| { | |
| "best_metric": 31.03382013835511, | |
| "best_model_checkpoint": "./checkpoint-4500", | |
| "epoch": 434.0018, | |
| "eval_steps": 500, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 2.5e-08, | |
| "loss": 1.0639, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 5e-08, | |
| "loss": 1.0537, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 7.5e-08, | |
| "loss": 1.0365, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 1e-07, | |
| "loss": 1.0122, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 1.25e-07, | |
| "loss": 0.9715, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 1.5e-07, | |
| "loss": 0.9443, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 1.75e-07, | |
| "loss": 0.9048, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 2e-07, | |
| "loss": 0.8754, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 2.25e-07, | |
| "loss": 0.8426, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "learning_rate": 2.5e-07, | |
| "loss": 0.8109, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "learning_rate": 2.75e-07, | |
| "loss": 0.7894, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "learning_rate": 3e-07, | |
| "loss": 0.7598, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "learning_rate": 3.25e-07, | |
| "loss": 0.726, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "learning_rate": 3.5e-07, | |
| "loss": 0.7121, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "learning_rate": 3.75e-07, | |
| "loss": 0.6979, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "learning_rate": 4e-07, | |
| "loss": 0.6697, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "learning_rate": 4.2499999999999995e-07, | |
| "loss": 0.6394, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "learning_rate": 4.5e-07, | |
| "loss": 0.6277, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "learning_rate": 4.7499999999999995e-07, | |
| "loss": 0.6061, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "learning_rate": 5e-07, | |
| "loss": 0.5823, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_loss": 0.7964072227478027, | |
| "eval_runtime": 172.6955, | |
| "eval_samples_per_second": 4.962, | |
| "eval_steps_per_second": 0.156, | |
| "eval_wer": 37.89777094542659, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "learning_rate": 4.972222222222222e-07, | |
| "loss": 0.5686, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "learning_rate": 4.944444444444445e-07, | |
| "loss": 0.5445, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "learning_rate": 4.916666666666666e-07, | |
| "loss": 0.5319, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "learning_rate": 4.888888888888889e-07, | |
| "loss": 0.513, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "learning_rate": 4.861111111111111e-07, | |
| "loss": 0.5007, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "learning_rate": 4.833333333333333e-07, | |
| "loss": 0.485, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "learning_rate": 4.805555555555555e-07, | |
| "loss": 0.4672, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "learning_rate": 4.777777777777778e-07, | |
| "loss": 0.4558, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 63.0, | |
| "learning_rate": 4.7499999999999995e-07, | |
| "loss": 0.4432, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "learning_rate": 4.722222222222222e-07, | |
| "loss": 0.4323, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 67.0, | |
| "learning_rate": 4.694444444444444e-07, | |
| "loss": 0.4206, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 69.0, | |
| "learning_rate": 4.6666666666666666e-07, | |
| "loss": 0.4055, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 71.0, | |
| "learning_rate": 4.6388888888888886e-07, | |
| "loss": 0.3968, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 73.0, | |
| "learning_rate": 4.611111111111111e-07, | |
| "loss": 0.3866, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "learning_rate": 4.5833333333333327e-07, | |
| "loss": 0.3767, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 78.0, | |
| "learning_rate": 4.555555555555555e-07, | |
| "loss": 0.3668, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "learning_rate": 4.527777777777778e-07, | |
| "loss": 0.3582, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 82.0, | |
| "learning_rate": 4.5e-07, | |
| "loss": 0.3489, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "learning_rate": 4.4722222222222223e-07, | |
| "loss": 0.34, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 86.0, | |
| "learning_rate": 4.444444444444444e-07, | |
| "loss": 0.3312, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 86.0, | |
| "eval_loss": 0.699677586555481, | |
| "eval_runtime": 130.3854, | |
| "eval_samples_per_second": 6.573, | |
| "eval_steps_per_second": 0.207, | |
| "eval_wer": 33.71252882398155, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 89.0, | |
| "learning_rate": 4.4166666666666664e-07, | |
| "loss": 0.327, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 91.0, | |
| "learning_rate": 4.3888888888888884e-07, | |
| "loss": 0.3128, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 93.0, | |
| "learning_rate": 4.361111111111111e-07, | |
| "loss": 0.3051, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "learning_rate": 4.3333333333333335e-07, | |
| "loss": 0.299, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 97.0, | |
| "learning_rate": 4.3055555555555555e-07, | |
| "loss": 0.292, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 99.0, | |
| "learning_rate": 4.2777777777777775e-07, | |
| "loss": 0.2866, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 102.0, | |
| "learning_rate": 4.2499999999999995e-07, | |
| "loss": 0.2782, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 104.0, | |
| "learning_rate": 4.222222222222222e-07, | |
| "loss": 0.2734, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 106.0, | |
| "learning_rate": 4.194444444444444e-07, | |
| "loss": 0.2634, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 108.0, | |
| "learning_rate": 4.1666666666666667e-07, | |
| "loss": 0.2572, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 110.0, | |
| "learning_rate": 4.1388888888888887e-07, | |
| "loss": 0.2564, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 113.0, | |
| "learning_rate": 4.1111111111111107e-07, | |
| "loss": 0.2471, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 115.0, | |
| "learning_rate": 4.083333333333333e-07, | |
| "loss": 0.2396, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 117.0, | |
| "learning_rate": 4.055555555555555e-07, | |
| "loss": 0.2369, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 119.0, | |
| "learning_rate": 4.027777777777778e-07, | |
| "loss": 0.2268, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 121.0, | |
| "learning_rate": 4e-07, | |
| "loss": 0.2246, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 123.0, | |
| "learning_rate": 3.972222222222222e-07, | |
| "loss": 0.2165, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 126.0, | |
| "learning_rate": 3.9444444444444444e-07, | |
| "loss": 0.2154, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 128.0, | |
| "learning_rate": 3.9166666666666664e-07, | |
| "loss": 0.2101, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 130.0, | |
| "learning_rate": 3.888888888888889e-07, | |
| "loss": 0.2009, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 130.0, | |
| "eval_loss": 0.6783652305603027, | |
| "eval_runtime": 133.5638, | |
| "eval_samples_per_second": 6.416, | |
| "eval_steps_per_second": 0.202, | |
| "eval_wer": 32.74788624135281, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 132.0, | |
| "learning_rate": 3.861111111111111e-07, | |
| "loss": 0.1995, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 134.0, | |
| "learning_rate": 3.8333333333333335e-07, | |
| "loss": 0.1942, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 136.0, | |
| "learning_rate": 3.805555555555555e-07, | |
| "loss": 0.1884, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 139.0, | |
| "learning_rate": 3.7777777777777775e-07, | |
| "loss": 0.1841, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 141.0, | |
| "learning_rate": 3.75e-07, | |
| "loss": 0.1775, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 143.0, | |
| "learning_rate": 3.722222222222222e-07, | |
| "loss": 0.1785, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 145.0, | |
| "learning_rate": 3.6944444444444447e-07, | |
| "loss": 0.1708, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 147.0, | |
| "learning_rate": 3.666666666666666e-07, | |
| "loss": 0.1679, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 149.0, | |
| "learning_rate": 3.6388888888888887e-07, | |
| "loss": 0.1635, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 152.0, | |
| "learning_rate": 3.6111111111111107e-07, | |
| "loss": 0.1589, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 154.0, | |
| "learning_rate": 3.583333333333333e-07, | |
| "loss": 0.1577, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 156.0, | |
| "learning_rate": 3.5555555555555553e-07, | |
| "loss": 0.1531, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 158.0, | |
| "learning_rate": 3.527777777777778e-07, | |
| "loss": 0.1488, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 160.0, | |
| "learning_rate": 3.5e-07, | |
| "loss": 0.1473, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 163.0, | |
| "learning_rate": 3.472222222222222e-07, | |
| "loss": 0.1406, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 165.0, | |
| "learning_rate": 3.4444444444444444e-07, | |
| "loss": 0.1383, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 167.0, | |
| "learning_rate": 3.4166666666666664e-07, | |
| "loss": 0.1368, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 169.0, | |
| "learning_rate": 3.388888888888889e-07, | |
| "loss": 0.1337, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 171.0, | |
| "learning_rate": 3.361111111111111e-07, | |
| "loss": 0.1296, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 173.0, | |
| "learning_rate": 3.333333333333333e-07, | |
| "loss": 0.1271, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 173.0, | |
| "eval_loss": 0.6759907007217407, | |
| "eval_runtime": 127.3116, | |
| "eval_samples_per_second": 6.732, | |
| "eval_steps_per_second": 0.212, | |
| "eval_wer": 31.998462720983863, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 176.0, | |
| "learning_rate": 3.3055555555555556e-07, | |
| "loss": 0.1241, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 178.0, | |
| "learning_rate": 3.2777777777777776e-07, | |
| "loss": 0.1225, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 180.0, | |
| "learning_rate": 3.25e-07, | |
| "loss": 0.1175, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 182.0, | |
| "learning_rate": 3.222222222222222e-07, | |
| "loss": 0.1165, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 184.0, | |
| "learning_rate": 3.194444444444444e-07, | |
| "loss": 0.1121, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 186.0, | |
| "learning_rate": 3.166666666666666e-07, | |
| "loss": 0.1112, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 189.0, | |
| "learning_rate": 3.1388888888888887e-07, | |
| "loss": 0.109, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 191.0, | |
| "learning_rate": 3.111111111111111e-07, | |
| "loss": 0.1051, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 193.0, | |
| "learning_rate": 3.0833333333333333e-07, | |
| "loss": 0.1047, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 195.0, | |
| "learning_rate": 3.055555555555556e-07, | |
| "loss": 0.102, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 197.0, | |
| "learning_rate": 3.0277777777777773e-07, | |
| "loss": 0.098, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 199.0, | |
| "learning_rate": 3e-07, | |
| "loss": 0.0967, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 202.0, | |
| "learning_rate": 2.972222222222222e-07, | |
| "loss": 0.0953, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 204.0, | |
| "learning_rate": 2.9444444444444444e-07, | |
| "loss": 0.093, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 206.0, | |
| "learning_rate": 2.916666666666667e-07, | |
| "loss": 0.091, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 208.0, | |
| "learning_rate": 2.8888888888888885e-07, | |
| "loss": 0.0884, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 210.0, | |
| "learning_rate": 2.861111111111111e-07, | |
| "loss": 0.0871, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 213.0, | |
| "learning_rate": 2.833333333333333e-07, | |
| "loss": 0.0853, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 215.0, | |
| "learning_rate": 2.8055555555555556e-07, | |
| "loss": 0.0837, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 217.0, | |
| "learning_rate": 2.7777777777777776e-07, | |
| "loss": 0.0815, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 217.0, | |
| "eval_loss": 0.6799281239509583, | |
| "eval_runtime": 127.8527, | |
| "eval_samples_per_second": 6.703, | |
| "eval_steps_per_second": 0.211, | |
| "eval_wer": 31.30284396617986, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 219.0, | |
| "learning_rate": 2.75e-07, | |
| "loss": 0.0789, | |
| "step": 2525 | |
| }, | |
| { | |
| "epoch": 221.0, | |
| "learning_rate": 2.7222222222222216e-07, | |
| "loss": 0.0787, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 223.0, | |
| "learning_rate": 2.694444444444444e-07, | |
| "loss": 0.0768, | |
| "step": 2575 | |
| }, | |
| { | |
| "epoch": 226.0, | |
| "learning_rate": 2.6666666666666667e-07, | |
| "loss": 0.0744, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 228.0, | |
| "learning_rate": 2.638888888888889e-07, | |
| "loss": 0.0741, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 230.0, | |
| "learning_rate": 2.6111111111111113e-07, | |
| "loss": 0.0718, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 232.0, | |
| "learning_rate": 2.5833333333333333e-07, | |
| "loss": 0.0709, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 234.0, | |
| "learning_rate": 2.5555555555555553e-07, | |
| "loss": 0.0691, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 236.0, | |
| "learning_rate": 2.5277777777777773e-07, | |
| "loss": 0.068, | |
| "step": 2725 | |
| }, | |
| { | |
| "epoch": 239.0, | |
| "learning_rate": 2.5e-07, | |
| "loss": 0.0664, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 241.0, | |
| "learning_rate": 2.4722222222222224e-07, | |
| "loss": 0.0661, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 243.0, | |
| "learning_rate": 2.4444444444444445e-07, | |
| "loss": 0.065, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 245.0, | |
| "learning_rate": 2.4166666666666665e-07, | |
| "loss": 0.0633, | |
| "step": 2825 | |
| }, | |
| { | |
| "epoch": 247.0, | |
| "learning_rate": 2.388888888888889e-07, | |
| "loss": 0.0621, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 249.0, | |
| "learning_rate": 2.361111111111111e-07, | |
| "loss": 0.0615, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 252.0, | |
| "learning_rate": 2.3333333333333333e-07, | |
| "loss": 0.0609, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 254.0, | |
| "learning_rate": 2.3055555555555556e-07, | |
| "loss": 0.0586, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 256.0, | |
| "learning_rate": 2.2777777777777776e-07, | |
| "loss": 0.0586, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 258.0, | |
| "learning_rate": 2.25e-07, | |
| "loss": 0.0579, | |
| "step": 2975 | |
| }, | |
| { | |
| "epoch": 260.0, | |
| "learning_rate": 2.222222222222222e-07, | |
| "loss": 0.0561, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 260.0, | |
| "eval_loss": 0.6851306557655334, | |
| "eval_runtime": 128.2048, | |
| "eval_samples_per_second": 6.685, | |
| "eval_steps_per_second": 0.211, | |
| "eval_wer": 31.2336664104535, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 263.0, | |
| "learning_rate": 2.1944444444444442e-07, | |
| "loss": 0.0556, | |
| "step": 3025 | |
| }, | |
| { | |
| "epoch": 265.0, | |
| "learning_rate": 2.1666666666666667e-07, | |
| "loss": 0.0548, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 267.0, | |
| "learning_rate": 2.1388888888888888e-07, | |
| "loss": 0.0545, | |
| "step": 3075 | |
| }, | |
| { | |
| "epoch": 269.0, | |
| "learning_rate": 2.111111111111111e-07, | |
| "loss": 0.0532, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 271.0, | |
| "learning_rate": 2.0833333333333333e-07, | |
| "loss": 0.0522, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 273.0, | |
| "learning_rate": 2.0555555555555553e-07, | |
| "loss": 0.0523, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 276.0, | |
| "learning_rate": 2.0277777777777776e-07, | |
| "loss": 0.0513, | |
| "step": 3175 | |
| }, | |
| { | |
| "epoch": 278.0, | |
| "learning_rate": 2e-07, | |
| "loss": 0.0502, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 280.0, | |
| "learning_rate": 1.9722222222222222e-07, | |
| "loss": 0.0496, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 282.0, | |
| "learning_rate": 1.9444444444444445e-07, | |
| "loss": 0.0496, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 284.0, | |
| "learning_rate": 1.9166666666666668e-07, | |
| "loss": 0.0482, | |
| "step": 3275 | |
| }, | |
| { | |
| "epoch": 286.0, | |
| "learning_rate": 1.8888888888888888e-07, | |
| "loss": 0.0481, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 289.0, | |
| "learning_rate": 1.861111111111111e-07, | |
| "loss": 0.0472, | |
| "step": 3325 | |
| }, | |
| { | |
| "epoch": 291.0, | |
| "learning_rate": 1.833333333333333e-07, | |
| "loss": 0.0468, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 293.0, | |
| "learning_rate": 1.8055555555555554e-07, | |
| "loss": 0.0459, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 295.0, | |
| "learning_rate": 1.7777777777777776e-07, | |
| "loss": 0.0461, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 297.0, | |
| "learning_rate": 1.75e-07, | |
| "loss": 0.0449, | |
| "step": 3425 | |
| }, | |
| { | |
| "epoch": 299.0, | |
| "learning_rate": 1.7222222222222222e-07, | |
| "loss": 0.0448, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 302.0, | |
| "learning_rate": 1.6944444444444445e-07, | |
| "loss": 0.0441, | |
| "step": 3475 | |
| }, | |
| { | |
| "epoch": 304.0, | |
| "learning_rate": 1.6666666666666665e-07, | |
| "loss": 0.0438, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 304.0, | |
| "eval_loss": 0.6896069645881653, | |
| "eval_runtime": 127.6277, | |
| "eval_samples_per_second": 6.715, | |
| "eval_steps_per_second": 0.212, | |
| "eval_wer": 31.725595695618754, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 306.0, | |
| "learning_rate": 1.6388888888888888e-07, | |
| "loss": 0.0432, | |
| "step": 3525 | |
| }, | |
| { | |
| "epoch": 308.0, | |
| "learning_rate": 1.611111111111111e-07, | |
| "loss": 0.0425, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 310.0, | |
| "learning_rate": 1.583333333333333e-07, | |
| "loss": 0.0426, | |
| "step": 3575 | |
| }, | |
| { | |
| "epoch": 313.0, | |
| "learning_rate": 1.5555555555555556e-07, | |
| "loss": 0.0418, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 315.0, | |
| "learning_rate": 1.527777777777778e-07, | |
| "loss": 0.0417, | |
| "step": 3625 | |
| }, | |
| { | |
| "epoch": 317.0, | |
| "learning_rate": 1.5e-07, | |
| "loss": 0.0407, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 319.0, | |
| "learning_rate": 1.4722222222222222e-07, | |
| "loss": 0.0415, | |
| "step": 3675 | |
| }, | |
| { | |
| "epoch": 321.0, | |
| "learning_rate": 1.4444444444444442e-07, | |
| "loss": 0.04, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 323.0, | |
| "learning_rate": 1.4166666666666665e-07, | |
| "loss": 0.0399, | |
| "step": 3725 | |
| }, | |
| { | |
| "epoch": 326.0, | |
| "learning_rate": 1.3888888888888888e-07, | |
| "loss": 0.0399, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 328.0, | |
| "learning_rate": 1.3611111111111108e-07, | |
| "loss": 0.0392, | |
| "step": 3775 | |
| }, | |
| { | |
| "epoch": 330.0, | |
| "learning_rate": 1.3333333333333334e-07, | |
| "loss": 0.0387, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 332.0, | |
| "learning_rate": 1.3055555555555556e-07, | |
| "loss": 0.0391, | |
| "step": 3825 | |
| }, | |
| { | |
| "epoch": 334.0, | |
| "learning_rate": 1.2777777777777777e-07, | |
| "loss": 0.0382, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 336.0, | |
| "learning_rate": 1.25e-07, | |
| "loss": 0.038, | |
| "step": 3875 | |
| }, | |
| { | |
| "epoch": 339.0, | |
| "learning_rate": 1.2222222222222222e-07, | |
| "loss": 0.0377, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 341.0, | |
| "learning_rate": 1.1944444444444445e-07, | |
| "loss": 0.0374, | |
| "step": 3925 | |
| }, | |
| { | |
| "epoch": 343.0, | |
| "learning_rate": 1.1666666666666667e-07, | |
| "loss": 0.037, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 345.0, | |
| "learning_rate": 1.1388888888888888e-07, | |
| "loss": 0.0367, | |
| "step": 3975 | |
| }, | |
| { | |
| "epoch": 347.0, | |
| "learning_rate": 1.111111111111111e-07, | |
| "loss": 0.0367, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 347.0, | |
| "eval_loss": 0.6927756071090698, | |
| "eval_runtime": 127.2394, | |
| "eval_samples_per_second": 6.735, | |
| "eval_steps_per_second": 0.212, | |
| "eval_wer": 31.594926979246736, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 349.0, | |
| "learning_rate": 1.0833333333333334e-07, | |
| "loss": 0.0362, | |
| "step": 4025 | |
| }, | |
| { | |
| "epoch": 352.0, | |
| "learning_rate": 1.0555555555555555e-07, | |
| "loss": 0.036, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 354.0, | |
| "learning_rate": 1.0277777777777777e-07, | |
| "loss": 0.036, | |
| "step": 4075 | |
| }, | |
| { | |
| "epoch": 356.0, | |
| "learning_rate": 1e-07, | |
| "loss": 0.0353, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 358.0, | |
| "learning_rate": 9.722222222222222e-08, | |
| "loss": 0.0355, | |
| "step": 4125 | |
| }, | |
| { | |
| "epoch": 360.0, | |
| "learning_rate": 9.444444444444444e-08, | |
| "loss": 0.0353, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 363.0, | |
| "learning_rate": 9.166666666666665e-08, | |
| "loss": 0.035, | |
| "step": 4175 | |
| }, | |
| { | |
| "epoch": 365.0, | |
| "learning_rate": 8.888888888888888e-08, | |
| "loss": 0.0346, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 367.0, | |
| "learning_rate": 8.611111111111111e-08, | |
| "loss": 0.0345, | |
| "step": 4225 | |
| }, | |
| { | |
| "epoch": 369.0, | |
| "learning_rate": 8.333333333333333e-08, | |
| "loss": 0.0344, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 371.0, | |
| "learning_rate": 8.055555555555555e-08, | |
| "loss": 0.0342, | |
| "step": 4275 | |
| }, | |
| { | |
| "epoch": 373.0, | |
| "learning_rate": 7.777777777777778e-08, | |
| "loss": 0.0344, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 376.0, | |
| "learning_rate": 7.5e-08, | |
| "loss": 0.0336, | |
| "step": 4325 | |
| }, | |
| { | |
| "epoch": 378.0, | |
| "learning_rate": 7.222222222222221e-08, | |
| "loss": 0.0339, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 380.0, | |
| "learning_rate": 6.944444444444444e-08, | |
| "loss": 0.0339, | |
| "step": 4375 | |
| }, | |
| { | |
| "epoch": 382.0, | |
| "learning_rate": 6.666666666666667e-08, | |
| "loss": 0.0334, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 384.0, | |
| "learning_rate": 6.388888888888888e-08, | |
| "loss": 0.0332, | |
| "step": 4425 | |
| }, | |
| { | |
| "epoch": 386.0, | |
| "learning_rate": 6.111111111111111e-08, | |
| "loss": 0.033, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 389.0, | |
| "learning_rate": 5.833333333333333e-08, | |
| "loss": 0.0328, | |
| "step": 4475 | |
| }, | |
| { | |
| "epoch": 391.0, | |
| "learning_rate": 5.555555555555555e-08, | |
| "loss": 0.0331, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 391.0, | |
| "eval_loss": 0.6949473023414612, | |
| "eval_runtime": 125.7772, | |
| "eval_samples_per_second": 6.814, | |
| "eval_steps_per_second": 0.215, | |
| "eval_wer": 31.03382013835511, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 393.0, | |
| "learning_rate": 5.2777777777777776e-08, | |
| "loss": 0.0329, | |
| "step": 4525 | |
| }, | |
| { | |
| "epoch": 395.0, | |
| "learning_rate": 5e-08, | |
| "loss": 0.0328, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 397.0, | |
| "learning_rate": 4.722222222222222e-08, | |
| "loss": 0.0328, | |
| "step": 4575 | |
| }, | |
| { | |
| "epoch": 399.0, | |
| "learning_rate": 4.444444444444444e-08, | |
| "loss": 0.0325, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 402.0, | |
| "learning_rate": 4.166666666666666e-08, | |
| "loss": 0.0325, | |
| "step": 4625 | |
| }, | |
| { | |
| "epoch": 404.0, | |
| "learning_rate": 3.888888888888889e-08, | |
| "loss": 0.032, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 406.0, | |
| "learning_rate": 3.6111111111111106e-08, | |
| "loss": 0.0325, | |
| "step": 4675 | |
| }, | |
| { | |
| "epoch": 408.0, | |
| "learning_rate": 3.3333333333333334e-08, | |
| "loss": 0.0319, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 410.0, | |
| "learning_rate": 3.0555555555555556e-08, | |
| "loss": 0.032, | |
| "step": 4725 | |
| }, | |
| { | |
| "epoch": 413.0, | |
| "learning_rate": 2.7777777777777774e-08, | |
| "loss": 0.0319, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 415.0, | |
| "learning_rate": 2.5e-08, | |
| "loss": 0.0319, | |
| "step": 4775 | |
| }, | |
| { | |
| "epoch": 417.0, | |
| "learning_rate": 2.222222222222222e-08, | |
| "loss": 0.0318, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 419.0, | |
| "learning_rate": 1.9444444444444445e-08, | |
| "loss": 0.0319, | |
| "step": 4825 | |
| }, | |
| { | |
| "epoch": 421.0, | |
| "learning_rate": 1.6666666666666667e-08, | |
| "loss": 0.0319, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 423.0, | |
| "learning_rate": 1.3888888888888887e-08, | |
| "loss": 0.0315, | |
| "step": 4875 | |
| }, | |
| { | |
| "epoch": 426.0, | |
| "learning_rate": 1.111111111111111e-08, | |
| "loss": 0.0318, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 428.0, | |
| "learning_rate": 8.333333333333334e-09, | |
| "loss": 0.0313, | |
| "step": 4925 | |
| }, | |
| { | |
| "epoch": 430.0, | |
| "learning_rate": 5.555555555555555e-09, | |
| "loss": 0.032, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 432.0, | |
| "learning_rate": 2.7777777777777776e-09, | |
| "loss": 0.0313, | |
| "step": 4975 | |
| }, | |
| { | |
| "epoch": 434.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.0317, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 434.0, | |
| "eval_loss": 0.6956606507301331, | |
| "eval_runtime": 132.3167, | |
| "eval_samples_per_second": 6.477, | |
| "eval_steps_per_second": 0.204, | |
| "eval_wer": 31.045349730976174, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 434.0, | |
| "step": 5000, | |
| "total_flos": 2.027666171953152e+19, | |
| "train_loss": 0.1991682630777359, | |
| "train_runtime": 20636.3225, | |
| "train_samples_per_second": 15.507, | |
| "train_steps_per_second": 0.242 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "total_flos": 2.027666171953152e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |