| { | |
| "best_metric": 27.72060783790989, | |
| "best_model_checkpoint": "./checkpoint-1250", | |
| "epoch": 666.001, | |
| "eval_steps": 250, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 2.5e-08, | |
| "loss": 0.5919, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 5e-08, | |
| "loss": 0.582, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 7.5e-08, | |
| "loss": 0.5785, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 1e-07, | |
| "loss": 0.5606, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 1.25e-07, | |
| "loss": 0.5503, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 1.5e-07, | |
| "loss": 0.5353, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "learning_rate": 1.75e-07, | |
| "loss": 0.5179, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "learning_rate": 2e-07, | |
| "loss": 0.5118, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "learning_rate": 2.25e-07, | |
| "loss": 0.4893, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "learning_rate": 2.5e-07, | |
| "loss": 0.4805, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_loss": 0.4867776334285736, | |
| "eval_runtime": 106.4902, | |
| "eval_samples_per_second": 3.925, | |
| "eval_steps_per_second": 0.131, | |
| "eval_wer": 30.418555051986136, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "learning_rate": 2.75e-07, | |
| "loss": 0.4629, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "learning_rate": 3e-07, | |
| "loss": 0.4567, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "learning_rate": 3.25e-07, | |
| "loss": 0.4397, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "learning_rate": 3.5e-07, | |
| "loss": 0.4314, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "learning_rate": 3.75e-07, | |
| "loss": 0.4165, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 53.0, | |
| "learning_rate": 4e-07, | |
| "loss": 0.4056, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "learning_rate": 4.2499999999999995e-07, | |
| "loss": 0.3921, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 59.0, | |
| "learning_rate": 4.5e-07, | |
| "loss": 0.3851, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 63.0, | |
| "learning_rate": 4.7499999999999995e-07, | |
| "loss": 0.375, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 66.0, | |
| "learning_rate": 5e-07, | |
| "loss": 0.3559, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 66.0, | |
| "eval_loss": 0.44173145294189453, | |
| "eval_runtime": 94.715, | |
| "eval_samples_per_second": 4.413, | |
| "eval_steps_per_second": 0.148, | |
| "eval_wer": 29.09090909090909, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 69.0, | |
| "learning_rate": 4.972222222222222e-07, | |
| "loss": 0.3491, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 73.0, | |
| "learning_rate": 4.944444444444445e-07, | |
| "loss": 0.3364, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "learning_rate": 4.916666666666666e-07, | |
| "loss": 0.3277, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 79.0, | |
| "learning_rate": 4.888888888888889e-07, | |
| "loss": 0.3178, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 83.0, | |
| "learning_rate": 4.861111111111111e-07, | |
| "loss": 0.3088, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 86.0, | |
| "learning_rate": 4.833333333333333e-07, | |
| "loss": 0.298, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 89.0, | |
| "learning_rate": 4.805555555555555e-07, | |
| "loss": 0.2904, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 93.0, | |
| "learning_rate": 4.777777777777778e-07, | |
| "loss": 0.2814, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 96.0, | |
| "learning_rate": 4.7499999999999995e-07, | |
| "loss": 0.2733, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 99.0, | |
| "learning_rate": 4.722222222222222e-07, | |
| "loss": 0.2655, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 99.0, | |
| "eval_loss": 0.43072518706321716, | |
| "eval_runtime": 93.9445, | |
| "eval_samples_per_second": 4.449, | |
| "eval_steps_per_second": 0.149, | |
| "eval_wer": 28.216475606504932, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 103.0, | |
| "learning_rate": 4.694444444444444e-07, | |
| "loss": 0.259, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 106.0, | |
| "learning_rate": 4.6666666666666666e-07, | |
| "loss": 0.2494, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 109.0, | |
| "learning_rate": 4.6388888888888886e-07, | |
| "loss": 0.2424, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 113.0, | |
| "learning_rate": 4.611111111111111e-07, | |
| "loss": 0.238, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 116.0, | |
| "learning_rate": 4.5833333333333327e-07, | |
| "loss": 0.2284, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 119.0, | |
| "learning_rate": 4.555555555555555e-07, | |
| "loss": 0.2216, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 123.0, | |
| "learning_rate": 4.527777777777778e-07, | |
| "loss": 0.2159, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 126.0, | |
| "learning_rate": 4.5e-07, | |
| "loss": 0.2114, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 129.0, | |
| "learning_rate": 4.4722222222222223e-07, | |
| "loss": 0.2032, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 133.0, | |
| "learning_rate": 4.444444444444444e-07, | |
| "loss": 0.1987, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 133.0, | |
| "eval_loss": 0.43504032492637634, | |
| "eval_runtime": 93.1688, | |
| "eval_samples_per_second": 4.486, | |
| "eval_steps_per_second": 0.15, | |
| "eval_wer": 27.832577979205546, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 136.0, | |
| "learning_rate": 4.4166666666666664e-07, | |
| "loss": 0.1914, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 139.0, | |
| "learning_rate": 4.3888888888888884e-07, | |
| "loss": 0.1879, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 143.0, | |
| "learning_rate": 4.361111111111111e-07, | |
| "loss": 0.1824, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 146.0, | |
| "learning_rate": 4.3333333333333335e-07, | |
| "loss": 0.1748, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 149.0, | |
| "learning_rate": 4.3055555555555555e-07, | |
| "loss": 0.1719, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 153.0, | |
| "learning_rate": 4.2777777777777775e-07, | |
| "loss": 0.1667, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 156.0, | |
| "learning_rate": 4.2499999999999995e-07, | |
| "loss": 0.1611, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 159.0, | |
| "learning_rate": 4.222222222222222e-07, | |
| "loss": 0.1553, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 163.0, | |
| "learning_rate": 4.194444444444444e-07, | |
| "loss": 0.1508, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 166.0, | |
| "learning_rate": 4.1666666666666667e-07, | |
| "loss": 0.1472, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 166.0, | |
| "eval_loss": 0.4467640519142151, | |
| "eval_runtime": 91.3178, | |
| "eval_samples_per_second": 4.577, | |
| "eval_steps_per_second": 0.153, | |
| "eval_wer": 27.72060783790989, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 169.0, | |
| "learning_rate": 4.1388888888888887e-07, | |
| "loss": 0.1425, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 173.0, | |
| "learning_rate": 4.1111111111111107e-07, | |
| "loss": 0.1378, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 176.0, | |
| "learning_rate": 4.083333333333333e-07, | |
| "loss": 0.1342, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 179.0, | |
| "learning_rate": 4.055555555555555e-07, | |
| "loss": 0.1294, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 183.0, | |
| "learning_rate": 4.027777777777778e-07, | |
| "loss": 0.1256, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 186.0, | |
| "learning_rate": 4e-07, | |
| "loss": 0.1219, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 189.0, | |
| "learning_rate": 3.972222222222222e-07, | |
| "loss": 0.1175, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 193.0, | |
| "learning_rate": 3.9444444444444444e-07, | |
| "loss": 0.1145, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 196.0, | |
| "learning_rate": 3.9166666666666664e-07, | |
| "loss": 0.1105, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 199.0, | |
| "learning_rate": 3.888888888888889e-07, | |
| "loss": 0.1061, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 199.0, | |
| "eval_loss": 0.46401092410087585, | |
| "eval_runtime": 92.2149, | |
| "eval_samples_per_second": 4.533, | |
| "eval_steps_per_second": 0.152, | |
| "eval_wer": 28.09917355371901, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 203.0, | |
| "learning_rate": 3.861111111111111e-07, | |
| "loss": 0.1033, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 206.0, | |
| "learning_rate": 3.8333333333333335e-07, | |
| "loss": 0.0998, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 209.0, | |
| "learning_rate": 3.805555555555555e-07, | |
| "loss": 0.0967, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 213.0, | |
| "learning_rate": 3.7777777777777775e-07, | |
| "loss": 0.0937, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 216.0, | |
| "learning_rate": 3.75e-07, | |
| "loss": 0.0894, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 219.0, | |
| "learning_rate": 3.722222222222222e-07, | |
| "loss": 0.0878, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 223.0, | |
| "learning_rate": 3.6944444444444447e-07, | |
| "loss": 0.0842, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 226.0, | |
| "learning_rate": 3.666666666666666e-07, | |
| "loss": 0.0822, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 229.0, | |
| "learning_rate": 3.6388888888888887e-07, | |
| "loss": 0.0784, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 233.0, | |
| "learning_rate": 3.6111111111111107e-07, | |
| "loss": 0.0767, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 233.0, | |
| "eval_loss": 0.48346972465515137, | |
| "eval_runtime": 94.3586, | |
| "eval_samples_per_second": 4.43, | |
| "eval_steps_per_second": 0.148, | |
| "eval_wer": 28.573713676352973, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 236.0, | |
| "learning_rate": 3.583333333333333e-07, | |
| "loss": 0.0734, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 239.0, | |
| "learning_rate": 3.5555555555555553e-07, | |
| "loss": 0.071, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 243.0, | |
| "learning_rate": 3.527777777777778e-07, | |
| "loss": 0.0687, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 246.0, | |
| "learning_rate": 3.5e-07, | |
| "loss": 0.0665, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 249.0, | |
| "learning_rate": 3.472222222222222e-07, | |
| "loss": 0.0638, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 253.0, | |
| "learning_rate": 3.4444444444444444e-07, | |
| "loss": 0.0617, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 256.0, | |
| "learning_rate": 3.4166666666666664e-07, | |
| "loss": 0.0599, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 259.0, | |
| "learning_rate": 3.388888888888889e-07, | |
| "loss": 0.0575, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 263.0, | |
| "learning_rate": 3.361111111111111e-07, | |
| "loss": 0.0562, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 266.0, | |
| "learning_rate": 3.333333333333333e-07, | |
| "loss": 0.0541, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 266.0, | |
| "eval_loss": 0.5032361745834351, | |
| "eval_runtime": 93.8219, | |
| "eval_samples_per_second": 4.455, | |
| "eval_steps_per_second": 0.149, | |
| "eval_wer": 28.685683817648627, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 269.0, | |
| "learning_rate": 3.3055555555555556e-07, | |
| "loss": 0.0518, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 273.0, | |
| "learning_rate": 3.2777777777777776e-07, | |
| "loss": 0.0508, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 276.0, | |
| "learning_rate": 3.25e-07, | |
| "loss": 0.0491, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 279.0, | |
| "learning_rate": 3.222222222222222e-07, | |
| "loss": 0.0475, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 283.0, | |
| "learning_rate": 3.194444444444444e-07, | |
| "loss": 0.046, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 286.0, | |
| "learning_rate": 3.166666666666666e-07, | |
| "loss": 0.045, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 289.0, | |
| "learning_rate": 3.1388888888888887e-07, | |
| "loss": 0.0432, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 293.0, | |
| "learning_rate": 3.111111111111111e-07, | |
| "loss": 0.042, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 296.0, | |
| "learning_rate": 3.0833333333333333e-07, | |
| "loss": 0.0408, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 299.0, | |
| "learning_rate": 3.055555555555556e-07, | |
| "loss": 0.0396, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 299.0, | |
| "eval_loss": 0.5201818943023682, | |
| "eval_runtime": 93.6995, | |
| "eval_samples_per_second": 4.461, | |
| "eval_steps_per_second": 0.149, | |
| "eval_wer": 28.7763263129832, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 303.0, | |
| "learning_rate": 3.0277777777777773e-07, | |
| "loss": 0.0381, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 306.0, | |
| "learning_rate": 3e-07, | |
| "loss": 0.0374, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 309.0, | |
| "learning_rate": 2.972222222222222e-07, | |
| "loss": 0.0365, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 313.0, | |
| "learning_rate": 2.9444444444444444e-07, | |
| "loss": 0.0351, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 316.0, | |
| "learning_rate": 2.916666666666667e-07, | |
| "loss": 0.0344, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 319.0, | |
| "learning_rate": 2.8888888888888885e-07, | |
| "loss": 0.0334, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 323.0, | |
| "learning_rate": 2.861111111111111e-07, | |
| "loss": 0.0326, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 326.0, | |
| "learning_rate": 2.833333333333333e-07, | |
| "loss": 0.0317, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 329.0, | |
| "learning_rate": 2.8055555555555556e-07, | |
| "loss": 0.0308, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 333.0, | |
| "learning_rate": 2.7777777777777776e-07, | |
| "loss": 0.03, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 333.0, | |
| "eval_loss": 0.5353468060493469, | |
| "eval_runtime": 92.1858, | |
| "eval_samples_per_second": 4.534, | |
| "eval_steps_per_second": 0.152, | |
| "eval_wer": 29.202879232204747, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 336.0, | |
| "learning_rate": 2.75e-07, | |
| "loss": 0.0294, | |
| "step": 2525 | |
| }, | |
| { | |
| "epoch": 339.0, | |
| "learning_rate": 2.7222222222222216e-07, | |
| "loss": 0.0287, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 343.0, | |
| "learning_rate": 2.694444444444444e-07, | |
| "loss": 0.0279, | |
| "step": 2575 | |
| }, | |
| { | |
| "epoch": 346.0, | |
| "learning_rate": 2.6666666666666667e-07, | |
| "loss": 0.0273, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 349.0, | |
| "learning_rate": 2.638888888888889e-07, | |
| "loss": 0.0267, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 353.0, | |
| "learning_rate": 2.6111111111111113e-07, | |
| "loss": 0.0261, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 356.0, | |
| "learning_rate": 2.5833333333333333e-07, | |
| "loss": 0.0253, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 359.0, | |
| "learning_rate": 2.5555555555555553e-07, | |
| "loss": 0.0249, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 363.0, | |
| "learning_rate": 2.5277777777777773e-07, | |
| "loss": 0.0245, | |
| "step": 2725 | |
| }, | |
| { | |
| "epoch": 366.0, | |
| "learning_rate": 2.5e-07, | |
| "loss": 0.0237, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 366.0, | |
| "eval_loss": 0.5479018688201904, | |
| "eval_runtime": 91.5542, | |
| "eval_samples_per_second": 4.566, | |
| "eval_steps_per_second": 0.153, | |
| "eval_wer": 28.909624100239935, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 369.0, | |
| "learning_rate": 2.4722222222222224e-07, | |
| "loss": 0.0233, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 373.0, | |
| "learning_rate": 2.4444444444444445e-07, | |
| "loss": 0.0228, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 376.0, | |
| "learning_rate": 2.4166666666666665e-07, | |
| "loss": 0.0223, | |
| "step": 2825 | |
| }, | |
| { | |
| "epoch": 379.0, | |
| "learning_rate": 2.388888888888889e-07, | |
| "loss": 0.022, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 383.0, | |
| "learning_rate": 2.361111111111111e-07, | |
| "loss": 0.0216, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 386.0, | |
| "learning_rate": 2.3333333333333333e-07, | |
| "loss": 0.0207, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 389.0, | |
| "learning_rate": 2.3055555555555556e-07, | |
| "loss": 0.0206, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 393.0, | |
| "learning_rate": 2.2777777777777776e-07, | |
| "loss": 0.0201, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 396.0, | |
| "learning_rate": 2.25e-07, | |
| "loss": 0.0198, | |
| "step": 2975 | |
| }, | |
| { | |
| "epoch": 399.0, | |
| "learning_rate": 2.222222222222222e-07, | |
| "loss": 0.0195, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 399.0, | |
| "eval_loss": 0.5586678981781006, | |
| "eval_runtime": 92.6086, | |
| "eval_samples_per_second": 4.514, | |
| "eval_steps_per_second": 0.151, | |
| "eval_wer": 28.909624100239935, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 403.0, | |
| "learning_rate": 2.1944444444444442e-07, | |
| "loss": 0.0191, | |
| "step": 3025 | |
| }, | |
| { | |
| "epoch": 406.0, | |
| "learning_rate": 2.1666666666666667e-07, | |
| "loss": 0.0186, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 409.0, | |
| "learning_rate": 2.1388888888888888e-07, | |
| "loss": 0.0185, | |
| "step": 3075 | |
| }, | |
| { | |
| "epoch": 413.0, | |
| "learning_rate": 2.111111111111111e-07, | |
| "loss": 0.0181, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 416.0, | |
| "learning_rate": 2.0833333333333333e-07, | |
| "loss": 0.0177, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 419.0, | |
| "learning_rate": 2.0555555555555553e-07, | |
| "loss": 0.0174, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 423.0, | |
| "learning_rate": 2.0277777777777776e-07, | |
| "loss": 0.0172, | |
| "step": 3175 | |
| }, | |
| { | |
| "epoch": 426.0, | |
| "learning_rate": 2e-07, | |
| "loss": 0.0168, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 429.0, | |
| "learning_rate": 1.9722222222222222e-07, | |
| "loss": 0.0166, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 433.0, | |
| "learning_rate": 1.9444444444444445e-07, | |
| "loss": 0.0163, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 433.0, | |
| "eval_loss": 0.5682737231254578, | |
| "eval_runtime": 92.9945, | |
| "eval_samples_per_second": 4.495, | |
| "eval_steps_per_second": 0.151, | |
| "eval_wer": 28.94694748067182, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 436.0, | |
| "learning_rate": 1.9166666666666668e-07, | |
| "loss": 0.0159, | |
| "step": 3275 | |
| }, | |
| { | |
| "epoch": 439.0, | |
| "learning_rate": 1.8888888888888888e-07, | |
| "loss": 0.0158, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 443.0, | |
| "learning_rate": 1.861111111111111e-07, | |
| "loss": 0.0155, | |
| "step": 3325 | |
| }, | |
| { | |
| "epoch": 446.0, | |
| "learning_rate": 1.833333333333333e-07, | |
| "loss": 0.0153, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 449.0, | |
| "learning_rate": 1.8055555555555554e-07, | |
| "loss": 0.015, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 453.0, | |
| "learning_rate": 1.7777777777777776e-07, | |
| "loss": 0.0149, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 456.0, | |
| "learning_rate": 1.75e-07, | |
| "loss": 0.0146, | |
| "step": 3425 | |
| }, | |
| { | |
| "epoch": 459.0, | |
| "learning_rate": 1.7222222222222222e-07, | |
| "loss": 0.0143, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 463.0, | |
| "learning_rate": 1.6944444444444445e-07, | |
| "loss": 0.0141, | |
| "step": 3475 | |
| }, | |
| { | |
| "epoch": 466.0, | |
| "learning_rate": 1.6666666666666665e-07, | |
| "loss": 0.014, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 466.0, | |
| "eval_loss": 0.5766630172729492, | |
| "eval_runtime": 92.3581, | |
| "eval_samples_per_second": 4.526, | |
| "eval_steps_per_second": 0.152, | |
| "eval_wer": 29.133564382831246, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 469.0, | |
| "learning_rate": 1.6388888888888888e-07, | |
| "loss": 0.0138, | |
| "step": 3525 | |
| }, | |
| { | |
| "epoch": 473.0, | |
| "learning_rate": 1.611111111111111e-07, | |
| "loss": 0.0137, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 476.0, | |
| "learning_rate": 1.583333333333333e-07, | |
| "loss": 0.0133, | |
| "step": 3575 | |
| }, | |
| { | |
| "epoch": 479.0, | |
| "learning_rate": 1.5555555555555556e-07, | |
| "loss": 0.0132, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 483.0, | |
| "learning_rate": 1.527777777777778e-07, | |
| "loss": 0.013, | |
| "step": 3625 | |
| }, | |
| { | |
| "epoch": 486.0, | |
| "learning_rate": 1.5e-07, | |
| "loss": 0.0128, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 489.0, | |
| "learning_rate": 1.4722222222222222e-07, | |
| "loss": 0.0126, | |
| "step": 3675 | |
| }, | |
| { | |
| "epoch": 493.0, | |
| "learning_rate": 1.4444444444444442e-07, | |
| "loss": 0.0125, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 496.0, | |
| "learning_rate": 1.4166666666666665e-07, | |
| "loss": 0.0123, | |
| "step": 3725 | |
| }, | |
| { | |
| "epoch": 499.0, | |
| "learning_rate": 1.3888888888888888e-07, | |
| "loss": 0.0121, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 499.0, | |
| "eval_loss": 0.5837976336479187, | |
| "eval_runtime": 93.0244, | |
| "eval_samples_per_second": 4.493, | |
| "eval_steps_per_second": 0.15, | |
| "eval_wer": 29.341508930951747, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 503.0, | |
| "learning_rate": 1.3611111111111108e-07, | |
| "loss": 0.012, | |
| "step": 3775 | |
| }, | |
| { | |
| "epoch": 506.0, | |
| "learning_rate": 1.3333333333333334e-07, | |
| "loss": 0.0117, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 509.0, | |
| "learning_rate": 1.3055555555555556e-07, | |
| "loss": 0.0117, | |
| "step": 3825 | |
| }, | |
| { | |
| "epoch": 513.0, | |
| "learning_rate": 1.2777777777777777e-07, | |
| "loss": 0.0115, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 516.0, | |
| "learning_rate": 1.25e-07, | |
| "loss": 0.0114, | |
| "step": 3875 | |
| }, | |
| { | |
| "epoch": 519.0, | |
| "learning_rate": 1.2222222222222222e-07, | |
| "loss": 0.0113, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 523.0, | |
| "learning_rate": 1.1944444444444445e-07, | |
| "loss": 0.0112, | |
| "step": 3925 | |
| }, | |
| { | |
| "epoch": 526.0, | |
| "learning_rate": 1.1666666666666667e-07, | |
| "loss": 0.011, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 529.0, | |
| "learning_rate": 1.1388888888888888e-07, | |
| "loss": 0.011, | |
| "step": 3975 | |
| }, | |
| { | |
| "epoch": 533.0, | |
| "learning_rate": 1.111111111111111e-07, | |
| "loss": 0.0108, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 533.0, | |
| "eval_loss": 0.5900186896324158, | |
| "eval_runtime": 92.7674, | |
| "eval_samples_per_second": 4.506, | |
| "eval_steps_per_second": 0.151, | |
| "eval_wer": 29.277525993068515, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 536.0, | |
| "learning_rate": 1.0833333333333334e-07, | |
| "loss": 0.0108, | |
| "step": 4025 | |
| }, | |
| { | |
| "epoch": 539.0, | |
| "learning_rate": 1.0555555555555555e-07, | |
| "loss": 0.0106, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 543.0, | |
| "learning_rate": 1.0277777777777777e-07, | |
| "loss": 0.0104, | |
| "step": 4075 | |
| }, | |
| { | |
| "epoch": 546.0, | |
| "learning_rate": 1e-07, | |
| "loss": 0.0105, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 549.0, | |
| "learning_rate": 9.722222222222222e-08, | |
| "loss": 0.0103, | |
| "step": 4125 | |
| }, | |
| { | |
| "epoch": 553.0, | |
| "learning_rate": 9.444444444444444e-08, | |
| "loss": 0.0102, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 556.0, | |
| "learning_rate": 9.166666666666665e-08, | |
| "loss": 0.0101, | |
| "step": 4175 | |
| }, | |
| { | |
| "epoch": 559.0, | |
| "learning_rate": 8.888888888888888e-08, | |
| "loss": 0.0101, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 563.0, | |
| "learning_rate": 8.611111111111111e-08, | |
| "loss": 0.01, | |
| "step": 4225 | |
| }, | |
| { | |
| "epoch": 566.0, | |
| "learning_rate": 8.333333333333333e-08, | |
| "loss": 0.01, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 566.0, | |
| "eval_loss": 0.5951334834098816, | |
| "eval_runtime": 92.8599, | |
| "eval_samples_per_second": 4.501, | |
| "eval_steps_per_second": 0.151, | |
| "eval_wer": 29.60810450546521, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 569.0, | |
| "learning_rate": 8.055555555555555e-08, | |
| "loss": 0.0098, | |
| "step": 4275 | |
| }, | |
| { | |
| "epoch": 573.0, | |
| "learning_rate": 7.777777777777778e-08, | |
| "loss": 0.0097, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 576.0, | |
| "learning_rate": 7.5e-08, | |
| "loss": 0.0097, | |
| "step": 4325 | |
| }, | |
| { | |
| "epoch": 579.0, | |
| "learning_rate": 7.222222222222221e-08, | |
| "loss": 0.0096, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 583.0, | |
| "learning_rate": 6.944444444444444e-08, | |
| "loss": 0.0095, | |
| "step": 4375 | |
| }, | |
| { | |
| "epoch": 586.0, | |
| "learning_rate": 6.666666666666667e-08, | |
| "loss": 0.0095, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 589.0, | |
| "learning_rate": 6.388888888888888e-08, | |
| "loss": 0.0095, | |
| "step": 4425 | |
| }, | |
| { | |
| "epoch": 593.0, | |
| "learning_rate": 6.111111111111111e-08, | |
| "loss": 0.0094, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 596.0, | |
| "learning_rate": 5.833333333333333e-08, | |
| "loss": 0.0093, | |
| "step": 4475 | |
| }, | |
| { | |
| "epoch": 599.0, | |
| "learning_rate": 5.555555555555555e-08, | |
| "loss": 0.0093, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 599.0, | |
| "eval_loss": 0.5988227128982544, | |
| "eval_runtime": 93.1273, | |
| "eval_samples_per_second": 4.488, | |
| "eval_steps_per_second": 0.15, | |
| "eval_wer": 29.485470541189017, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 603.0, | |
| "learning_rate": 5.2777777777777776e-08, | |
| "loss": 0.0092, | |
| "step": 4525 | |
| }, | |
| { | |
| "epoch": 606.0, | |
| "learning_rate": 5e-08, | |
| "loss": 0.0092, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 609.0, | |
| "learning_rate": 4.722222222222222e-08, | |
| "loss": 0.0091, | |
| "step": 4575 | |
| }, | |
| { | |
| "epoch": 613.0, | |
| "learning_rate": 4.444444444444444e-08, | |
| "loss": 0.0091, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 616.0, | |
| "learning_rate": 4.166666666666666e-08, | |
| "loss": 0.0091, | |
| "step": 4625 | |
| }, | |
| { | |
| "epoch": 619.0, | |
| "learning_rate": 3.888888888888889e-08, | |
| "loss": 0.009, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 623.0, | |
| "learning_rate": 3.6111111111111106e-08, | |
| "loss": 0.009, | |
| "step": 4675 | |
| }, | |
| { | |
| "epoch": 626.0, | |
| "learning_rate": 3.3333333333333334e-08, | |
| "loss": 0.009, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 629.0, | |
| "learning_rate": 3.0555555555555556e-08, | |
| "loss": 0.0089, | |
| "step": 4725 | |
| }, | |
| { | |
| "epoch": 633.0, | |
| "learning_rate": 2.7777777777777774e-08, | |
| "loss": 0.0088, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 633.0, | |
| "eval_loss": 0.6012450456619263, | |
| "eval_runtime": 91.6852, | |
| "eval_samples_per_second": 4.559, | |
| "eval_steps_per_second": 0.153, | |
| "eval_wer": 29.52812583311117, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 636.0, | |
| "learning_rate": 2.5e-08, | |
| "loss": 0.0089, | |
| "step": 4775 | |
| }, | |
| { | |
| "epoch": 639.0, | |
| "learning_rate": 2.222222222222222e-08, | |
| "loss": 0.0088, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 643.0, | |
| "learning_rate": 1.9444444444444445e-08, | |
| "loss": 0.0088, | |
| "step": 4825 | |
| }, | |
| { | |
| "epoch": 646.0, | |
| "learning_rate": 1.6666666666666667e-08, | |
| "loss": 0.0088, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 649.0, | |
| "learning_rate": 1.3888888888888887e-08, | |
| "loss": 0.0088, | |
| "step": 4875 | |
| }, | |
| { | |
| "epoch": 653.0, | |
| "learning_rate": 1.111111111111111e-08, | |
| "loss": 0.0088, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 656.0, | |
| "learning_rate": 8.333333333333334e-09, | |
| "loss": 0.0088, | |
| "step": 4925 | |
| }, | |
| { | |
| "epoch": 659.0, | |
| "learning_rate": 5.555555555555555e-09, | |
| "loss": 0.0087, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 663.0, | |
| "learning_rate": 2.7777777777777776e-09, | |
| "loss": 0.0088, | |
| "step": 4975 | |
| }, | |
| { | |
| "epoch": 666.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.0087, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 666.0, | |
| "eval_loss": 0.6019512414932251, | |
| "eval_runtime": 92.0662, | |
| "eval_samples_per_second": 4.54, | |
| "eval_steps_per_second": 0.152, | |
| "eval_wer": 29.426819514796055, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 666.0, | |
| "step": 5000, | |
| "total_flos": 2.062560849887232e+19, | |
| "train_loss": 0.10753952994346619, | |
| "train_runtime": 22112.1896, | |
| "train_samples_per_second": 14.472, | |
| "train_steps_per_second": 0.226 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 250, | |
| "total_flos": 2.062560849887232e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |