| { | |
| "best_metric": 17.036199095022624, | |
| "best_model_checkpoint": "./whisper-small-common-voice-16-pt/checkpoint-4000", | |
| "epoch": 3.7009622501850483, | |
| "eval_steps": 500, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 4.2000000000000006e-07, | |
| "loss": 4.9785, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 9.000000000000001e-07, | |
| "loss": 3.8237, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 1.4000000000000001e-06, | |
| "loss": 2.8164, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 1.9000000000000002e-06, | |
| "loss": 1.6381, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 2.4000000000000003e-06, | |
| "loss": 1.0913, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 2.9e-06, | |
| "loss": 0.9218, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 3.4000000000000005e-06, | |
| "loss": 0.8171, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 3.900000000000001e-06, | |
| "loss": 0.7009, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 4.38e-06, | |
| "loss": 0.6581, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 4.880000000000001e-06, | |
| "loss": 0.4877, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 5.380000000000001e-06, | |
| "loss": 0.3432, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 5.8800000000000005e-06, | |
| "loss": 0.2772, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 6.380000000000001e-06, | |
| "loss": 0.3046, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 6.88e-06, | |
| "loss": 0.3073, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 7.3800000000000005e-06, | |
| "loss": 0.2705, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 7.88e-06, | |
| "loss": 0.2777, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 8.380000000000001e-06, | |
| "loss": 0.2733, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 8.880000000000001e-06, | |
| "loss": 0.2705, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 9.38e-06, | |
| "loss": 0.2593, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 9.88e-06, | |
| "loss": 0.2591, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_loss": 0.2776341736316681, | |
| "eval_runtime": 2710.5305, | |
| "eval_samples_per_second": 3.468, | |
| "eval_steps_per_second": 0.433, | |
| "eval_wer": 20.07272139625081, | |
| "eval_wer_normalized": 13.67072759666779, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 9.957777777777779e-06, | |
| "loss": 0.2517, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 9.902222222222223e-06, | |
| "loss": 0.2477, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 9.846666666666668e-06, | |
| "loss": 0.2491, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 9.791111111111112e-06, | |
| "loss": 0.2574, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 9.735555555555556e-06, | |
| "loss": 0.2692, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 9.68e-06, | |
| "loss": 0.2729, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 9.624444444444445e-06, | |
| "loss": 0.2401, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 9.56888888888889e-06, | |
| "loss": 0.2345, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 9.513333333333334e-06, | |
| "loss": 0.2319, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 9.457777777777778e-06, | |
| "loss": 0.283, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 9.402222222222222e-06, | |
| "loss": 0.2324, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 9.346666666666666e-06, | |
| "loss": 0.2459, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 9.291111111111112e-06, | |
| "loss": 0.2373, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 9.235555555555556e-06, | |
| "loss": 0.2413, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 9.180000000000002e-06, | |
| "loss": 0.2633, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 9.124444444444444e-06, | |
| "loss": 0.23, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 9.06888888888889e-06, | |
| "loss": 0.2402, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 9.013333333333334e-06, | |
| "loss": 0.2298, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 8.957777777777778e-06, | |
| "loss": 0.205, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 8.902222222222224e-06, | |
| "loss": 0.2626, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "eval_loss": 0.2599042057991028, | |
| "eval_runtime": 2724.1583, | |
| "eval_samples_per_second": 3.45, | |
| "eval_steps_per_second": 0.431, | |
| "eval_wer": 19.400452488687783, | |
| "eval_wer_normalized": 13.333654355467809, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 8.846666666666668e-06, | |
| "loss": 0.2616, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 8.791111111111112e-06, | |
| "loss": 0.2532, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 8.735555555555556e-06, | |
| "loss": 0.2405, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 8.68e-06, | |
| "loss": 0.2162, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 8.624444444444446e-06, | |
| "loss": 0.2431, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 8.56888888888889e-06, | |
| "loss": 0.2352, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 8.513333333333335e-06, | |
| "loss": 0.2682, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 8.457777777777778e-06, | |
| "loss": 0.2161, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 8.402222222222223e-06, | |
| "loss": 0.2299, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 8.346666666666668e-06, | |
| "loss": 0.246, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 8.291111111111112e-06, | |
| "loss": 0.2518, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 8.235555555555557e-06, | |
| "loss": 0.2467, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 8.18e-06, | |
| "loss": 0.2147, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 8.124444444444445e-06, | |
| "loss": 0.2452, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 8.06888888888889e-06, | |
| "loss": 0.1274, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 8.013333333333333e-06, | |
| "loss": 0.1139, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 7.957777777777779e-06, | |
| "loss": 0.1065, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 7.902222222222223e-06, | |
| "loss": 0.1072, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 7.846666666666667e-06, | |
| "loss": 0.1226, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 7.791111111111111e-06, | |
| "loss": 0.1131, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "eval_loss": 0.2515513598918915, | |
| "eval_runtime": 2716.9869, | |
| "eval_samples_per_second": 3.459, | |
| "eval_steps_per_second": 0.432, | |
| "eval_wer": 18.04137039431157, | |
| "eval_wer_normalized": 12.133031572526926, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 7.735555555555557e-06, | |
| "loss": 0.1119, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 7.680000000000001e-06, | |
| "loss": 0.1057, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 7.624444444444445e-06, | |
| "loss": 0.1145, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 7.56888888888889e-06, | |
| "loss": 0.1256, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 7.513333333333334e-06, | |
| "loss": 0.1126, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 7.457777777777778e-06, | |
| "loss": 0.1097, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 7.402222222222223e-06, | |
| "loss": 0.1052, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 7.346666666666668e-06, | |
| "loss": 0.1293, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 7.291111111111112e-06, | |
| "loss": 0.1116, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 7.235555555555556e-06, | |
| "loss": 0.1174, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 7.180000000000001e-06, | |
| "loss": 0.1105, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 7.124444444444445e-06, | |
| "loss": 0.1221, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 7.06888888888889e-06, | |
| "loss": 0.1139, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 7.0133333333333345e-06, | |
| "loss": 0.1201, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 6.9577777777777785e-06, | |
| "loss": 0.1211, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 6.902222222222223e-06, | |
| "loss": 0.1192, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 6.846666666666667e-06, | |
| "loss": 0.1133, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 6.7911111111111115e-06, | |
| "loss": 0.1013, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 6.735555555555556e-06, | |
| "loss": 0.1074, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 6.680000000000001e-06, | |
| "loss": 0.1016, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "eval_loss": 0.24819953739643097, | |
| "eval_runtime": 2711.2399, | |
| "eval_samples_per_second": 3.467, | |
| "eval_steps_per_second": 0.433, | |
| "eval_wer": 18.35972850678733, | |
| "eval_wer_normalized": 11.924367185117415, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 6.6244444444444445e-06, | |
| "loss": 0.1255, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 6.568888888888889e-06, | |
| "loss": 0.1113, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 6.513333333333333e-06, | |
| "loss": 0.11, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 6.457777777777778e-06, | |
| "loss": 0.1096, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 6.402222222222223e-06, | |
| "loss": 0.11, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 6.346666666666668e-06, | |
| "loss": 0.12, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 6.291111111111111e-06, | |
| "loss": 0.1101, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 6.235555555555556e-06, | |
| "loss": 0.0962, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 6.18e-06, | |
| "loss": 0.104, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 6.124444444444445e-06, | |
| "loss": 0.0933, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 6.06888888888889e-06, | |
| "loss": 0.1207, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 6.013333333333335e-06, | |
| "loss": 0.111, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 5.957777777777778e-06, | |
| "loss": 0.1154, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 5.902222222222223e-06, | |
| "loss": 0.1024, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 5.846666666666667e-06, | |
| "loss": 0.1036, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 5.791111111111112e-06, | |
| "loss": 0.1235, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 5.735555555555557e-06, | |
| "loss": 0.1019, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 5.68e-06, | |
| "loss": 0.1182, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 5.624444444444445e-06, | |
| "loss": 0.1065, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 5.56888888888889e-06, | |
| "loss": 0.1094, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "eval_loss": 0.24113476276397705, | |
| "eval_runtime": 2716.3575, | |
| "eval_samples_per_second": 3.46, | |
| "eval_steps_per_second": 0.433, | |
| "eval_wer": 17.419198448610214, | |
| "eval_wer_normalized": 11.60173993996886, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 5.513333333333334e-06, | |
| "loss": 0.094, | |
| "step": 2525 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 5.4577777777777785e-06, | |
| "loss": 0.1107, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 5.402222222222223e-06, | |
| "loss": 0.1221, | |
| "step": 2575 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 5.346666666666667e-06, | |
| "loss": 0.1111, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 5.2911111111111115e-06, | |
| "loss": 0.108, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 5.235555555555556e-06, | |
| "loss": 0.114, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 5.18e-06, | |
| "loss": 0.1039, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 5.124444444444445e-06, | |
| "loss": 0.1123, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 5.06888888888889e-06, | |
| "loss": 0.0625, | |
| "step": 2725 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 5.013333333333333e-06, | |
| "loss": 0.0488, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 4.957777777777778e-06, | |
| "loss": 0.0493, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 4.902222222222222e-06, | |
| "loss": 0.0465, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 4.846666666666667e-06, | |
| "loss": 0.0451, | |
| "step": 2825 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 4.791111111111111e-06, | |
| "loss": 0.0436, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 4.735555555555556e-06, | |
| "loss": 0.0606, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 4.680000000000001e-06, | |
| "loss": 0.0505, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 4.624444444444445e-06, | |
| "loss": 0.0562, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 4.568888888888889e-06, | |
| "loss": 0.051, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 4.513333333333333e-06, | |
| "loss": 0.0407, | |
| "step": 2975 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 4.457777777777778e-06, | |
| "loss": 0.0524, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "eval_loss": 0.25118857622146606, | |
| "eval_runtime": 2728.2785, | |
| "eval_samples_per_second": 3.445, | |
| "eval_steps_per_second": 0.431, | |
| "eval_wer": 17.354557207498384, | |
| "eval_wer_normalized": 11.463700422144107, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 4.402222222222223e-06, | |
| "loss": 0.0478, | |
| "step": 3025 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 4.346666666666667e-06, | |
| "loss": 0.0487, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 4.291111111111112e-06, | |
| "loss": 0.046, | |
| "step": 3075 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 4.235555555555556e-06, | |
| "loss": 0.0425, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 4.18e-06, | |
| "loss": 0.0428, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 4.124444444444445e-06, | |
| "loss": 0.0455, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 4.0688888888888896e-06, | |
| "loss": 0.0482, | |
| "step": 3175 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 4.013333333333334e-06, | |
| "loss": 0.0428, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 3.9577777777777785e-06, | |
| "loss": 0.0501, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 3.9022222222222225e-06, | |
| "loss": 0.0486, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 3.8466666666666665e-06, | |
| "loss": 0.0554, | |
| "step": 3275 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 3.7911111111111114e-06, | |
| "loss": 0.0443, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 3.7355555555555555e-06, | |
| "loss": 0.0536, | |
| "step": 3325 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 3.6800000000000003e-06, | |
| "loss": 0.0562, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 3.624444444444445e-06, | |
| "loss": 0.0541, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 3.568888888888889e-06, | |
| "loss": 0.0472, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 3.5133333333333337e-06, | |
| "loss": 0.0449, | |
| "step": 3425 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 3.457777777777778e-06, | |
| "loss": 0.0424, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 3.4022222222222222e-06, | |
| "loss": 0.0478, | |
| "step": 3475 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 3.346666666666667e-06, | |
| "loss": 0.0433, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "eval_loss": 0.24957023561000824, | |
| "eval_runtime": 2729.044, | |
| "eval_samples_per_second": 3.444, | |
| "eval_steps_per_second": 0.431, | |
| "eval_wer": 17.089528118939885, | |
| "eval_wer_normalized": 11.298374022888877, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 3.2911111111111116e-06, | |
| "loss": 0.0438, | |
| "step": 3525 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 3.2355555555555556e-06, | |
| "loss": 0.0501, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 3.1800000000000005e-06, | |
| "loss": 0.0532, | |
| "step": 3575 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 3.124444444444445e-06, | |
| "loss": 0.0478, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 3.068888888888889e-06, | |
| "loss": 0.0477, | |
| "step": 3625 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 3.013333333333334e-06, | |
| "loss": 0.0545, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 2.957777777777778e-06, | |
| "loss": 0.0532, | |
| "step": 3675 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 2.9022222222222223e-06, | |
| "loss": 0.0441, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 2.8466666666666672e-06, | |
| "loss": 0.0473, | |
| "step": 3725 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 2.7911111111111113e-06, | |
| "loss": 0.0473, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 2.7355555555555557e-06, | |
| "loss": 0.049, | |
| "step": 3775 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 2.68e-06, | |
| "loss": 0.044, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 2.6244444444444446e-06, | |
| "loss": 0.0484, | |
| "step": 3825 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 2.568888888888889e-06, | |
| "loss": 0.0446, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "learning_rate": 2.5133333333333336e-06, | |
| "loss": 0.0479, | |
| "step": 3875 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 2.457777777777778e-06, | |
| "loss": 0.045, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 2.4022222222222225e-06, | |
| "loss": 0.0535, | |
| "step": 3925 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 2.346666666666667e-06, | |
| "loss": 0.0459, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 2.2911111111111114e-06, | |
| "loss": 0.0515, | |
| "step": 3975 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 2.235555555555556e-06, | |
| "loss": 0.0453, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "eval_loss": 0.24788813292980194, | |
| "eval_runtime": 2717.3489, | |
| "eval_samples_per_second": 3.459, | |
| "eval_steps_per_second": 0.432, | |
| "eval_wer": 17.036199095022624, | |
| "eval_wer_normalized": 11.267876920113643, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 2.1800000000000003e-06, | |
| "loss": 0.0497, | |
| "step": 4025 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 2.1244444444444443e-06, | |
| "loss": 0.0448, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 2.0688888888888892e-06, | |
| "loss": 0.0283, | |
| "step": 4075 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 2.0133333333333337e-06, | |
| "loss": 0.0222, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 1.9577777777777777e-06, | |
| "loss": 0.0215, | |
| "step": 4125 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 1.9022222222222222e-06, | |
| "loss": 0.0248, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "learning_rate": 1.8466666666666668e-06, | |
| "loss": 0.0227, | |
| "step": 4175 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "learning_rate": 1.7911111111111113e-06, | |
| "loss": 0.0241, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "learning_rate": 1.7355555555555555e-06, | |
| "loss": 0.0204, | |
| "step": 4225 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "learning_rate": 1.6800000000000002e-06, | |
| "loss": 0.0178, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "learning_rate": 1.6244444444444447e-06, | |
| "loss": 0.0182, | |
| "step": 4275 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "learning_rate": 1.568888888888889e-06, | |
| "loss": 0.0217, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 1.5133333333333334e-06, | |
| "loss": 0.0222, | |
| "step": 4325 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 1.457777777777778e-06, | |
| "loss": 0.022, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "learning_rate": 1.4022222222222223e-06, | |
| "loss": 0.018, | |
| "step": 4375 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "learning_rate": 1.3466666666666668e-06, | |
| "loss": 0.0211, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 1.2911111111111112e-06, | |
| "loss": 0.0215, | |
| "step": 4425 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "learning_rate": 1.2355555555555557e-06, | |
| "loss": 0.0197, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 1.1800000000000001e-06, | |
| "loss": 0.0227, | |
| "step": 4475 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 1.1244444444444446e-06, | |
| "loss": 0.0201, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "eval_loss": 0.269341379404068, | |
| "eval_runtime": 2723.6438, | |
| "eval_samples_per_second": 3.451, | |
| "eval_steps_per_second": 0.431, | |
| "eval_wer": 17.66321913380737, | |
| "eval_wer_normalized": 11.71088746569076, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 1.068888888888889e-06, | |
| "loss": 0.0198, | |
| "step": 4525 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 1.0133333333333333e-06, | |
| "loss": 0.021, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 9.57777777777778e-07, | |
| "loss": 0.0221, | |
| "step": 4575 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "learning_rate": 9.022222222222222e-07, | |
| "loss": 0.0191, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 8.466666666666668e-07, | |
| "loss": 0.0198, | |
| "step": 4625 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "learning_rate": 7.911111111111111e-07, | |
| "loss": 0.0191, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 7.355555555555556e-07, | |
| "loss": 0.02, | |
| "step": 4675 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "learning_rate": 6.800000000000001e-07, | |
| "loss": 0.0253, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 6.244444444444445e-07, | |
| "loss": 0.0237, | |
| "step": 4725 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "learning_rate": 5.68888888888889e-07, | |
| "loss": 0.0162, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 5.133333333333334e-07, | |
| "loss": 0.0191, | |
| "step": 4775 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 4.5777777777777784e-07, | |
| "loss": 0.0221, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "learning_rate": 4.0222222222222224e-07, | |
| "loss": 0.0225, | |
| "step": 4825 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 3.466666666666667e-07, | |
| "loss": 0.0216, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "learning_rate": 2.9111111111111116e-07, | |
| "loss": 0.0203, | |
| "step": 4875 | |
| }, | |
| { | |
| "epoch": 3.63, | |
| "learning_rate": 2.3555555555555556e-07, | |
| "loss": 0.0216, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "learning_rate": 1.8e-07, | |
| "loss": 0.0202, | |
| "step": 4925 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "learning_rate": 1.2444444444444446e-07, | |
| "loss": 0.0185, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 6.888888888888889e-08, | |
| "loss": 0.0195, | |
| "step": 4975 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 1.3333333333333334e-08, | |
| "loss": 0.0229, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "eval_loss": 0.2711798846721649, | |
| "eval_runtime": 2712.8452, | |
| "eval_samples_per_second": 3.465, | |
| "eval_steps_per_second": 0.433, | |
| "eval_wer": 17.33354880413704, | |
| "eval_wer_normalized": 11.332081347008877, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "step": 5000, | |
| "total_flos": 2.307990597599232e+19, | |
| "train_loss": 0.1926749886929989, | |
| "train_runtime": 37692.9225, | |
| "train_samples_per_second": 2.122, | |
| "train_steps_per_second": 0.133 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 500, | |
| "total_flos": 2.307990597599232e+19, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |