| { | |
| "best_metric": 25.160615568504408, | |
| "best_model_checkpoint": "./checkpoint-5000", | |
| "epoch": 2.1788, | |
| "eval_steps": 500, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 4.2000000000000006e-07, | |
| "loss": 1.9502, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 8.8e-07, | |
| "loss": 1.3508, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 1.3800000000000001e-06, | |
| "loss": 0.7956, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.8800000000000002e-06, | |
| "loss": 0.5718, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 2.38e-06, | |
| "loss": 0.479, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 2.88e-06, | |
| "loss": 0.4334, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 3.3800000000000007e-06, | |
| "loss": 0.3928, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 3.88e-06, | |
| "loss": 0.3953, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 4.38e-06, | |
| "loss": 0.5018, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 4.880000000000001e-06, | |
| "loss": 0.5229, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 5.380000000000001e-06, | |
| "loss": 0.4696, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 5.8800000000000005e-06, | |
| "loss": 0.4296, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 6.380000000000001e-06, | |
| "loss": 0.534, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 6.88e-06, | |
| "loss": 0.559, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 7.3800000000000005e-06, | |
| "loss": 0.5479, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 7.88e-06, | |
| "loss": 0.5153, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 8.380000000000001e-06, | |
| "loss": 0.5327, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 8.880000000000001e-06, | |
| "loss": 0.3963, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.38e-06, | |
| "loss": 0.4182, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.88e-06, | |
| "loss": 0.3422, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 0.5929203629493713, | |
| "eval_runtime": 2188.7966, | |
| "eval_samples_per_second": 3.434, | |
| "eval_steps_per_second": 0.215, | |
| "eval_wer": 44.65361820807809, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.957777777777779e-06, | |
| "loss": 0.3705, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 9.902222222222223e-06, | |
| "loss": 0.3795, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 9.846666666666668e-06, | |
| "loss": 0.303, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 9.791111111111112e-06, | |
| "loss": 0.2878, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 9.735555555555556e-06, | |
| "loss": 0.4061, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 9.68e-06, | |
| "loss": 0.4774, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 9.624444444444445e-06, | |
| "loss": 0.4538, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 9.56888888888889e-06, | |
| "loss": 0.4927, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 9.513333333333334e-06, | |
| "loss": 0.5113, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 9.457777777777778e-06, | |
| "loss": 0.5401, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 9.402222222222222e-06, | |
| "loss": 0.5035, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 9.346666666666666e-06, | |
| "loss": 0.4589, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 9.291111111111112e-06, | |
| "loss": 0.4781, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 9.235555555555556e-06, | |
| "loss": 0.4842, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 9.180000000000002e-06, | |
| "loss": 0.5072, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 9.124444444444444e-06, | |
| "loss": 0.4898, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 9.06888888888889e-06, | |
| "loss": 0.4351, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 9.013333333333334e-06, | |
| "loss": 0.4532, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 8.957777777777778e-06, | |
| "loss": 0.459, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 8.902222222222224e-06, | |
| "loss": 0.4805, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 0.47376352548599243, | |
| "eval_runtime": 2134.6267, | |
| "eval_samples_per_second": 3.521, | |
| "eval_steps_per_second": 0.22, | |
| "eval_wer": 37.58902335773694, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 8.846666666666668e-06, | |
| "loss": 0.4535, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 8.791111111111112e-06, | |
| "loss": 0.455, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 8.735555555555556e-06, | |
| "loss": 0.432, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 8.68e-06, | |
| "loss": 0.4285, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 8.624444444444446e-06, | |
| "loss": 0.4101, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 8.56888888888889e-06, | |
| "loss": 0.4072, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 8.513333333333335e-06, | |
| "loss": 0.4049, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 8.457777777777778e-06, | |
| "loss": 0.4103, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 8.402222222222223e-06, | |
| "loss": 0.3725, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 8.346666666666668e-06, | |
| "loss": 0.3715, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 8.291111111111112e-06, | |
| "loss": 0.3856, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 8.235555555555557e-06, | |
| "loss": 0.3634, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 8.182222222222222e-06, | |
| "loss": 0.3889, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 8.126666666666668e-06, | |
| "loss": 0.3985, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 8.071111111111112e-06, | |
| "loss": 0.4281, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 8.015555555555556e-06, | |
| "loss": 0.3856, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 7.960000000000002e-06, | |
| "loss": 0.3661, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 7.904444444444444e-06, | |
| "loss": 0.3715, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 7.84888888888889e-06, | |
| "loss": 0.3989, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 7.793333333333334e-06, | |
| "loss": 0.3886, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_loss": 0.42949435114860535, | |
| "eval_runtime": 2131.5524, | |
| "eval_samples_per_second": 3.526, | |
| "eval_steps_per_second": 0.22, | |
| "eval_wer": 34.21734150107077, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 7.737777777777778e-06, | |
| "loss": 0.4242, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 7.682222222222224e-06, | |
| "loss": 0.3673, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 7.626666666666668e-06, | |
| "loss": 0.4005, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 7.571111111111112e-06, | |
| "loss": 0.4209, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 7.515555555555556e-06, | |
| "loss": 0.4032, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 7.4600000000000006e-06, | |
| "loss": 0.3914, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 7.4044444444444455e-06, | |
| "loss": 0.4295, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 7.3488888888888895e-06, | |
| "loss": 0.425, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 7.2933333333333335e-06, | |
| "loss": 0.3943, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 7.237777777777778e-06, | |
| "loss": 0.3964, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 7.1822222222222224e-06, | |
| "loss": 0.3265, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 7.126666666666667e-06, | |
| "loss": 0.3585, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 7.071111111111112e-06, | |
| "loss": 0.3808, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 7.015555555555556e-06, | |
| "loss": 0.3954, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 6.96e-06, | |
| "loss": 0.3809, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 6.904444444444444e-06, | |
| "loss": 0.368, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 6.848888888888889e-06, | |
| "loss": 0.3716, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 6.793333333333334e-06, | |
| "loss": 0.3772, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 6.737777777777779e-06, | |
| "loss": 0.3657, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 6.682222222222223e-06, | |
| "loss": 0.3929, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 0.3847246468067169, | |
| "eval_runtime": 2144.4869, | |
| "eval_samples_per_second": 3.505, | |
| "eval_steps_per_second": 0.219, | |
| "eval_wer": 30.955226853926987, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 6.626666666666667e-06, | |
| "loss": 0.3886, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 6.571111111111111e-06, | |
| "loss": 0.3978, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 6.515555555555556e-06, | |
| "loss": 0.3373, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 6.460000000000001e-06, | |
| "loss": 0.2841, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 6.404444444444446e-06, | |
| "loss": 0.2347, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 6.348888888888889e-06, | |
| "loss": 0.1821, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 6.293333333333334e-06, | |
| "loss": 0.1804, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 6.237777777777778e-06, | |
| "loss": 0.1867, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 6.182222222222223e-06, | |
| "loss": 0.1801, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 6.126666666666668e-06, | |
| "loss": 0.1615, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 6.0711111111111125e-06, | |
| "loss": 0.2176, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 6.015555555555556e-06, | |
| "loss": 0.245, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 5.9600000000000005e-06, | |
| "loss": 0.2334, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 5.9044444444444446e-06, | |
| "loss": 0.2344, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 5.8488888888888895e-06, | |
| "loss": 0.1908, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 5.793333333333334e-06, | |
| "loss": 0.2087, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 5.737777777777778e-06, | |
| "loss": 0.2023, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 5.682222222222222e-06, | |
| "loss": 0.2127, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 5.626666666666667e-06, | |
| "loss": 0.2259, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 5.571111111111111e-06, | |
| "loss": 0.2195, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "eval_loss": 0.36449727416038513, | |
| "eval_runtime": 2132.4653, | |
| "eval_samples_per_second": 3.525, | |
| "eval_steps_per_second": 0.22, | |
| "eval_wer": 28.213556452014544, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 5.515555555555556e-06, | |
| "loss": 0.2321, | |
| "step": 2525 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 5.460000000000001e-06, | |
| "loss": 0.2166, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 5.404444444444444e-06, | |
| "loss": 0.2025, | |
| "step": 2575 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 5.348888888888889e-06, | |
| "loss": 0.2383, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 5.293333333333334e-06, | |
| "loss": 0.2111, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 5.237777777777778e-06, | |
| "loss": 0.2259, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 5.182222222222223e-06, | |
| "loss": 0.1921, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 5.126666666666668e-06, | |
| "loss": 0.2208, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 5.071111111111111e-06, | |
| "loss": 0.2216, | |
| "step": 2725 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 5.015555555555556e-06, | |
| "loss": 0.2349, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 4.960000000000001e-06, | |
| "loss": 0.2296, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 4.904444444444445e-06, | |
| "loss": 0.1777, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 4.848888888888889e-06, | |
| "loss": 0.1949, | |
| "step": 2825 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 4.793333333333334e-06, | |
| "loss": 0.2204, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 4.737777777777779e-06, | |
| "loss": 0.185, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 4.682222222222223e-06, | |
| "loss": 0.2093, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 4.626666666666667e-06, | |
| "loss": 0.1687, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 4.571111111111112e-06, | |
| "loss": 0.1694, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 4.515555555555556e-06, | |
| "loss": 0.1798, | |
| "step": 2975 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 4.4600000000000005e-06, | |
| "loss": 0.1844, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "eval_loss": 0.36267614364624023, | |
| "eval_runtime": 2126.4881, | |
| "eval_samples_per_second": 3.534, | |
| "eval_steps_per_second": 0.221, | |
| "eval_wer": 27.705563026047113, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 4.404444444444445e-06, | |
| "loss": 0.1883, | |
| "step": 3025 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 4.348888888888889e-06, | |
| "loss": 0.1842, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 4.2933333333333334e-06, | |
| "loss": 0.2411, | |
| "step": 3075 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 4.2377777777777775e-06, | |
| "loss": 0.2348, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 4.182222222222222e-06, | |
| "loss": 0.2429, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 4.126666666666667e-06, | |
| "loss": 0.2574, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 4.071111111111111e-06, | |
| "loss": 0.1975, | |
| "step": 3175 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 4.015555555555556e-06, | |
| "loss": 0.1545, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 3.96e-06, | |
| "loss": 0.1459, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 3.904444444444444e-06, | |
| "loss": 0.1355, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 3.848888888888889e-06, | |
| "loss": 0.1444, | |
| "step": 3275 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 3.793333333333334e-06, | |
| "loss": 0.4791, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 3.737777777777778e-06, | |
| "loss": 0.1154, | |
| "step": 3325 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 3.6822222222222225e-06, | |
| "loss": 0.1299, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 3.6266666666666674e-06, | |
| "loss": 0.1766, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 3.5711111111111114e-06, | |
| "loss": 0.1968, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 3.515555555555556e-06, | |
| "loss": 0.2011, | |
| "step": 3425 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 3.46e-06, | |
| "loss": 0.1754, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 3.404444444444445e-06, | |
| "loss": 0.1766, | |
| "step": 3475 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 3.3488888888888892e-06, | |
| "loss": 0.1873, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "eval_loss": 0.35195156931877136, | |
| "eval_runtime": 2146.1663, | |
| "eval_samples_per_second": 3.502, | |
| "eval_steps_per_second": 0.219, | |
| "eval_wer": 27.14029583146571, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 3.2933333333333333e-06, | |
| "loss": 0.1885, | |
| "step": 3525 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 3.237777777777778e-06, | |
| "loss": 0.2021, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 3.1822222222222226e-06, | |
| "loss": 0.1724, | |
| "step": 3575 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 3.1266666666666667e-06, | |
| "loss": 0.1851, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 3.0711111111111115e-06, | |
| "loss": 0.1778, | |
| "step": 3625 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 3.015555555555556e-06, | |
| "loss": 0.1739, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 2.96e-06, | |
| "loss": 0.1636, | |
| "step": 3675 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 2.904444444444445e-06, | |
| "loss": 0.1849, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 2.8488888888888894e-06, | |
| "loss": 0.1927, | |
| "step": 3725 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 2.7933333333333334e-06, | |
| "loss": 0.1924, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 2.7377777777777783e-06, | |
| "loss": 0.1909, | |
| "step": 3775 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 2.6822222222222223e-06, | |
| "loss": 0.1635, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 2.6266666666666668e-06, | |
| "loss": 0.1689, | |
| "step": 3825 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 2.5711111111111112e-06, | |
| "loss": 0.1451, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 2.5155555555555557e-06, | |
| "loss": 0.1915, | |
| "step": 3875 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 2.46e-06, | |
| "loss": 0.1547, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 2.4044444444444446e-06, | |
| "loss": 0.1683, | |
| "step": 3925 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 2.348888888888889e-06, | |
| "loss": 0.1762, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 2.2933333333333335e-06, | |
| "loss": 0.1606, | |
| "step": 3975 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 2.237777777777778e-06, | |
| "loss": 0.1695, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "eval_loss": 0.3335447907447815, | |
| "eval_runtime": 2135.6594, | |
| "eval_samples_per_second": 3.519, | |
| "eval_steps_per_second": 0.22, | |
| "eval_wer": 26.181582748144827, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 2.1822222222222225e-06, | |
| "loss": 0.2014, | |
| "step": 4025 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 2.126666666666667e-06, | |
| "loss": 0.1779, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 2.0711111111111114e-06, | |
| "loss": 0.1998, | |
| "step": 4075 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 2.0155555555555554e-06, | |
| "loss": 0.2256, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.9600000000000003e-06, | |
| "loss": 0.1313, | |
| "step": 4125 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 1.9044444444444445e-06, | |
| "loss": 0.0907, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 1.848888888888889e-06, | |
| "loss": 0.096, | |
| "step": 4175 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 1.7933333333333337e-06, | |
| "loss": 0.0868, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 1.737777777777778e-06, | |
| "loss": 0.0832, | |
| "step": 4225 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 1.6822222222222224e-06, | |
| "loss": 0.0965, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 1.6266666666666666e-06, | |
| "loss": 0.0879, | |
| "step": 4275 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 1.5711111111111113e-06, | |
| "loss": 0.1016, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 1.5155555555555558e-06, | |
| "loss": 0.0834, | |
| "step": 4325 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 1.46e-06, | |
| "loss": 0.094, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 1.4044444444444447e-06, | |
| "loss": 0.0845, | |
| "step": 4375 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 1.3488888888888891e-06, | |
| "loss": 0.077, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 1.2933333333333334e-06, | |
| "loss": 0.0814, | |
| "step": 4425 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 1.2377777777777778e-06, | |
| "loss": 0.0974, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 1.1822222222222223e-06, | |
| "loss": 0.0956, | |
| "step": 4475 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 1.1266666666666667e-06, | |
| "loss": 0.089, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "eval_loss": 0.36234843730926514, | |
| "eval_runtime": 2138.092, | |
| "eval_samples_per_second": 3.515, | |
| "eval_steps_per_second": 0.22, | |
| "eval_wer": 26.10936799641416, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 1.0711111111111112e-06, | |
| "loss": 0.0838, | |
| "step": 4525 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 1.0155555555555557e-06, | |
| "loss": 0.1, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 9.600000000000001e-07, | |
| "loss": 0.105, | |
| "step": 4575 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 9.044444444444445e-07, | |
| "loss": 0.0716, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 8.488888888888889e-07, | |
| "loss": 0.0882, | |
| "step": 4625 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 7.933333333333335e-07, | |
| "loss": 0.0834, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 7.377777777777779e-07, | |
| "loss": 0.071, | |
| "step": 4675 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 6.822222222222223e-07, | |
| "loss": 0.0831, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 6.266666666666667e-07, | |
| "loss": 0.084, | |
| "step": 4725 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 5.711111111111111e-07, | |
| "loss": 0.0818, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 5.155555555555556e-07, | |
| "loss": 0.079, | |
| "step": 4775 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 4.6000000000000004e-07, | |
| "loss": 0.1148, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 4.0444444444444445e-07, | |
| "loss": 0.1241, | |
| "step": 4825 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 3.488888888888889e-07, | |
| "loss": 0.1287, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 2.9333333333333337e-07, | |
| "loss": 0.0923, | |
| "step": 4875 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 2.3777777777777777e-07, | |
| "loss": 0.0654, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 1.8222222222222226e-07, | |
| "loss": 0.0748, | |
| "step": 4925 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 1.2666666666666666e-07, | |
| "loss": 0.0624, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 7.111111111111112e-08, | |
| "loss": 0.0722, | |
| "step": 4975 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 1.5555555555555557e-08, | |
| "loss": 0.072, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "eval_loss": 0.3393059968948364, | |
| "eval_runtime": 2130.566, | |
| "eval_samples_per_second": 3.528, | |
| "eval_steps_per_second": 0.221, | |
| "eval_wer": 25.160615568504408, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "step": 5000, | |
| "total_flos": 1.69853681664e+20, | |
| "train_loss": 0.28622923114299775, | |
| "train_runtime": 48261.9735, | |
| "train_samples_per_second": 1.658, | |
| "train_steps_per_second": 0.104 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 1000, | |
| "total_flos": 1.69853681664e+20, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |