| { | |
| "best_metric": 10.977473664409331, | |
| "best_model_checkpoint": "./checkpoint-2000", | |
| "epoch": 3.1285, | |
| "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": 4.4e-07, | |
| "loss": 1.2598, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 9.400000000000001e-07, | |
| "loss": 1.0197, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.42e-06, | |
| "loss": 0.6361, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 1.9200000000000003e-06, | |
| "loss": 0.4259, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 2.4000000000000003e-06, | |
| "loss": 0.3557, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 2.9e-06, | |
| "loss": 0.374, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 3.4000000000000005e-06, | |
| "loss": 0.3457, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 3.88e-06, | |
| "loss": 0.3889, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 4.38e-06, | |
| "loss": 0.3082, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 4.880000000000001e-06, | |
| "loss": 0.2432, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 5.380000000000001e-06, | |
| "loss": 0.1686, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 5.8800000000000005e-06, | |
| "loss": 0.1364, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 6.380000000000001e-06, | |
| "loss": 0.0969, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 6.88e-06, | |
| "loss": 0.1039, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 7.3800000000000005e-06, | |
| "loss": 0.1364, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 7.88e-06, | |
| "loss": 0.2883, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 8.380000000000001e-06, | |
| "loss": 0.4871, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 8.880000000000001e-06, | |
| "loss": 0.5703, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 9.38e-06, | |
| "loss": 0.4629, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 9.88e-06, | |
| "loss": 0.3639, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_loss": 0.25124552845954895, | |
| "eval_runtime": 1743.999, | |
| "eval_samples_per_second": 2.135, | |
| "eval_steps_per_second": 0.067, | |
| "eval_wer": 12.959697140707299, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 9.945714285714286e-06, | |
| "loss": 0.2625, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 9.874285714285715e-06, | |
| "loss": 0.275, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 9.802857142857144e-06, | |
| "loss": 0.2818, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 9.731428571428573e-06, | |
| "loss": 0.264, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 9.66e-06, | |
| "loss": 0.2065, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 9.58857142857143e-06, | |
| "loss": 0.2068, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 9.517142857142859e-06, | |
| "loss": 0.1838, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 9.445714285714288e-06, | |
| "loss": 0.2076, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 9.374285714285715e-06, | |
| "loss": 0.2077, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 9.302857142857144e-06, | |
| "loss": 0.3448, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 9.231428571428573e-06, | |
| "loss": 0.4413, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 9.16e-06, | |
| "loss": 0.2996, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 9.08857142857143e-06, | |
| "loss": 0.2807, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 9.017142857142858e-06, | |
| "loss": 0.2254, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 8.945714285714286e-06, | |
| "loss": 0.2145, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 8.874285714285715e-06, | |
| "loss": 0.2132, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 8.802857142857144e-06, | |
| "loss": 0.1821, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 8.731428571428571e-06, | |
| "loss": 0.264, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 8.66e-06, | |
| "loss": 0.1965, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 8.588571428571429e-06, | |
| "loss": 0.1931, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 0.21230609714984894, | |
| "eval_runtime": 1727.7901, | |
| "eval_samples_per_second": 2.155, | |
| "eval_steps_per_second": 0.068, | |
| "eval_wer": 12.141412716328066, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 8.517142857142858e-06, | |
| "loss": 0.2059, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 8.445714285714285e-06, | |
| "loss": 0.2057, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 8.374285714285714e-06, | |
| "loss": 0.2104, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 8.302857142857143e-06, | |
| "loss": 0.2285, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 8.231428571428572e-06, | |
| "loss": 0.2429, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 8.16e-06, | |
| "loss": 0.247, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 8.088571428571429e-06, | |
| "loss": 0.1586, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 8.017142857142858e-06, | |
| "loss": 0.0678, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 7.945714285714287e-06, | |
| "loss": 0.1091, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 7.874285714285716e-06, | |
| "loss": 0.0976, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 7.802857142857143e-06, | |
| "loss": 0.0975, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 7.731428571428572e-06, | |
| "loss": 0.0977, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 7.660000000000001e-06, | |
| "loss": 0.1087, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 7.588571428571429e-06, | |
| "loss": 0.0979, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 7.5171428571428575e-06, | |
| "loss": 0.1594, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 7.445714285714286e-06, | |
| "loss": 0.1942, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 7.374285714285715e-06, | |
| "loss": 0.2218, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 7.302857142857144e-06, | |
| "loss": 0.3135, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 7.231428571428573e-06, | |
| "loss": 0.4001, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 7.16e-06, | |
| "loss": 0.329, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "eval_loss": 0.20638370513916016, | |
| "eval_runtime": 1710.5053, | |
| "eval_samples_per_second": 2.177, | |
| "eval_steps_per_second": 0.068, | |
| "eval_wer": 11.581781414597442, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 7.088571428571429e-06, | |
| "loss": 0.2257, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 7.017142857142858e-06, | |
| "loss": 0.1623, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 6.945714285714287e-06, | |
| "loss": 0.1411, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 6.874285714285714e-06, | |
| "loss": 0.1541, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 6.8028571428571434e-06, | |
| "loss": 0.1714, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 6.7314285714285724e-06, | |
| "loss": 0.1389, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 6.660000000000001e-06, | |
| "loss": 0.1406, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 6.588571428571429e-06, | |
| "loss": 0.108, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 6.517142857142858e-06, | |
| "loss": 0.0681, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 6.445714285714286e-06, | |
| "loss": 0.0646, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 6.374285714285715e-06, | |
| "loss": 0.2618, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 6.302857142857144e-06, | |
| "loss": 0.2948, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 6.231428571428571e-06, | |
| "loss": 0.1995, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 6.16e-06, | |
| "loss": 0.1951, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 6.088571428571429e-06, | |
| "loss": 0.1935, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 6.017142857142858e-06, | |
| "loss": 0.1738, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 5.945714285714286e-06, | |
| "loss": 0.1357, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 5.874285714285715e-06, | |
| "loss": 0.1112, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 5.802857142857144e-06, | |
| "loss": 0.1284, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 5.731428571428572e-06, | |
| "loss": 0.097, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "eval_loss": 0.20503124594688416, | |
| "eval_runtime": 1729.8614, | |
| "eval_samples_per_second": 2.153, | |
| "eval_steps_per_second": 0.068, | |
| "eval_wer": 10.977473664409331, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 5.66e-06, | |
| "loss": 0.0764, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 5.588571428571429e-06, | |
| "loss": 0.0894, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 5.517142857142857e-06, | |
| "loss": 0.1047, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 5.445714285714286e-06, | |
| "loss": 0.1001, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 5.374285714285715e-06, | |
| "loss": 0.0944, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 5.3028571428571425e-06, | |
| "loss": 0.0636, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 5.2314285714285716e-06, | |
| "loss": 0.1307, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 5.1600000000000006e-06, | |
| "loss": 0.2306, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 5.08857142857143e-06, | |
| "loss": 0.2631, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 5.017142857142857e-06, | |
| "loss": 0.2219, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 4.945714285714286e-06, | |
| "loss": 0.2094, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 4.874285714285715e-06, | |
| "loss": 0.16, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.802857142857143e-06, | |
| "loss": 0.1558, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 4.731428571428572e-06, | |
| "loss": 0.0687, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 4.66e-06, | |
| "loss": 0.0572, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 4.588571428571429e-06, | |
| "loss": 0.064, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 4.5171428571428575e-06, | |
| "loss": 0.0722, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 4.445714285714286e-06, | |
| "loss": 0.0647, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 4.374285714285715e-06, | |
| "loss": 0.0475, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 4.302857142857143e-06, | |
| "loss": 0.0522, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "eval_loss": 0.22582364082336426, | |
| "eval_runtime": 1686.8156, | |
| "eval_samples_per_second": 2.208, | |
| "eval_steps_per_second": 0.069, | |
| "eval_wer": 10.439004890895411, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 4.231428571428572e-06, | |
| "loss": 0.0725, | |
| "step": 2525 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 4.16e-06, | |
| "loss": 0.087, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 4.088571428571429e-06, | |
| "loss": 0.092, | |
| "step": 2575 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 4.017142857142857e-06, | |
| "loss": 0.1035, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 3.945714285714286e-06, | |
| "loss": 0.0851, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 3.874285714285715e-06, | |
| "loss": 0.0792, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 3.802857142857143e-06, | |
| "loss": 0.0712, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 3.731428571428572e-06, | |
| "loss": 0.0599, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 3.66e-06, | |
| "loss": 0.0514, | |
| "step": 2725 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 3.588571428571429e-06, | |
| "loss": 0.0469, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 3.5171428571428573e-06, | |
| "loss": 0.0497, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 3.4457142857142863e-06, | |
| "loss": 0.0181, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 3.3742857142857145e-06, | |
| "loss": 0.0241, | |
| "step": 2825 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 3.302857142857143e-06, | |
| "loss": 0.0265, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 3.2314285714285716e-06, | |
| "loss": 0.1353, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 3.1600000000000002e-06, | |
| "loss": 0.1835, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 3.0885714285714284e-06, | |
| "loss": 0.1452, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 3.0171428571428574e-06, | |
| "loss": 0.0891, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 2.945714285714286e-06, | |
| "loss": 0.0867, | |
| "step": 2975 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 2.8742857142857146e-06, | |
| "loss": 0.1026, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "eval_loss": 0.22012001276016235, | |
| "eval_runtime": 1713.363, | |
| "eval_samples_per_second": 2.174, | |
| "eval_steps_per_second": 0.068, | |
| "eval_wer": 11.701702407825433, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 2.802857142857143e-06, | |
| "loss": 0.1778, | |
| "step": 3025 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 2.7314285714285714e-06, | |
| "loss": 0.1322, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 2.6600000000000004e-06, | |
| "loss": 0.0836, | |
| "step": 3075 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 2.5885714285714285e-06, | |
| "loss": 0.0632, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 2.5171428571428575e-06, | |
| "loss": 0.0584, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 2.445714285714286e-06, | |
| "loss": 0.1935, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 2.3742857142857147e-06, | |
| "loss": 0.2403, | |
| "step": 3175 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 2.302857142857143e-06, | |
| "loss": 0.2374, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 2.2314285714285715e-06, | |
| "loss": 0.1677, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 2.16e-06, | |
| "loss": 0.1603, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 2.0885714285714287e-06, | |
| "loss": 0.106, | |
| "step": 3275 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 2.0171428571428573e-06, | |
| "loss": 0.066, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 1.945714285714286e-06, | |
| "loss": 0.0887, | |
| "step": 3325 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 1.8742857142857142e-06, | |
| "loss": 0.0698, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 1.8028571428571432e-06, | |
| "loss": 0.0736, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 1.7314285714285716e-06, | |
| "loss": 0.0723, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 1.6600000000000002e-06, | |
| "loss": 0.0496, | |
| "step": 3425 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 1.5885714285714288e-06, | |
| "loss": 0.0292, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 1.5171428571428574e-06, | |
| "loss": 0.015, | |
| "step": 3475 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 1.4457142857142858e-06, | |
| "loss": 0.0448, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.2286548763513565, | |
| "eval_runtime": 1674.9882, | |
| "eval_samples_per_second": 2.223, | |
| "eval_steps_per_second": 0.07, | |
| "eval_wer": 10.387274266365688, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 1.3742857142857143e-06, | |
| "loss": 0.0398, | |
| "step": 3525 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 1.302857142857143e-06, | |
| "loss": 0.0815, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 1.2314285714285715e-06, | |
| "loss": 0.0988, | |
| "step": 3575 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 1.1600000000000001e-06, | |
| "loss": 0.0707, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 1.0885714285714287e-06, | |
| "loss": 0.0515, | |
| "step": 3625 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 1.0171428571428573e-06, | |
| "loss": 0.0431, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 9.457142857142858e-07, | |
| "loss": 0.0286, | |
| "step": 3675 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 8.742857142857144e-07, | |
| "loss": 0.0196, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 8.028571428571429e-07, | |
| "loss": 0.0147, | |
| "step": 3725 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 7.314285714285715e-07, | |
| "loss": 0.012, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 6.6e-07, | |
| "loss": 0.0195, | |
| "step": 3775 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 5.885714285714286e-07, | |
| "loss": 0.0205, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 5.171428571428572e-07, | |
| "loss": 0.0284, | |
| "step": 3825 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "learning_rate": 4.457142857142858e-07, | |
| "loss": 0.0671, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 3.7428571428571434e-07, | |
| "loss": 0.0644, | |
| "step": 3875 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 3.028571428571429e-07, | |
| "loss": 0.0505, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "learning_rate": 2.3142857142857144e-07, | |
| "loss": 0.031, | |
| "step": 3925 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 1.6e-07, | |
| "loss": 0.0319, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 8.857142857142858e-08, | |
| "loss": 0.0381, | |
| "step": 3975 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "learning_rate": 1.7142857142857143e-08, | |
| "loss": 0.0455, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "eval_loss": 0.2380528599023819, | |
| "eval_runtime": 1750.3543, | |
| "eval_samples_per_second": 2.128, | |
| "eval_steps_per_second": 0.067, | |
| "eval_wer": 11.124435665914222, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "step": 4000, | |
| "total_flos": 1.358319892267008e+20, | |
| "train_loss": 0.16877834644168616, | |
| "train_runtime": 42775.4284, | |
| "train_samples_per_second": 1.496, | |
| "train_steps_per_second": 0.094 | |
| } | |
| ], | |
| "max_steps": 4000, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 1.358319892267008e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |