.1205, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 7.654285714285715e-06, | |
| "loss": 0.191, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 7.5828571428571444e-06, | |
| "loss": 0.416, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 7.511428571428572e-06, | |
| "loss": 0.2462, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 7.440000000000001e-06, | |
| "loss": 0.1218, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 7.36857142857143e-06, | |
| "loss": 0.1181, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 7.297142857142858e-06, | |
| "loss": 0.1055, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 7.225714285714286e-06, | |
| "loss": 0.1087, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 7.154285714285715e-06, | |
| "loss": 0.0984, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "eval_loss": 0.20248055458068848, | |
| "eval_runtime": 1249.5933, | |
| "eval_samples_per_second": 3.585, | |
| "eval_steps_per_second": 0.112, | |
| "eval_wer": 10.407253601925088, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 7.082857142857143e-06, | |
| "loss": 0.1287, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 7.011428571428572e-06, | |
| "loss": 0.084, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 6.9400000000000005e-06, | |
| "loss": 0.0439, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 6.868571428571429e-06, | |
| "loss": 0.0856, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 6.797142857142858e-06, | |
| "loss": 0.0517, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 6.725714285714287e-06, | |
| "loss": 0.051, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 6.654285714285716e-06, | |
| "loss": 0.0597, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 6.582857142857143e-06, | |
| "loss": 0.0487, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 6.511428571428572e-06, | |
| "loss": 0.0821, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 6.440000000000001e-06, | |
| "loss": 0.0649, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 6.368571428571429e-06, | |
| "loss": 0.0238, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 6.297142857142857e-06, | |
| "loss": 0.0313, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 6.225714285714286e-06, | |
| "loss": 0.0547, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 6.1542857142857145e-06, | |
| "loss": 0.0612, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 6.0828571428571435e-06, | |
| "loss": 0.0647, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 6.011428571428572e-06, | |
| "loss": 0.0536, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 5.94e-06, | |
| "loss": 0.0565, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 5.868571428571429e-06, | |
| "loss": 0.078, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 5.797142857142858e-06, | |
| "loss": 0.0753, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 5.725714285714287e-06, | |
| "loss": 0.0782, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "eval_loss": 0.2072889655828476, | |
| "eval_runtime": 1303.2739, | |
| "eval_samples_per_second": 3.437, | |
| "eval_steps_per_second": 0.107, | |
| "eval_wer": 10.904771091187849, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 5.654285714285714e-06, | |
| "loss": 0.147, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 5.582857142857143e-06, | |
| "loss": 0.1905, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 5.511428571428572e-06, | |
| "loss": 0.3605, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 5.4400000000000004e-06, | |
| "loss": 0.2344, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 5.368571428571429e-06, | |
| "loss": 0.1066, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 5.297142857142858e-06, | |
| "loss": 0.0752, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 5.225714285714286e-06, | |
| "loss": 0.1042, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 5.154285714285715e-06, | |
| "loss": 0.1903, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 5.082857142857144e-06, | |
| "loss": 0.1248, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 5.011428571428571e-06, | |
| "loss": 0.0754, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 4.94e-06, | |
| "loss": 0.0596, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 4.868571428571429e-06, | |
| "loss": 0.1033, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.797142857142857e-06, | |
| "loss": 0.0651, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 4.725714285714286e-06, | |
| "loss": 0.0289, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 4.6542857142857145e-06, | |
| "loss": 0.0956, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 4.5828571428571435e-06, | |
| "loss": 0.2166, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 4.511428571428572e-06, | |
| "loss": 0.1193, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 4.440000000000001e-06, | |
| "loss": 0.0627, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 4.368571428571429e-06, | |
| "loss": 0.045, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 4.297142857142858e-06, | |
| "loss": 0.0301, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "eval_loss": 0.22605395317077637, | |
| "eval_runtime": 1242.3125, | |
| "eval_samples_per_second": 3.606, | |
| "eval_steps_per_second": 0.113, | |
| "eval_wer": 10.178801693590147, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 4.225714285714286e-06, | |
| "loss": 0.0428, | |
| "step": 2525 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 4.154285714285714e-06, | |
| "loss": 0.03, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 4.082857142857143e-06, | |
| "loss": 0.0148, | |
| "step": 2575 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 4.011428571428571e-06, | |
| "loss": 0.0359, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 3.94e-06, | |
| "loss": 0.0384, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 3.8685714285714286e-06, | |
| "loss": 0.0219, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "learning_rate": 3.7971428571428576e-06, | |
| "loss": 0.0183, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 3.7257142857142857e-06, | |
| "loss": 0.0127, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 3.6542857142857148e-06, | |
| "loss": 0.0214, | |
| "step": 2725 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "learning_rate": 3.582857142857143e-06, | |
| "loss": 0.0519, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 3.511428571428572e-06, | |
| "loss": 0.0502, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 3.44e-06, | |
| "loss": 0.0795, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "learning_rate": 3.3685714285714287e-06, | |
| "loss": 0.0423, | |
| "step": 2825 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "learning_rate": 3.2971428571428577e-06, | |
| "loss": 0.0254, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 3.225714285714286e-06, | |
| "loss": 0.0108, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "learning_rate": 3.154285714285715e-06, | |
| "loss": 0.0253, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "learning_rate": 3.082857142857143e-06, | |
| "loss": 0.0946, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "learning_rate": 3.0114285714285716e-06, | |
| "loss": 0.0742, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 2.9400000000000002e-06, | |
| "loss": 0.0344, | |
| "step": 2975 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 2.868571428571429e-06, | |
| "loss": 0.0284, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "eval_loss": 0.2303255945444107, | |
| "eval_runtime": 1235.1133, | |
| "eval_samples_per_second": 3.627, | |
| "eval_steps_per_second": 0.113, | |
| "eval_wer": 10.063052726700445, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "learning_rate": 2.797142857142857e-06, | |
| "loss": 0.0402, | |
| "step": 3025 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 2.725714285714286e-06, | |
| "loss": 0.0446, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 2.654285714285714e-06, | |
| "loss": 0.0908, | |
| "step": 3075 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 2.582857142857143e-06, | |
| "loss": 0.0283, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 2.5114285714285718e-06, | |
| "loss": 0.007, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 2.4400000000000004e-06, | |
| "loss": 0.004, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 2.3685714285714285e-06, | |
| "loss": 0.0035, | |
| "step": 3175 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 2.297142857142857e-06, | |
| "loss": 0.0145, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 2.2257142857142857e-06, | |
| "loss": 0.0748, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 2.1542857142857147e-06, | |
| "loss": 0.0918, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 2.0828571428571433e-06, | |
| "loss": 0.0311, | |
| "step": 3275 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 2.0114285714285715e-06, | |
| "loss": 0.0149, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 1.94e-06, | |
| "loss": 0.0068, | |
| "step": 3325 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 1.8685714285714289e-06, | |
| "loss": 0.0278, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 1.7971428571428572e-06, | |
| "loss": 0.0141, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "learning_rate": 1.7257142857142858e-06, | |
| "loss": 0.0152, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "learning_rate": 1.6542857142857144e-06, | |
| "loss": 0.0234, | |
| "step": 3425 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "learning_rate": 1.582857142857143e-06, | |
| "loss": 0.0427, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "learning_rate": 1.5114285714285714e-06, | |
| "loss": 0.0313, | |
| "step": 3475 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "learning_rate": 1.44e-06, | |
| "loss": 0.0153, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "eval_loss": 0.2513434886932373, | |
| "eval_runtime": 1269.5341, | |
| "eval_samples_per_second": 3.529, | |
| "eval_steps_per_second": 0.11, | |
| "eval_wer": 10.48035821259227, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "learning_rate": 1.3685714285714286e-06, | |
| "loss": 0.0099, | |
| "step": 3525 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "learning_rate": 1.2971428571428574e-06, | |
| "loss": 0.0066, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "learning_rate": 1.2257142857142857e-06, | |
| "loss": 0.0121, | |
| "step": 3575 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "learning_rate": 1.1542857142857143e-06, | |
| "loss": 0.0208, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "learning_rate": 1.082857142857143e-06, | |
| "loss": 0.0544, | |
| "step": 3625 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "learning_rate": 1.0114285714285715e-06, | |
| "loss": 0.0481, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 4.15, | |
| "learning_rate": 9.400000000000001e-07, | |
| "loss": 0.0175, | |
| "step": 3675 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "learning_rate": 8.685714285714286e-07, | |
| "loss": 0.0103, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "learning_rate": 7.971428571428572e-07, | |
| "loss": 0.0164, | |
| "step": 3725 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "learning_rate": 7.257142857142857e-07, | |
| "loss": 0.0128, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "learning_rate": 6.542857142857144e-07, | |
| "loss": 0.0064, | |
| "step": 3775 | |
| }, | |
| { | |
| "epoch": 4.18, | |
| "learning_rate": 5.82857142857143e-07, | |
| "loss": 0.0122, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "learning_rate": 5.114285714285714e-07, | |
| "loss": 0.0179, | |
| "step": 3825 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "learning_rate": 4.4e-07, | |
| "loss": 0.0121, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 3.685714285714286e-07, | |
| "loss": 0.008, | |
| "step": 3875 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 2.9714285714285715e-07, | |
| "loss": 0.0064, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "learning_rate": 2.2571428571428574e-07, | |
| "loss": 0.0012, | |
| "step": 3925 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 1.542857142857143e-07, | |
| "loss": 0.009, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 8.285714285714285e-08, | |
| "loss": 0.0567, | |
| "step": 3975 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "learning_rate": 1.142857142857143e-08, | |
| "loss": 0.0386, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "eval_loss": 0.2617354393005371, | |
| "eval_runtime": 1232.317, | |
| "eval_samples_per_second": 3.635, | |
| "eval_steps_per_second": 0.114, | |
| "eval_wer": 9.891459959995533, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "step": 4000, | |
| "total_flos": 1.305761918681088e+20, | |
| "train_loss": 0.14665698340721428, | |
| "train_runtime": 48253.4406, | |
| "train_samples_per_second": 2.653, | |
| "train_steps_per_second": 0.083 | |
| } | |
| ], | |
| "max_steps": 4000, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 1.305761918681088e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |