| { | |
| "best_metric": 28.809032414714096, | |
| "best_model_checkpoint": "./checkpoint-1500", | |
| "epoch": 526.0006, | |
| "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": 2.2999999999999998e-08, | |
| "loss": 0.5935, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.8e-08, | |
| "loss": 0.595, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 7.299999999999999e-08, | |
| "loss": 0.5715, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 9.8e-08, | |
| "loss": 0.5452, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 1.23e-07, | |
| "loss": 0.5321, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 1.4799999999999998e-07, | |
| "loss": 0.5084, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 1.7299999999999997e-07, | |
| "loss": 0.4796, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "learning_rate": 1.98e-07, | |
| "loss": 0.4607, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "learning_rate": 2.23e-07, | |
| "loss": 0.4399, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "learning_rate": 2.48e-07, | |
| "loss": 0.42, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "learning_rate": 2.73e-07, | |
| "loss": 0.4021, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "learning_rate": 2.98e-07, | |
| "loss": 0.383, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "learning_rate": 3.23e-07, | |
| "loss": 0.3694, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "learning_rate": 3.4799999999999994e-07, | |
| "loss": 0.3495, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "learning_rate": 3.7299999999999997e-07, | |
| "loss": 0.3251, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "learning_rate": 3.98e-07, | |
| "loss": 0.3137, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "learning_rate": 4.2299999999999996e-07, | |
| "loss": 0.2936, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "learning_rate": 4.48e-07, | |
| "loss": 0.2721, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "learning_rate": 4.7299999999999996e-07, | |
| "loss": 0.2608, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "learning_rate": 4.979999999999999e-07, | |
| "loss": 0.2414, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_loss": 0.39878302812576294, | |
| "eval_runtime": 168.8098, | |
| "eval_samples_per_second": 2.535, | |
| "eval_steps_per_second": 0.083, | |
| "eval_wer": 30.569382056573996, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "learning_rate": 4.974444444444444e-07, | |
| "loss": 0.2243, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 57.0, | |
| "learning_rate": 4.946666666666666e-07, | |
| "loss": 0.2054, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "learning_rate": 4.918888888888889e-07, | |
| "loss": 0.1952, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 63.0, | |
| "learning_rate": 4.891111111111111e-07, | |
| "loss": 0.1732, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "learning_rate": 4.863333333333333e-07, | |
| "loss": 0.164, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "learning_rate": 4.835555555555555e-07, | |
| "loss": 0.1481, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 71.0, | |
| "learning_rate": 4.807777777777778e-07, | |
| "loss": 0.1369, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 73.0, | |
| "learning_rate": 4.779999999999999e-07, | |
| "loss": 0.1249, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "learning_rate": 4.7522222222222217e-07, | |
| "loss": 0.1153, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 78.0, | |
| "learning_rate": 4.724444444444444e-07, | |
| "loss": 0.1045, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 81.0, | |
| "learning_rate": 4.696666666666667e-07, | |
| "loss": 0.0952, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "learning_rate": 4.668888888888889e-07, | |
| "loss": 0.0871, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 86.0, | |
| "learning_rate": 4.641111111111111e-07, | |
| "loss": 0.0789, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 89.0, | |
| "learning_rate": 4.613333333333333e-07, | |
| "loss": 0.072, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 92.0, | |
| "learning_rate": 4.5855555555555554e-07, | |
| "loss": 0.0644, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 94.0, | |
| "learning_rate": 4.5577777777777774e-07, | |
| "loss": 0.0598, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 97.0, | |
| "learning_rate": 4.53e-07, | |
| "loss": 0.0527, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 99.0, | |
| "learning_rate": 4.5022222222222225e-07, | |
| "loss": 0.0493, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 102.0, | |
| "learning_rate": 4.474444444444444e-07, | |
| "loss": 0.0452, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 105.0, | |
| "learning_rate": 4.4466666666666665e-07, | |
| "loss": 0.0412, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 105.0, | |
| "eval_loss": 0.42843055725097656, | |
| "eval_runtime": 176.206, | |
| "eval_samples_per_second": 2.429, | |
| "eval_steps_per_second": 0.079, | |
| "eval_wer": 30.569382056573996, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 107.0, | |
| "learning_rate": 4.4188888888888885e-07, | |
| "loss": 0.0381, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 110.0, | |
| "learning_rate": 4.391111111111111e-07, | |
| "loss": 0.0357, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 113.0, | |
| "learning_rate": 4.363333333333333e-07, | |
| "loss": 0.0333, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 115.0, | |
| "learning_rate": 4.3355555555555556e-07, | |
| "loss": 0.0306, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 118.0, | |
| "learning_rate": 4.3077777777777777e-07, | |
| "loss": 0.0288, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 121.0, | |
| "learning_rate": 4.2799999999999997e-07, | |
| "loss": 0.0273, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 123.0, | |
| "learning_rate": 4.252222222222222e-07, | |
| "loss": 0.0258, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 126.0, | |
| "learning_rate": 4.224444444444444e-07, | |
| "loss": 0.0241, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 128.0, | |
| "learning_rate": 4.196666666666667e-07, | |
| "loss": 0.0233, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 131.0, | |
| "learning_rate": 4.1688888888888883e-07, | |
| "loss": 0.0219, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 134.0, | |
| "learning_rate": 4.141111111111111e-07, | |
| "loss": 0.0211, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 136.0, | |
| "learning_rate": 4.113333333333333e-07, | |
| "loss": 0.0199, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 139.0, | |
| "learning_rate": 4.0855555555555554e-07, | |
| "loss": 0.0191, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 142.0, | |
| "learning_rate": 4.057777777777778e-07, | |
| "loss": 0.0185, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 144.0, | |
| "learning_rate": 4.03e-07, | |
| "loss": 0.0176, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 147.0, | |
| "learning_rate": 4.002222222222222e-07, | |
| "loss": 0.017, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 149.0, | |
| "learning_rate": 3.974444444444444e-07, | |
| "loss": 0.0163, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 152.0, | |
| "learning_rate": 3.9466666666666665e-07, | |
| "loss": 0.0156, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 155.0, | |
| "learning_rate": 3.9188888888888886e-07, | |
| "loss": 0.0151, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 157.0, | |
| "learning_rate": 3.891111111111111e-07, | |
| "loss": 0.0147, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 157.0, | |
| "eval_loss": 0.4547523856163025, | |
| "eval_runtime": 158.1807, | |
| "eval_samples_per_second": 2.706, | |
| "eval_steps_per_second": 0.089, | |
| "eval_wer": 28.809032414714096, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 160.0, | |
| "learning_rate": 3.863333333333333e-07, | |
| "loss": 0.0141, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 163.0, | |
| "learning_rate": 3.835555555555555e-07, | |
| "loss": 0.0137, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 165.0, | |
| "learning_rate": 3.8077777777777777e-07, | |
| "loss": 0.0132, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 168.0, | |
| "learning_rate": 3.7799999999999997e-07, | |
| "loss": 0.0129, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 171.0, | |
| "learning_rate": 3.752222222222222e-07, | |
| "loss": 0.0124, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 173.0, | |
| "learning_rate": 3.7244444444444443e-07, | |
| "loss": 0.0121, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 176.0, | |
| "learning_rate": 3.6966666666666663e-07, | |
| "loss": 0.0117, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 178.0, | |
| "learning_rate": 3.668888888888889e-07, | |
| "loss": 0.0114, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 181.0, | |
| "learning_rate": 3.641111111111111e-07, | |
| "loss": 0.0111, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 184.0, | |
| "learning_rate": 3.6133333333333334e-07, | |
| "loss": 0.0108, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 186.0, | |
| "learning_rate": 3.5855555555555554e-07, | |
| "loss": 0.0105, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 189.0, | |
| "learning_rate": 3.557777777777778e-07, | |
| "loss": 0.0102, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 192.0, | |
| "learning_rate": 3.5299999999999994e-07, | |
| "loss": 0.01, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 194.0, | |
| "learning_rate": 3.502222222222222e-07, | |
| "loss": 0.0097, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 197.0, | |
| "learning_rate": 3.4744444444444445e-07, | |
| "loss": 0.0094, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 199.0, | |
| "learning_rate": 3.4466666666666666e-07, | |
| "loss": 0.0093, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 202.0, | |
| "learning_rate": 3.418888888888889e-07, | |
| "loss": 0.009, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 205.0, | |
| "learning_rate": 3.3911111111111106e-07, | |
| "loss": 0.0088, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 207.0, | |
| "learning_rate": 3.363333333333333e-07, | |
| "loss": 0.0086, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 210.0, | |
| "learning_rate": 3.335555555555555e-07, | |
| "loss": 0.0084, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 210.0, | |
| "eval_loss": 0.4738238751888275, | |
| "eval_runtime": 158.8439, | |
| "eval_samples_per_second": 2.694, | |
| "eval_steps_per_second": 0.088, | |
| "eval_wer": 29.11254097365546, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 213.0, | |
| "learning_rate": 3.3077777777777777e-07, | |
| "loss": 0.0083, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 215.0, | |
| "learning_rate": 3.28e-07, | |
| "loss": 0.0081, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 218.0, | |
| "learning_rate": 3.2522222222222223e-07, | |
| "loss": 0.0079, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 221.0, | |
| "learning_rate": 3.2244444444444443e-07, | |
| "loss": 0.0078, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 223.0, | |
| "learning_rate": 3.1966666666666663e-07, | |
| "loss": 0.0075, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 226.0, | |
| "learning_rate": 3.168888888888889e-07, | |
| "loss": 0.0074, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 228.0, | |
| "learning_rate": 3.141111111111111e-07, | |
| "loss": 0.0073, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 231.0, | |
| "learning_rate": 3.1133333333333334e-07, | |
| "loss": 0.0071, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 234.0, | |
| "learning_rate": 3.0855555555555554e-07, | |
| "loss": 0.0069, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 236.0, | |
| "learning_rate": 3.0577777777777775e-07, | |
| "loss": 0.0069, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 239.0, | |
| "learning_rate": 3.03e-07, | |
| "loss": 0.0068, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 242.0, | |
| "learning_rate": 3.002222222222222e-07, | |
| "loss": 0.0066, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 244.0, | |
| "learning_rate": 2.9744444444444446e-07, | |
| "loss": 0.0065, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 247.0, | |
| "learning_rate": 2.9466666666666666e-07, | |
| "loss": 0.0063, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 249.0, | |
| "learning_rate": 2.9188888888888886e-07, | |
| "loss": 0.0063, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 252.0, | |
| "learning_rate": 2.8911111111111106e-07, | |
| "loss": 0.0061, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 255.0, | |
| "learning_rate": 2.863333333333333e-07, | |
| "loss": 0.006, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 257.0, | |
| "learning_rate": 2.8355555555555557e-07, | |
| "loss": 0.0059, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 260.0, | |
| "learning_rate": 2.8077777777777777e-07, | |
| "loss": 0.0059, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 263.0, | |
| "learning_rate": 2.7800000000000003e-07, | |
| "loss": 0.0057, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 263.0, | |
| "eval_loss": 0.48884743452072144, | |
| "eval_runtime": 158.8642, | |
| "eval_samples_per_second": 2.694, | |
| "eval_steps_per_second": 0.088, | |
| "eval_wer": 29.355347820808547, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 265.0, | |
| "learning_rate": 2.752222222222222e-07, | |
| "loss": 0.0057, | |
| "step": 2525 | |
| }, | |
| { | |
| "epoch": 268.0, | |
| "learning_rate": 2.7244444444444443e-07, | |
| "loss": 0.0055, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 271.0, | |
| "learning_rate": 2.6966666666666663e-07, | |
| "loss": 0.0055, | |
| "step": 2575 | |
| }, | |
| { | |
| "epoch": 273.0, | |
| "learning_rate": 2.668888888888889e-07, | |
| "loss": 0.0054, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 276.0, | |
| "learning_rate": 2.6411111111111114e-07, | |
| "loss": 0.0053, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 278.0, | |
| "learning_rate": 2.613333333333333e-07, | |
| "loss": 0.0052, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 281.0, | |
| "learning_rate": 2.5855555555555555e-07, | |
| "loss": 0.0051, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 284.0, | |
| "learning_rate": 2.5577777777777775e-07, | |
| "loss": 0.0051, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 286.0, | |
| "learning_rate": 2.53e-07, | |
| "loss": 0.005, | |
| "step": 2725 | |
| }, | |
| { | |
| "epoch": 289.0, | |
| "learning_rate": 2.502222222222222e-07, | |
| "loss": 0.0049, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 292.0, | |
| "learning_rate": 2.474444444444444e-07, | |
| "loss": 0.0048, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 294.0, | |
| "learning_rate": 2.4466666666666666e-07, | |
| "loss": 0.0048, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 297.0, | |
| "learning_rate": 2.4188888888888886e-07, | |
| "loss": 0.0047, | |
| "step": 2825 | |
| }, | |
| { | |
| "epoch": 299.0, | |
| "learning_rate": 2.391111111111111e-07, | |
| "loss": 0.0046, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 302.0, | |
| "learning_rate": 2.3633333333333332e-07, | |
| "loss": 0.0046, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 305.0, | |
| "learning_rate": 2.3355555555555552e-07, | |
| "loss": 0.0045, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 307.0, | |
| "learning_rate": 2.3077777777777778e-07, | |
| "loss": 0.0045, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 310.0, | |
| "learning_rate": 2.28e-07, | |
| "loss": 0.0044, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 313.0, | |
| "learning_rate": 2.252222222222222e-07, | |
| "loss": 0.0043, | |
| "step": 2975 | |
| }, | |
| { | |
| "epoch": 315.0, | |
| "learning_rate": 2.2244444444444443e-07, | |
| "loss": 0.0043, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 315.0, | |
| "eval_loss": 0.501005232334137, | |
| "eval_runtime": 157.999, | |
| "eval_samples_per_second": 2.709, | |
| "eval_steps_per_second": 0.089, | |
| "eval_wer": 29.221804054874344, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 318.0, | |
| "learning_rate": 2.1966666666666666e-07, | |
| "loss": 0.0042, | |
| "step": 3025 | |
| }, | |
| { | |
| "epoch": 321.0, | |
| "learning_rate": 2.1688888888888886e-07, | |
| "loss": 0.0042, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 323.0, | |
| "learning_rate": 2.141111111111111e-07, | |
| "loss": 0.0041, | |
| "step": 3075 | |
| }, | |
| { | |
| "epoch": 326.0, | |
| "learning_rate": 2.1133333333333335e-07, | |
| "loss": 0.0041, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 328.0, | |
| "learning_rate": 2.0855555555555555e-07, | |
| "loss": 0.004, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 331.0, | |
| "learning_rate": 2.0577777777777778e-07, | |
| "loss": 0.004, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 334.0, | |
| "learning_rate": 2.03e-07, | |
| "loss": 0.0039, | |
| "step": 3175 | |
| }, | |
| { | |
| "epoch": 336.0, | |
| "learning_rate": 2.002222222222222e-07, | |
| "loss": 0.0039, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 339.0, | |
| "learning_rate": 1.9744444444444443e-07, | |
| "loss": 0.0038, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 342.0, | |
| "learning_rate": 1.9466666666666664e-07, | |
| "loss": 0.0038, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 344.0, | |
| "learning_rate": 1.918888888888889e-07, | |
| "loss": 0.0037, | |
| "step": 3275 | |
| }, | |
| { | |
| "epoch": 347.0, | |
| "learning_rate": 1.8911111111111112e-07, | |
| "loss": 0.0037, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 349.0, | |
| "learning_rate": 1.8633333333333332e-07, | |
| "loss": 0.0037, | |
| "step": 3325 | |
| }, | |
| { | |
| "epoch": 352.0, | |
| "learning_rate": 1.8355555555555555e-07, | |
| "loss": 0.0036, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 355.0, | |
| "learning_rate": 1.8077777777777778e-07, | |
| "loss": 0.0036, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 357.0, | |
| "learning_rate": 1.7799999999999998e-07, | |
| "loss": 0.0035, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 360.0, | |
| "learning_rate": 1.752222222222222e-07, | |
| "loss": 0.0035, | |
| "step": 3425 | |
| }, | |
| { | |
| "epoch": 363.0, | |
| "learning_rate": 1.7244444444444444e-07, | |
| "loss": 0.0035, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 365.0, | |
| "learning_rate": 1.6966666666666666e-07, | |
| "loss": 0.0034, | |
| "step": 3475 | |
| }, | |
| { | |
| "epoch": 368.0, | |
| "learning_rate": 1.668888888888889e-07, | |
| "loss": 0.0034, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 368.0, | |
| "eval_loss": 0.5108353495597839, | |
| "eval_runtime": 159.2225, | |
| "eval_samples_per_second": 2.688, | |
| "eval_steps_per_second": 0.088, | |
| "eval_wer": 29.488891586742742, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 371.0, | |
| "learning_rate": 1.6411111111111112e-07, | |
| "loss": 0.0034, | |
| "step": 3525 | |
| }, | |
| { | |
| "epoch": 373.0, | |
| "learning_rate": 1.6133333333333332e-07, | |
| "loss": 0.0033, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 376.0, | |
| "learning_rate": 1.5855555555555555e-07, | |
| "loss": 0.0033, | |
| "step": 3575 | |
| }, | |
| { | |
| "epoch": 378.0, | |
| "learning_rate": 1.5577777777777775e-07, | |
| "loss": 0.0033, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 381.0, | |
| "learning_rate": 1.5299999999999998e-07, | |
| "loss": 0.0032, | |
| "step": 3625 | |
| }, | |
| { | |
| "epoch": 384.0, | |
| "learning_rate": 1.5022222222222224e-07, | |
| "loss": 0.0033, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 386.0, | |
| "learning_rate": 1.4744444444444444e-07, | |
| "loss": 0.0032, | |
| "step": 3675 | |
| }, | |
| { | |
| "epoch": 389.0, | |
| "learning_rate": 1.4466666666666667e-07, | |
| "loss": 0.0032, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 392.0, | |
| "learning_rate": 1.418888888888889e-07, | |
| "loss": 0.0031, | |
| "step": 3725 | |
| }, | |
| { | |
| "epoch": 394.0, | |
| "learning_rate": 1.391111111111111e-07, | |
| "loss": 0.0031, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 397.0, | |
| "learning_rate": 1.3633333333333332e-07, | |
| "loss": 0.0031, | |
| "step": 3775 | |
| }, | |
| { | |
| "epoch": 399.0, | |
| "learning_rate": 1.3355555555555555e-07, | |
| "loss": 0.003, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 402.0, | |
| "learning_rate": 1.3077777777777775e-07, | |
| "loss": 0.003, | |
| "step": 3825 | |
| }, | |
| { | |
| "epoch": 405.0, | |
| "learning_rate": 1.28e-07, | |
| "loss": 0.003, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 407.0, | |
| "learning_rate": 1.2522222222222224e-07, | |
| "loss": 0.003, | |
| "step": 3875 | |
| }, | |
| { | |
| "epoch": 410.0, | |
| "learning_rate": 1.2244444444444444e-07, | |
| "loss": 0.0029, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 413.0, | |
| "learning_rate": 1.1966666666666667e-07, | |
| "loss": 0.003, | |
| "step": 3925 | |
| }, | |
| { | |
| "epoch": 415.0, | |
| "learning_rate": 1.1688888888888888e-07, | |
| "loss": 0.0029, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 418.0, | |
| "learning_rate": 1.141111111111111e-07, | |
| "loss": 0.0029, | |
| "step": 3975 | |
| }, | |
| { | |
| "epoch": 421.0, | |
| "learning_rate": 1.1133333333333334e-07, | |
| "loss": 0.0029, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 421.0, | |
| "eval_loss": 0.5185180306434631, | |
| "eval_runtime": 158.7036, | |
| "eval_samples_per_second": 2.697, | |
| "eval_steps_per_second": 0.088, | |
| "eval_wer": 29.501031929100403, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 423.0, | |
| "learning_rate": 1.0855555555555555e-07, | |
| "loss": 0.0028, | |
| "step": 4025 | |
| }, | |
| { | |
| "epoch": 426.0, | |
| "learning_rate": 1.0577777777777777e-07, | |
| "loss": 0.0028, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 428.0, | |
| "learning_rate": 1.03e-07, | |
| "loss": 0.0028, | |
| "step": 4075 | |
| }, | |
| { | |
| "epoch": 431.0, | |
| "learning_rate": 1.0022222222222222e-07, | |
| "loss": 0.0028, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 434.0, | |
| "learning_rate": 9.744444444444444e-08, | |
| "loss": 0.0028, | |
| "step": 4125 | |
| }, | |
| { | |
| "epoch": 436.0, | |
| "learning_rate": 9.466666666666665e-08, | |
| "loss": 0.0028, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 439.0, | |
| "learning_rate": 9.18888888888889e-08, | |
| "loss": 0.0027, | |
| "step": 4175 | |
| }, | |
| { | |
| "epoch": 442.0, | |
| "learning_rate": 8.911111111111111e-08, | |
| "loss": 0.0027, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 444.0, | |
| "learning_rate": 8.633333333333333e-08, | |
| "loss": 0.0027, | |
| "step": 4225 | |
| }, | |
| { | |
| "epoch": 447.0, | |
| "learning_rate": 8.355555555555554e-08, | |
| "loss": 0.0027, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 449.0, | |
| "learning_rate": 8.077777777777778e-08, | |
| "loss": 0.0027, | |
| "step": 4275 | |
| }, | |
| { | |
| "epoch": 452.0, | |
| "learning_rate": 7.8e-08, | |
| "loss": 0.0027, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 455.0, | |
| "learning_rate": 7.522222222222221e-08, | |
| "loss": 0.0026, | |
| "step": 4325 | |
| }, | |
| { | |
| "epoch": 457.0, | |
| "learning_rate": 7.244444444444445e-08, | |
| "loss": 0.0027, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 460.0, | |
| "learning_rate": 6.966666666666667e-08, | |
| "loss": 0.0026, | |
| "step": 4375 | |
| }, | |
| { | |
| "epoch": 463.0, | |
| "learning_rate": 6.688888888888888e-08, | |
| "loss": 0.0026, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 465.0, | |
| "learning_rate": 6.41111111111111e-08, | |
| "loss": 0.0026, | |
| "step": 4425 | |
| }, | |
| { | |
| "epoch": 468.0, | |
| "learning_rate": 6.133333333333333e-08, | |
| "loss": 0.0026, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 471.0, | |
| "learning_rate": 5.8555555555555555e-08, | |
| "loss": 0.0026, | |
| "step": 4475 | |
| }, | |
| { | |
| "epoch": 473.0, | |
| "learning_rate": 5.5777777777777777e-08, | |
| "loss": 0.0026, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 473.0, | |
| "eval_loss": 0.5236162543296814, | |
| "eval_runtime": 158.5575, | |
| "eval_samples_per_second": 2.699, | |
| "eval_steps_per_second": 0.088, | |
| "eval_wer": 29.488891586742742, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 476.0, | |
| "learning_rate": 5.3e-08, | |
| "loss": 0.0026, | |
| "step": 4525 | |
| }, | |
| { | |
| "epoch": 478.0, | |
| "learning_rate": 5.022222222222222e-08, | |
| "loss": 0.0025, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 481.0, | |
| "learning_rate": 4.744444444444444e-08, | |
| "loss": 0.0025, | |
| "step": 4575 | |
| }, | |
| { | |
| "epoch": 484.0, | |
| "learning_rate": 4.466666666666666e-08, | |
| "loss": 0.0025, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 486.0, | |
| "learning_rate": 4.188888888888889e-08, | |
| "loss": 0.0025, | |
| "step": 4625 | |
| }, | |
| { | |
| "epoch": 489.0, | |
| "learning_rate": 3.9111111111111106e-08, | |
| "loss": 0.0025, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 492.0, | |
| "learning_rate": 3.6333333333333334e-08, | |
| "loss": 0.0025, | |
| "step": 4675 | |
| }, | |
| { | |
| "epoch": 494.0, | |
| "learning_rate": 3.355555555555555e-08, | |
| "loss": 0.0025, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 497.0, | |
| "learning_rate": 3.077777777777778e-08, | |
| "loss": 0.0025, | |
| "step": 4725 | |
| }, | |
| { | |
| "epoch": 499.0, | |
| "learning_rate": 2.8e-08, | |
| "loss": 0.0025, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 502.0, | |
| "learning_rate": 2.522222222222222e-08, | |
| "loss": 0.0025, | |
| "step": 4775 | |
| }, | |
| { | |
| "epoch": 505.0, | |
| "learning_rate": 2.2444444444444442e-08, | |
| "loss": 0.0025, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 507.0, | |
| "learning_rate": 1.9666666666666664e-08, | |
| "loss": 0.0025, | |
| "step": 4825 | |
| }, | |
| { | |
| "epoch": 510.0, | |
| "learning_rate": 1.6888888888888886e-08, | |
| "loss": 0.0025, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 513.0, | |
| "learning_rate": 1.411111111111111e-08, | |
| "loss": 0.0025, | |
| "step": 4875 | |
| }, | |
| { | |
| "epoch": 515.0, | |
| "learning_rate": 1.1333333333333334e-08, | |
| "loss": 0.0025, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 518.0, | |
| "learning_rate": 8.555555555555556e-09, | |
| "loss": 0.0025, | |
| "step": 4925 | |
| }, | |
| { | |
| "epoch": 521.0, | |
| "learning_rate": 5.777777777777777e-09, | |
| "loss": 0.0025, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 523.0, | |
| "learning_rate": 3e-09, | |
| "loss": 0.0025, | |
| "step": 4975 | |
| }, | |
| { | |
| "epoch": 526.0, | |
| "learning_rate": 2.2222222222222221e-10, | |
| "loss": 0.0024, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 526.0, | |
| "eval_loss": 0.5256302952766418, | |
| "eval_runtime": 158.8325, | |
| "eval_samples_per_second": 2.695, | |
| "eval_steps_per_second": 0.088, | |
| "eval_wer": 29.537452956173365, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 526.0, | |
| "step": 5000, | |
| "total_flos": 8.946320561160192e+19, | |
| "train_loss": 0.05876407126933336, | |
| "train_runtime": 46109.0625, | |
| "train_samples_per_second": 6.94, | |
| "train_steps_per_second": 0.108 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "total_flos": 8.946320561160192e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |