| { | |
| "best_metric": 44.287968920723564, | |
| "best_model_checkpoint": "./checkpoint-3000", | |
| "epoch": 526.3157894736842, | |
| "eval_steps": 500, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 2.4e-08, | |
| "loss": 0.8932, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 5.26, | |
| "learning_rate": 4.9e-08, | |
| "loss": 0.8854, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 7.89, | |
| "learning_rate": 7.399999999999999e-08, | |
| "loss": 0.8644, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 10.53, | |
| "learning_rate": 9.9e-08, | |
| "loss": 0.8475, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 13.16, | |
| "learning_rate": 1.24e-07, | |
| "loss": 0.8335, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 15.79, | |
| "learning_rate": 1.49e-07, | |
| "loss": 0.8167, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 18.42, | |
| "learning_rate": 1.7399999999999997e-07, | |
| "loss": 0.7933, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 21.05, | |
| "learning_rate": 1.99e-07, | |
| "loss": 0.7764, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 23.68, | |
| "learning_rate": 2.24e-07, | |
| "loss": 0.7498, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 26.32, | |
| "learning_rate": 2.4899999999999997e-07, | |
| "loss": 0.7433, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 28.95, | |
| "learning_rate": 2.74e-07, | |
| "loss": 0.7188, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 31.58, | |
| "learning_rate": 2.9899999999999996e-07, | |
| "loss": 0.7038, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 34.21, | |
| "learning_rate": 3.24e-07, | |
| "loss": 0.6835, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 36.84, | |
| "learning_rate": 3.4899999999999996e-07, | |
| "loss": 0.6601, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 39.47, | |
| "learning_rate": 3.74e-07, | |
| "loss": 0.6519, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 42.11, | |
| "learning_rate": 3.99e-07, | |
| "loss": 0.6359, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 44.74, | |
| "learning_rate": 4.24e-07, | |
| "loss": 0.6136, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 47.37, | |
| "learning_rate": 4.49e-07, | |
| "loss": 0.605, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 4.7399999999999993e-07, | |
| "loss": 0.5872, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 52.63, | |
| "learning_rate": 4.99e-07, | |
| "loss": 0.5692, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 52.63, | |
| "eval_loss": 0.6252978444099426, | |
| "eval_runtime": 62.0235, | |
| "eval_samples_per_second": 6.901, | |
| "eval_steps_per_second": 0.226, | |
| "eval_wer": 54.789365060094696, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 55.26, | |
| "learning_rate": 4.973333333333333e-07, | |
| "loss": 0.5509, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 57.89, | |
| "learning_rate": 4.945555555555556e-07, | |
| "loss": 0.5416, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 60.53, | |
| "learning_rate": 4.917777777777777e-07, | |
| "loss": 0.5232, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 63.16, | |
| "learning_rate": 4.89e-07, | |
| "loss": 0.5112, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 65.79, | |
| "learning_rate": 4.862222222222222e-07, | |
| "loss": 0.495, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 68.42, | |
| "learning_rate": 4.834444444444444e-07, | |
| "loss": 0.4867, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 71.05, | |
| "learning_rate": 4.806666666666667e-07, | |
| "loss": 0.4673, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 73.68, | |
| "learning_rate": 4.778888888888888e-07, | |
| "loss": 0.4554, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 76.32, | |
| "learning_rate": 4.7511111111111106e-07, | |
| "loss": 0.4528, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 78.95, | |
| "learning_rate": 4.723333333333333e-07, | |
| "loss": 0.4346, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 81.58, | |
| "learning_rate": 4.695555555555555e-07, | |
| "loss": 0.4237, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 84.21, | |
| "learning_rate": 4.6677777777777777e-07, | |
| "loss": 0.4114, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 86.84, | |
| "learning_rate": 4.64e-07, | |
| "loss": 0.4071, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 89.47, | |
| "learning_rate": 4.612222222222222e-07, | |
| "loss": 0.3896, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 92.11, | |
| "learning_rate": 4.5844444444444443e-07, | |
| "loss": 0.3856, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 94.74, | |
| "learning_rate": 4.5566666666666663e-07, | |
| "loss": 0.3765, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 97.37, | |
| "learning_rate": 4.528888888888889e-07, | |
| "loss": 0.3642, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "learning_rate": 4.501111111111111e-07, | |
| "loss": 0.3536, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 102.63, | |
| "learning_rate": 4.4733333333333334e-07, | |
| "loss": 0.347, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 105.26, | |
| "learning_rate": 4.445555555555555e-07, | |
| "loss": 0.3447, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 105.26, | |
| "eval_loss": 0.6001198887825012, | |
| "eval_runtime": 54.8595, | |
| "eval_samples_per_second": 7.802, | |
| "eval_steps_per_second": 0.255, | |
| "eval_wer": 45.210634939905304, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 107.89, | |
| "learning_rate": 4.4177777777777775e-07, | |
| "loss": 0.3318, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 110.53, | |
| "learning_rate": 4.39e-07, | |
| "loss": 0.3221, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 113.16, | |
| "learning_rate": 4.362222222222222e-07, | |
| "loss": 0.3154, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 115.79, | |
| "learning_rate": 4.3344444444444446e-07, | |
| "loss": 0.3073, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 118.42, | |
| "learning_rate": 4.306666666666666e-07, | |
| "loss": 0.3016, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 121.05, | |
| "learning_rate": 4.2788888888888886e-07, | |
| "loss": 0.2947, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 123.68, | |
| "learning_rate": 4.2511111111111106e-07, | |
| "loss": 0.2861, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 126.32, | |
| "learning_rate": 4.223333333333333e-07, | |
| "loss": 0.2751, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 128.95, | |
| "learning_rate": 4.1955555555555557e-07, | |
| "loss": 0.2756, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 131.58, | |
| "learning_rate": 4.1677777777777777e-07, | |
| "loss": 0.2671, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 134.21, | |
| "learning_rate": 4.14e-07, | |
| "loss": 0.2597, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 136.84, | |
| "learning_rate": 4.112222222222222e-07, | |
| "loss": 0.2523, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 139.47, | |
| "learning_rate": 4.0844444444444443e-07, | |
| "loss": 0.2475, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 142.11, | |
| "learning_rate": 4.0566666666666663e-07, | |
| "loss": 0.2412, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 144.74, | |
| "learning_rate": 4.028888888888889e-07, | |
| "loss": 0.2316, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 147.37, | |
| "learning_rate": 4.001111111111111e-07, | |
| "loss": 0.2303, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 150.0, | |
| "learning_rate": 3.973333333333333e-07, | |
| "loss": 0.2278, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 152.63, | |
| "learning_rate": 3.9455555555555555e-07, | |
| "loss": 0.2176, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 155.26, | |
| "learning_rate": 3.9177777777777775e-07, | |
| "loss": 0.2152, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 157.89, | |
| "learning_rate": 3.89e-07, | |
| "loss": 0.2067, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 157.89, | |
| "eval_loss": 0.6108792424201965, | |
| "eval_runtime": 54.1831, | |
| "eval_samples_per_second": 7.899, | |
| "eval_steps_per_second": 0.258, | |
| "eval_wer": 44.737161587956784, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 160.53, | |
| "learning_rate": 3.862222222222222e-07, | |
| "loss": 0.2053, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 163.16, | |
| "learning_rate": 3.834444444444444e-07, | |
| "loss": 0.1996, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 165.79, | |
| "learning_rate": 3.8066666666666666e-07, | |
| "loss": 0.1913, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 168.42, | |
| "learning_rate": 3.7788888888888886e-07, | |
| "loss": 0.1899, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 171.05, | |
| "learning_rate": 3.751111111111111e-07, | |
| "loss": 0.1844, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 173.68, | |
| "learning_rate": 3.723333333333333e-07, | |
| "loss": 0.1813, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 176.32, | |
| "learning_rate": 3.695555555555556e-07, | |
| "loss": 0.1762, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 178.95, | |
| "learning_rate": 3.667777777777777e-07, | |
| "loss": 0.1717, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 181.58, | |
| "learning_rate": 3.64e-07, | |
| "loss": 0.1687, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 184.21, | |
| "learning_rate": 3.6122222222222223e-07, | |
| "loss": 0.1599, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 186.84, | |
| "learning_rate": 3.5844444444444443e-07, | |
| "loss": 0.16, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 189.47, | |
| "learning_rate": 3.556666666666667e-07, | |
| "loss": 0.1564, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 192.11, | |
| "learning_rate": 3.5288888888888884e-07, | |
| "loss": 0.1521, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 194.74, | |
| "learning_rate": 3.501111111111111e-07, | |
| "loss": 0.1476, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 197.37, | |
| "learning_rate": 3.473333333333333e-07, | |
| "loss": 0.1448, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 200.0, | |
| "learning_rate": 3.4455555555555555e-07, | |
| "loss": 0.1394, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 202.63, | |
| "learning_rate": 3.4177777777777775e-07, | |
| "loss": 0.1372, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 205.26, | |
| "learning_rate": 3.39e-07, | |
| "loss": 0.1345, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 207.89, | |
| "learning_rate": 3.362222222222222e-07, | |
| "loss": 0.1302, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 210.53, | |
| "learning_rate": 3.334444444444444e-07, | |
| "loss": 0.1273, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 210.53, | |
| "eval_loss": 0.6303355097770691, | |
| "eval_runtime": 54.9028, | |
| "eval_samples_per_second": 7.796, | |
| "eval_steps_per_second": 0.255, | |
| "eval_wer": 44.737161587956784, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 213.16, | |
| "learning_rate": 3.3066666666666666e-07, | |
| "loss": 0.1247, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 215.79, | |
| "learning_rate": 3.2788888888888886e-07, | |
| "loss": 0.1207, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 218.42, | |
| "learning_rate": 3.251111111111111e-07, | |
| "loss": 0.1177, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 221.05, | |
| "learning_rate": 3.223333333333333e-07, | |
| "loss": 0.1152, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 223.68, | |
| "learning_rate": 3.195555555555555e-07, | |
| "loss": 0.1132, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 226.32, | |
| "learning_rate": 3.167777777777778e-07, | |
| "loss": 0.1107, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 228.95, | |
| "learning_rate": 3.14e-07, | |
| "loss": 0.1066, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 231.58, | |
| "learning_rate": 3.1122222222222223e-07, | |
| "loss": 0.106, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 234.21, | |
| "learning_rate": 3.0844444444444444e-07, | |
| "loss": 0.0998, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 236.84, | |
| "learning_rate": 3.0566666666666664e-07, | |
| "loss": 0.1006, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 239.47, | |
| "learning_rate": 3.0288888888888884e-07, | |
| "loss": 0.0972, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 242.11, | |
| "learning_rate": 3.001111111111111e-07, | |
| "loss": 0.094, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 244.74, | |
| "learning_rate": 2.9733333333333335e-07, | |
| "loss": 0.093, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 247.37, | |
| "learning_rate": 2.9455555555555555e-07, | |
| "loss": 0.0898, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 250.0, | |
| "learning_rate": 2.917777777777778e-07, | |
| "loss": 0.0887, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 252.63, | |
| "learning_rate": 2.8899999999999995e-07, | |
| "loss": 0.0864, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 255.26, | |
| "learning_rate": 2.862222222222222e-07, | |
| "loss": 0.0832, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 257.89, | |
| "learning_rate": 2.834444444444444e-07, | |
| "loss": 0.0835, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 260.53, | |
| "learning_rate": 2.8066666666666667e-07, | |
| "loss": 0.0807, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 263.16, | |
| "learning_rate": 2.778888888888889e-07, | |
| "loss": 0.0788, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 263.16, | |
| "eval_loss": 0.6507540345191956, | |
| "eval_runtime": 51.9133, | |
| "eval_samples_per_second": 8.245, | |
| "eval_steps_per_second": 0.27, | |
| "eval_wer": 44.45793371373073, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 265.79, | |
| "learning_rate": 2.7511111111111107e-07, | |
| "loss": 0.0767, | |
| "step": 2525 | |
| }, | |
| { | |
| "epoch": 268.42, | |
| "learning_rate": 2.723333333333333e-07, | |
| "loss": 0.0749, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 271.05, | |
| "learning_rate": 2.695555555555555e-07, | |
| "loss": 0.0735, | |
| "step": 2575 | |
| }, | |
| { | |
| "epoch": 273.68, | |
| "learning_rate": 2.667777777777778e-07, | |
| "loss": 0.0712, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 276.32, | |
| "learning_rate": 2.64e-07, | |
| "loss": 0.0704, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 278.95, | |
| "learning_rate": 2.6122222222222224e-07, | |
| "loss": 0.0693, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 281.58, | |
| "learning_rate": 2.584444444444444e-07, | |
| "loss": 0.0671, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 284.21, | |
| "learning_rate": 2.5566666666666664e-07, | |
| "loss": 0.0654, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 286.84, | |
| "learning_rate": 2.528888888888889e-07, | |
| "loss": 0.0658, | |
| "step": 2725 | |
| }, | |
| { | |
| "epoch": 289.47, | |
| "learning_rate": 2.501111111111111e-07, | |
| "loss": 0.0628, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 292.11, | |
| "learning_rate": 2.473333333333333e-07, | |
| "loss": 0.0615, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 294.74, | |
| "learning_rate": 2.4455555555555555e-07, | |
| "loss": 0.0613, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 297.37, | |
| "learning_rate": 2.4177777777777775e-07, | |
| "loss": 0.0595, | |
| "step": 2825 | |
| }, | |
| { | |
| "epoch": 300.0, | |
| "learning_rate": 2.3899999999999996e-07, | |
| "loss": 0.0585, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 302.63, | |
| "learning_rate": 2.362222222222222e-07, | |
| "loss": 0.0573, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 305.26, | |
| "learning_rate": 2.3344444444444444e-07, | |
| "loss": 0.0567, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 307.89, | |
| "learning_rate": 2.3066666666666664e-07, | |
| "loss": 0.055, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 310.53, | |
| "learning_rate": 2.2788888888888887e-07, | |
| "loss": 0.0539, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 313.16, | |
| "learning_rate": 2.2511111111111112e-07, | |
| "loss": 0.0548, | |
| "step": 2975 | |
| }, | |
| { | |
| "epoch": 315.79, | |
| "learning_rate": 2.2233333333333333e-07, | |
| "loss": 0.0526, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 315.79, | |
| "eval_loss": 0.6669550538063049, | |
| "eval_runtime": 52.0245, | |
| "eval_samples_per_second": 8.227, | |
| "eval_steps_per_second": 0.269, | |
| "eval_wer": 44.287968920723564, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 318.42, | |
| "learning_rate": 2.1955555555555555e-07, | |
| "loss": 0.0523, | |
| "step": 3025 | |
| }, | |
| { | |
| "epoch": 321.05, | |
| "learning_rate": 2.1677777777777778e-07, | |
| "loss": 0.0509, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 323.68, | |
| "learning_rate": 2.1399999999999998e-07, | |
| "loss": 0.0498, | |
| "step": 3075 | |
| }, | |
| { | |
| "epoch": 326.32, | |
| "learning_rate": 2.112222222222222e-07, | |
| "loss": 0.0496, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 328.95, | |
| "learning_rate": 2.0844444444444441e-07, | |
| "loss": 0.0493, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 331.58, | |
| "learning_rate": 2.0566666666666664e-07, | |
| "loss": 0.0476, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 334.21, | |
| "learning_rate": 2.028888888888889e-07, | |
| "loss": 0.0477, | |
| "step": 3175 | |
| }, | |
| { | |
| "epoch": 336.84, | |
| "learning_rate": 2.001111111111111e-07, | |
| "loss": 0.0467, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 339.47, | |
| "learning_rate": 1.9733333333333333e-07, | |
| "loss": 0.046, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 342.11, | |
| "learning_rate": 1.9455555555555556e-07, | |
| "loss": 0.0457, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 344.74, | |
| "learning_rate": 1.9177777777777776e-07, | |
| "loss": 0.0451, | |
| "step": 3275 | |
| }, | |
| { | |
| "epoch": 347.37, | |
| "learning_rate": 1.8899999999999999e-07, | |
| "loss": 0.0441, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 350.0, | |
| "learning_rate": 1.8622222222222221e-07, | |
| "loss": 0.0436, | |
| "step": 3325 | |
| }, | |
| { | |
| "epoch": 352.63, | |
| "learning_rate": 1.8344444444444444e-07, | |
| "loss": 0.0431, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 355.26, | |
| "learning_rate": 1.8066666666666667e-07, | |
| "loss": 0.0427, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 357.89, | |
| "learning_rate": 1.778888888888889e-07, | |
| "loss": 0.0422, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 360.53, | |
| "learning_rate": 1.751111111111111e-07, | |
| "loss": 0.0416, | |
| "step": 3425 | |
| }, | |
| { | |
| "epoch": 363.16, | |
| "learning_rate": 1.7233333333333333e-07, | |
| "loss": 0.0411, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 365.79, | |
| "learning_rate": 1.6955555555555553e-07, | |
| "loss": 0.0402, | |
| "step": 3475 | |
| }, | |
| { | |
| "epoch": 368.42, | |
| "learning_rate": 1.6677777777777776e-07, | |
| "loss": 0.0404, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 368.42, | |
| "eval_loss": 0.6783625483512878, | |
| "eval_runtime": 51.7698, | |
| "eval_samples_per_second": 8.267, | |
| "eval_steps_per_second": 0.27, | |
| "eval_wer": 44.71288090324147, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 371.05, | |
| "learning_rate": 1.64e-07, | |
| "loss": 0.0396, | |
| "step": 3525 | |
| }, | |
| { | |
| "epoch": 373.68, | |
| "learning_rate": 1.6122222222222221e-07, | |
| "loss": 0.039, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 376.32, | |
| "learning_rate": 1.5844444444444444e-07, | |
| "loss": 0.039, | |
| "step": 3575 | |
| }, | |
| { | |
| "epoch": 378.95, | |
| "learning_rate": 1.5566666666666667e-07, | |
| "loss": 0.0384, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 381.58, | |
| "learning_rate": 1.5288888888888887e-07, | |
| "loss": 0.0377, | |
| "step": 3625 | |
| }, | |
| { | |
| "epoch": 384.21, | |
| "learning_rate": 1.501111111111111e-07, | |
| "loss": 0.0379, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 386.84, | |
| "learning_rate": 1.4733333333333333e-07, | |
| "loss": 0.037, | |
| "step": 3675 | |
| }, | |
| { | |
| "epoch": 389.47, | |
| "learning_rate": 1.4455555555555553e-07, | |
| "loss": 0.0366, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 392.11, | |
| "learning_rate": 1.4177777777777779e-07, | |
| "loss": 0.0365, | |
| "step": 3725 | |
| }, | |
| { | |
| "epoch": 394.74, | |
| "learning_rate": 1.3900000000000001e-07, | |
| "loss": 0.0361, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 397.37, | |
| "learning_rate": 1.3622222222222222e-07, | |
| "loss": 0.0358, | |
| "step": 3775 | |
| }, | |
| { | |
| "epoch": 400.0, | |
| "learning_rate": 1.3344444444444444e-07, | |
| "loss": 0.0356, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 402.63, | |
| "learning_rate": 1.3066666666666665e-07, | |
| "loss": 0.0352, | |
| "step": 3825 | |
| }, | |
| { | |
| "epoch": 405.26, | |
| "learning_rate": 1.2788888888888887e-07, | |
| "loss": 0.0349, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 407.89, | |
| "learning_rate": 1.251111111111111e-07, | |
| "loss": 0.0345, | |
| "step": 3875 | |
| }, | |
| { | |
| "epoch": 410.53, | |
| "learning_rate": 1.2233333333333333e-07, | |
| "loss": 0.0343, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 413.16, | |
| "learning_rate": 1.1955555555555556e-07, | |
| "loss": 0.0338, | |
| "step": 3925 | |
| }, | |
| { | |
| "epoch": 415.79, | |
| "learning_rate": 1.1677777777777776e-07, | |
| "loss": 0.0338, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 418.42, | |
| "learning_rate": 1.14e-07, | |
| "loss": 0.0333, | |
| "step": 3975 | |
| }, | |
| { | |
| "epoch": 421.05, | |
| "learning_rate": 1.1122222222222222e-07, | |
| "loss": 0.0335, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 421.05, | |
| "eval_loss": 0.6860424876213074, | |
| "eval_runtime": 54.8692, | |
| "eval_samples_per_second": 7.8, | |
| "eval_steps_per_second": 0.255, | |
| "eval_wer": 46.266844725021244, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 423.68, | |
| "learning_rate": 1.0844444444444443e-07, | |
| "loss": 0.0329, | |
| "step": 4025 | |
| }, | |
| { | |
| "epoch": 426.32, | |
| "learning_rate": 1.0566666666666667e-07, | |
| "loss": 0.0329, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 428.95, | |
| "learning_rate": 1.0288888888888889e-07, | |
| "loss": 0.0324, | |
| "step": 4075 | |
| }, | |
| { | |
| "epoch": 431.58, | |
| "learning_rate": 1.001111111111111e-07, | |
| "loss": 0.0324, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 434.21, | |
| "learning_rate": 9.733333333333332e-08, | |
| "loss": 0.0321, | |
| "step": 4125 | |
| }, | |
| { | |
| "epoch": 436.84, | |
| "learning_rate": 9.455555555555556e-08, | |
| "loss": 0.032, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 439.47, | |
| "learning_rate": 9.177777777777777e-08, | |
| "loss": 0.0316, | |
| "step": 4175 | |
| }, | |
| { | |
| "epoch": 442.11, | |
| "learning_rate": 8.899999999999999e-08, | |
| "loss": 0.0314, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 444.74, | |
| "learning_rate": 8.622222222222222e-08, | |
| "loss": 0.0313, | |
| "step": 4225 | |
| }, | |
| { | |
| "epoch": 447.37, | |
| "learning_rate": 8.344444444444445e-08, | |
| "loss": 0.0311, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 450.0, | |
| "learning_rate": 8.066666666666666e-08, | |
| "loss": 0.0311, | |
| "step": 4275 | |
| }, | |
| { | |
| "epoch": 452.63, | |
| "learning_rate": 7.788888888888888e-08, | |
| "loss": 0.031, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 455.26, | |
| "learning_rate": 7.511111111111112e-08, | |
| "loss": 0.0305, | |
| "step": 4325 | |
| }, | |
| { | |
| "epoch": 457.89, | |
| "learning_rate": 7.233333333333333e-08, | |
| "loss": 0.0305, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 460.53, | |
| "learning_rate": 6.955555555555555e-08, | |
| "loss": 0.0304, | |
| "step": 4375 | |
| }, | |
| { | |
| "epoch": 463.16, | |
| "learning_rate": 6.677777777777778e-08, | |
| "loss": 0.03, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 465.79, | |
| "learning_rate": 6.4e-08, | |
| "loss": 0.0302, | |
| "step": 4425 | |
| }, | |
| { | |
| "epoch": 468.42, | |
| "learning_rate": 6.122222222222222e-08, | |
| "loss": 0.03, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 471.05, | |
| "learning_rate": 5.844444444444444e-08, | |
| "loss": 0.03, | |
| "step": 4475 | |
| }, | |
| { | |
| "epoch": 473.68, | |
| "learning_rate": 5.566666666666667e-08, | |
| "loss": 0.0296, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 473.68, | |
| "eval_loss": 0.6907175779342651, | |
| "eval_runtime": 51.855, | |
| "eval_samples_per_second": 8.254, | |
| "eval_steps_per_second": 0.27, | |
| "eval_wer": 44.591477479664924, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 476.32, | |
| "learning_rate": 5.2888888888888884e-08, | |
| "loss": 0.0299, | |
| "step": 4525 | |
| }, | |
| { | |
| "epoch": 478.95, | |
| "learning_rate": 5.011111111111111e-08, | |
| "loss": 0.0294, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 481.58, | |
| "learning_rate": 4.733333333333333e-08, | |
| "loss": 0.0295, | |
| "step": 4575 | |
| }, | |
| { | |
| "epoch": 484.21, | |
| "learning_rate": 4.4555555555555555e-08, | |
| "loss": 0.0293, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 486.84, | |
| "learning_rate": 4.177777777777777e-08, | |
| "loss": 0.0292, | |
| "step": 4625 | |
| }, | |
| { | |
| "epoch": 489.47, | |
| "learning_rate": 3.9e-08, | |
| "loss": 0.0294, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 492.11, | |
| "learning_rate": 3.622222222222223e-08, | |
| "loss": 0.029, | |
| "step": 4675 | |
| }, | |
| { | |
| "epoch": 494.74, | |
| "learning_rate": 3.344444444444444e-08, | |
| "loss": 0.0291, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 497.37, | |
| "learning_rate": 3.0666666666666663e-08, | |
| "loss": 0.0288, | |
| "step": 4725 | |
| }, | |
| { | |
| "epoch": 500.0, | |
| "learning_rate": 2.7888888888888888e-08, | |
| "loss": 0.0289, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 502.63, | |
| "learning_rate": 2.511111111111111e-08, | |
| "loss": 0.0288, | |
| "step": 4775 | |
| }, | |
| { | |
| "epoch": 505.26, | |
| "learning_rate": 2.233333333333333e-08, | |
| "loss": 0.0289, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 507.89, | |
| "learning_rate": 1.9555555555555553e-08, | |
| "loss": 0.0286, | |
| "step": 4825 | |
| }, | |
| { | |
| "epoch": 510.53, | |
| "learning_rate": 1.6777777777777775e-08, | |
| "loss": 0.0285, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 513.16, | |
| "learning_rate": 1.4e-08, | |
| "loss": 0.0287, | |
| "step": 4875 | |
| }, | |
| { | |
| "epoch": 515.79, | |
| "learning_rate": 1.1222222222222221e-08, | |
| "loss": 0.0287, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 518.42, | |
| "learning_rate": 8.444444444444443e-09, | |
| "loss": 0.0287, | |
| "step": 4925 | |
| }, | |
| { | |
| "epoch": 521.05, | |
| "learning_rate": 5.666666666666667e-09, | |
| "loss": 0.0287, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 523.68, | |
| "learning_rate": 2.8888888888888886e-09, | |
| "loss": 0.0286, | |
| "step": 4975 | |
| }, | |
| { | |
| "epoch": 526.32, | |
| "learning_rate": 1.1111111111111111e-10, | |
| "loss": 0.0287, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 526.32, | |
| "eval_loss": 0.6924209594726562, | |
| "eval_runtime": 51.963, | |
| "eval_samples_per_second": 8.237, | |
| "eval_steps_per_second": 0.269, | |
| "eval_wer": 44.62789850673789, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 526.32, | |
| "step": 5000, | |
| "total_flos": 2.010698796957696e+19, | |
| "train_loss": 0.19006389085054398, | |
| "train_runtime": 13585.8827, | |
| "train_samples_per_second": 23.554, | |
| "train_steps_per_second": 0.368 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 556, | |
| "save_steps": 500, | |
| "total_flos": 2.010698796957696e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |