| { | |
| "best_metric": 30.810565352304547, | |
| "best_model_checkpoint": "./checkpoint-2000", | |
| "epoch": 384.0016, | |
| "eval_steps": 500, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.8e-08, | |
| "loss": 1.5595, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 9.8e-08, | |
| "loss": 1.5603, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 1.4799999999999998e-07, | |
| "loss": 1.5213, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 1.98e-07, | |
| "loss": 1.4821, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 2.48e-07, | |
| "loss": 1.41, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 2.98e-07, | |
| "loss": 1.3263, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 3.4799999999999994e-07, | |
| "loss": 1.2489, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 3.98e-07, | |
| "loss": 1.1653, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 4.48e-07, | |
| "loss": 1.0787, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 4.979999999999999e-07, | |
| "loss": 1.0111, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "learning_rate": 5.48e-07, | |
| "loss": 0.9545, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "learning_rate": 5.979999999999999e-07, | |
| "loss": 0.9098, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "learning_rate": 6.48e-07, | |
| "loss": 0.8647, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "learning_rate": 6.979999999999999e-07, | |
| "loss": 0.8163, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "learning_rate": 7.48e-07, | |
| "loss": 0.7699, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "learning_rate": 7.98e-07, | |
| "loss": 0.727, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "learning_rate": 8.48e-07, | |
| "loss": 0.6914, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "learning_rate": 8.98e-07, | |
| "loss": 0.6487, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "learning_rate": 9.479999999999999e-07, | |
| "loss": 0.6132, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "learning_rate": 9.98e-07, | |
| "loss": 0.5804, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_loss": 0.7836360931396484, | |
| "eval_runtime": 223.452, | |
| "eval_samples_per_second": 4.314, | |
| "eval_steps_per_second": 0.139, | |
| "eval_wer": 36.047831680920304, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "learning_rate": 9.946666666666666e-07, | |
| "loss": 0.5424, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "learning_rate": 9.891111111111112e-07, | |
| "loss": 0.5229, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "learning_rate": 9.835555555555554e-07, | |
| "loss": 0.4847, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "learning_rate": 9.78e-07, | |
| "loss": 0.4584, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "learning_rate": 9.724444444444444e-07, | |
| "loss": 0.4327, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "learning_rate": 9.668888888888888e-07, | |
| "loss": 0.4075, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 51.0, | |
| "learning_rate": 9.613333333333334e-07, | |
| "loss": 0.3873, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 53.0, | |
| "learning_rate": 9.557777777777776e-07, | |
| "loss": 0.3639, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "learning_rate": 9.502222222222221e-07, | |
| "loss": 0.3513, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 57.0, | |
| "learning_rate": 9.446666666666666e-07, | |
| "loss": 0.3285, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 59.0, | |
| "learning_rate": 9.39111111111111e-07, | |
| "loss": 0.3143, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 61.0, | |
| "learning_rate": 9.335555555555555e-07, | |
| "loss": 0.2926, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 63.0, | |
| "learning_rate": 9.28e-07, | |
| "loss": 0.2823, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "learning_rate": 9.224444444444443e-07, | |
| "loss": 0.2668, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 67.0, | |
| "learning_rate": 9.168888888888889e-07, | |
| "loss": 0.252, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 69.0, | |
| "learning_rate": 9.113333333333333e-07, | |
| "loss": 0.2404, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 71.0, | |
| "learning_rate": 9.057777777777778e-07, | |
| "loss": 0.2254, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 73.0, | |
| "learning_rate": 9.002222222222222e-07, | |
| "loss": 0.2138, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 74.0, | |
| "learning_rate": 8.946666666666667e-07, | |
| "loss": 0.2057, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "learning_rate": 8.89111111111111e-07, | |
| "loss": 0.1934, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_loss": 0.6860684752464294, | |
| "eval_runtime": 195.5892, | |
| "eval_samples_per_second": 4.929, | |
| "eval_steps_per_second": 0.158, | |
| "eval_wer": 31.521985922954666, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 78.0, | |
| "learning_rate": 8.835555555555555e-07, | |
| "loss": 0.1809, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "learning_rate": 8.78e-07, | |
| "loss": 0.171, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 82.0, | |
| "learning_rate": 8.724444444444444e-07, | |
| "loss": 0.1647, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "learning_rate": 8.668888888888889e-07, | |
| "loss": 0.1516, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 86.0, | |
| "learning_rate": 8.613333333333332e-07, | |
| "loss": 0.147, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 88.0, | |
| "learning_rate": 8.557777777777777e-07, | |
| "loss": 0.1364, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "learning_rate": 8.502222222222221e-07, | |
| "loss": 0.1278, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 92.0, | |
| "learning_rate": 8.446666666666666e-07, | |
| "loss": 0.1219, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 94.0, | |
| "learning_rate": 8.391111111111111e-07, | |
| "loss": 0.1163, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 96.0, | |
| "learning_rate": 8.335555555555555e-07, | |
| "loss": 0.1064, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 98.0, | |
| "learning_rate": 8.28e-07, | |
| "loss": 0.1029, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 99.0, | |
| "learning_rate": 8.224444444444444e-07, | |
| "loss": 0.0957, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 101.0, | |
| "learning_rate": 8.168888888888889e-07, | |
| "loss": 0.0903, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 103.0, | |
| "learning_rate": 8.113333333333333e-07, | |
| "loss": 0.0839, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 105.0, | |
| "learning_rate": 8.057777777777778e-07, | |
| "loss": 0.0817, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 107.0, | |
| "learning_rate": 8.002222222222222e-07, | |
| "loss": 0.0745, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 109.0, | |
| "learning_rate": 7.946666666666666e-07, | |
| "loss": 0.07, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 111.0, | |
| "learning_rate": 7.891111111111111e-07, | |
| "loss": 0.0657, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 113.0, | |
| "learning_rate": 7.835555555555555e-07, | |
| "loss": 0.0629, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 115.0, | |
| "learning_rate": 7.78e-07, | |
| "loss": 0.0589, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 115.0, | |
| "eval_loss": 0.7040443420410156, | |
| "eval_runtime": 157.6277, | |
| "eval_samples_per_second": 6.116, | |
| "eval_steps_per_second": 0.197, | |
| "eval_wer": 32.44153485203965, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 117.0, | |
| "learning_rate": 7.724444444444444e-07, | |
| "loss": 0.056, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 119.0, | |
| "learning_rate": 7.668888888888888e-07, | |
| "loss": 0.0528, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 121.0, | |
| "learning_rate": 7.613333333333333e-07, | |
| "loss": 0.0498, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 123.0, | |
| "learning_rate": 7.557777777777777e-07, | |
| "loss": 0.0471, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 124.0, | |
| "learning_rate": 7.502222222222222e-07, | |
| "loss": 0.045, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 126.0, | |
| "learning_rate": 7.446666666666666e-07, | |
| "loss": 0.0429, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 128.0, | |
| "learning_rate": 7.391111111111111e-07, | |
| "loss": 0.041, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 130.0, | |
| "learning_rate": 7.335555555555554e-07, | |
| "loss": 0.0394, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 132.0, | |
| "learning_rate": 7.28e-07, | |
| "loss": 0.0374, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 134.0, | |
| "learning_rate": 7.224444444444445e-07, | |
| "loss": 0.0362, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 136.0, | |
| "learning_rate": 7.168888888888889e-07, | |
| "loss": 0.0343, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 138.0, | |
| "learning_rate": 7.113333333333334e-07, | |
| "loss": 0.0329, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 140.0, | |
| "learning_rate": 7.057777777777777e-07, | |
| "loss": 0.0322, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 142.0, | |
| "learning_rate": 7.002222222222222e-07, | |
| "loss": 0.0306, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 144.0, | |
| "learning_rate": 6.946666666666666e-07, | |
| "loss": 0.0304, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 146.0, | |
| "learning_rate": 6.891111111111111e-07, | |
| "loss": 0.0287, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 148.0, | |
| "learning_rate": 6.835555555555555e-07, | |
| "loss": 0.0281, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 149.0, | |
| "learning_rate": 6.78e-07, | |
| "loss": 0.0268, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 151.0, | |
| "learning_rate": 6.724444444444444e-07, | |
| "loss": 0.0261, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 153.0, | |
| "learning_rate": 6.668888888888888e-07, | |
| "loss": 0.0251, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 153.0, | |
| "eval_loss": 0.7221878170967102, | |
| "eval_runtime": 154.3176, | |
| "eval_samples_per_second": 6.247, | |
| "eval_steps_per_second": 0.201, | |
| "eval_wer": 30.810565352304547, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 155.0, | |
| "learning_rate": 6.613333333333333e-07, | |
| "loss": 0.0245, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 157.0, | |
| "learning_rate": 6.557777777777777e-07, | |
| "loss": 0.0237, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 159.0, | |
| "learning_rate": 6.502222222222222e-07, | |
| "loss": 0.0229, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 161.0, | |
| "learning_rate": 6.446666666666666e-07, | |
| "loss": 0.0227, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 163.0, | |
| "learning_rate": 6.39111111111111e-07, | |
| "loss": 0.0216, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 165.0, | |
| "learning_rate": 6.335555555555556e-07, | |
| "loss": 0.0211, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 167.0, | |
| "learning_rate": 6.28e-07, | |
| "loss": 0.0207, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 169.0, | |
| "learning_rate": 6.224444444444445e-07, | |
| "loss": 0.0197, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 171.0, | |
| "learning_rate": 6.168888888888889e-07, | |
| "loss": 0.0196, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 173.0, | |
| "learning_rate": 6.113333333333333e-07, | |
| "loss": 0.0191, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 174.0, | |
| "learning_rate": 6.057777777777777e-07, | |
| "loss": 0.0186, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 176.0, | |
| "learning_rate": 6.002222222222222e-07, | |
| "loss": 0.0182, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 178.0, | |
| "learning_rate": 5.946666666666667e-07, | |
| "loss": 0.0178, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 180.0, | |
| "learning_rate": 5.891111111111111e-07, | |
| "loss": 0.0172, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 182.0, | |
| "learning_rate": 5.835555555555556e-07, | |
| "loss": 0.0171, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 184.0, | |
| "learning_rate": 5.779999999999999e-07, | |
| "loss": 0.0166, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 186.0, | |
| "learning_rate": 5.724444444444444e-07, | |
| "loss": 0.0162, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 188.0, | |
| "learning_rate": 5.668888888888888e-07, | |
| "loss": 0.0159, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 190.0, | |
| "learning_rate": 5.613333333333333e-07, | |
| "loss": 0.0156, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 192.0, | |
| "learning_rate": 5.557777777777778e-07, | |
| "loss": 0.0154, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 192.0, | |
| "eval_loss": 0.7361734509468079, | |
| "eval_runtime": 154.9833, | |
| "eval_samples_per_second": 6.22, | |
| "eval_steps_per_second": 0.2, | |
| "eval_wer": 31.359267388178306, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 194.0, | |
| "learning_rate": 5.502222222222221e-07, | |
| "loss": 0.0149, | |
| "step": 2525 | |
| }, | |
| { | |
| "epoch": 196.0, | |
| "learning_rate": 5.446666666666666e-07, | |
| "loss": 0.0147, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 198.0, | |
| "learning_rate": 5.39111111111111e-07, | |
| "loss": 0.0144, | |
| "step": 2575 | |
| }, | |
| { | |
| "epoch": 199.0, | |
| "learning_rate": 5.335555555555556e-07, | |
| "loss": 0.0142, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 201.0, | |
| "learning_rate": 5.28e-07, | |
| "loss": 0.0139, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 203.0, | |
| "learning_rate": 5.224444444444445e-07, | |
| "loss": 0.0136, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 205.0, | |
| "learning_rate": 5.168888888888888e-07, | |
| "loss": 0.0133, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 207.0, | |
| "learning_rate": 5.113333333333333e-07, | |
| "loss": 0.0132, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 209.0, | |
| "learning_rate": 5.057777777777778e-07, | |
| "loss": 0.013, | |
| "step": 2725 | |
| }, | |
| { | |
| "epoch": 211.0, | |
| "learning_rate": 5.002222222222222e-07, | |
| "loss": 0.0127, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 213.0, | |
| "learning_rate": 4.946666666666666e-07, | |
| "loss": 0.0124, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 215.0, | |
| "learning_rate": 4.891111111111111e-07, | |
| "loss": 0.0123, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 217.0, | |
| "learning_rate": 4.835555555555555e-07, | |
| "loss": 0.0121, | |
| "step": 2825 | |
| }, | |
| { | |
| "epoch": 219.0, | |
| "learning_rate": 4.779999999999999e-07, | |
| "loss": 0.0119, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 221.0, | |
| "learning_rate": 4.724444444444444e-07, | |
| "loss": 0.0117, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 223.0, | |
| "learning_rate": 4.668888888888889e-07, | |
| "loss": 0.0116, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 224.0, | |
| "learning_rate": 4.613333333333333e-07, | |
| "loss": 0.0114, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 226.0, | |
| "learning_rate": 4.5577777777777774e-07, | |
| "loss": 0.0112, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 228.0, | |
| "learning_rate": 4.5022222222222225e-07, | |
| "loss": 0.0111, | |
| "step": 2975 | |
| }, | |
| { | |
| "epoch": 230.0, | |
| "learning_rate": 4.4466666666666665e-07, | |
| "loss": 0.0109, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 230.0, | |
| "eval_loss": 0.7470462918281555, | |
| "eval_runtime": 156.2204, | |
| "eval_samples_per_second": 6.171, | |
| "eval_steps_per_second": 0.198, | |
| "eval_wer": 31.760387497161886, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 232.0, | |
| "learning_rate": 4.391111111111111e-07, | |
| "loss": 0.0108, | |
| "step": 3025 | |
| }, | |
| { | |
| "epoch": 234.0, | |
| "learning_rate": 4.3355555555555556e-07, | |
| "loss": 0.0104, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 236.0, | |
| "learning_rate": 4.2799999999999997e-07, | |
| "loss": 0.0105, | |
| "step": 3075 | |
| }, | |
| { | |
| "epoch": 238.0, | |
| "learning_rate": 4.224444444444444e-07, | |
| "loss": 0.0103, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 240.0, | |
| "learning_rate": 4.1688888888888883e-07, | |
| "loss": 0.0102, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 242.0, | |
| "learning_rate": 4.113333333333333e-07, | |
| "loss": 0.01, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 244.0, | |
| "learning_rate": 4.057777777777778e-07, | |
| "loss": 0.0099, | |
| "step": 3175 | |
| }, | |
| { | |
| "epoch": 246.0, | |
| "learning_rate": 4.002222222222222e-07, | |
| "loss": 0.0098, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 248.0, | |
| "learning_rate": 3.9466666666666665e-07, | |
| "loss": 0.0096, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 249.0, | |
| "learning_rate": 3.891111111111111e-07, | |
| "loss": 0.0095, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 251.0, | |
| "learning_rate": 3.835555555555555e-07, | |
| "loss": 0.0094, | |
| "step": 3275 | |
| }, | |
| { | |
| "epoch": 253.0, | |
| "learning_rate": 3.7799999999999997e-07, | |
| "loss": 0.0093, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 255.0, | |
| "learning_rate": 3.7244444444444443e-07, | |
| "loss": 0.0091, | |
| "step": 3325 | |
| }, | |
| { | |
| "epoch": 257.0, | |
| "learning_rate": 3.668888888888889e-07, | |
| "loss": 0.0091, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 259.0, | |
| "learning_rate": 3.6133333333333334e-07, | |
| "loss": 0.009, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 261.0, | |
| "learning_rate": 3.557777777777778e-07, | |
| "loss": 0.0088, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 263.0, | |
| "learning_rate": 3.502222222222222e-07, | |
| "loss": 0.0088, | |
| "step": 3425 | |
| }, | |
| { | |
| "epoch": 265.0, | |
| "learning_rate": 3.4466666666666666e-07, | |
| "loss": 0.0087, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 267.0, | |
| "learning_rate": 3.3911111111111106e-07, | |
| "loss": 0.0085, | |
| "step": 3475 | |
| }, | |
| { | |
| "epoch": 269.0, | |
| "learning_rate": 3.335555555555555e-07, | |
| "loss": 0.0085, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 269.0, | |
| "eval_loss": 0.7561882734298706, | |
| "eval_runtime": 153.6646, | |
| "eval_samples_per_second": 6.273, | |
| "eval_steps_per_second": 0.202, | |
| "eval_wer": 31.71119352153183, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 271.0, | |
| "learning_rate": 3.28e-07, | |
| "loss": 0.0085, | |
| "step": 3525 | |
| }, | |
| { | |
| "epoch": 273.0, | |
| "learning_rate": 3.2244444444444443e-07, | |
| "loss": 0.0083, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 274.0, | |
| "learning_rate": 3.168888888888889e-07, | |
| "loss": 0.0083, | |
| "step": 3575 | |
| }, | |
| { | |
| "epoch": 276.0, | |
| "learning_rate": 3.1133333333333334e-07, | |
| "loss": 0.0081, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 278.0, | |
| "learning_rate": 3.0577777777777775e-07, | |
| "loss": 0.0081, | |
| "step": 3625 | |
| }, | |
| { | |
| "epoch": 280.0, | |
| "learning_rate": 3.002222222222222e-07, | |
| "loss": 0.008, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 282.0, | |
| "learning_rate": 2.9466666666666666e-07, | |
| "loss": 0.0078, | |
| "step": 3675 | |
| }, | |
| { | |
| "epoch": 284.0, | |
| "learning_rate": 2.8911111111111106e-07, | |
| "loss": 0.0079, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 286.0, | |
| "learning_rate": 2.8355555555555557e-07, | |
| "loss": 0.0077, | |
| "step": 3725 | |
| }, | |
| { | |
| "epoch": 288.0, | |
| "learning_rate": 2.7800000000000003e-07, | |
| "loss": 0.0077, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 290.0, | |
| "learning_rate": 2.7244444444444443e-07, | |
| "loss": 0.0076, | |
| "step": 3775 | |
| }, | |
| { | |
| "epoch": 292.0, | |
| "learning_rate": 2.668888888888889e-07, | |
| "loss": 0.0076, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 294.0, | |
| "learning_rate": 2.613333333333333e-07, | |
| "loss": 0.0075, | |
| "step": 3825 | |
| }, | |
| { | |
| "epoch": 296.0, | |
| "learning_rate": 2.5577777777777775e-07, | |
| "loss": 0.0075, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 298.0, | |
| "learning_rate": 2.502222222222222e-07, | |
| "loss": 0.0074, | |
| "step": 3875 | |
| }, | |
| { | |
| "epoch": 299.0, | |
| "learning_rate": 2.4466666666666666e-07, | |
| "loss": 0.0073, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 301.0, | |
| "learning_rate": 2.391111111111111e-07, | |
| "loss": 0.0072, | |
| "step": 3925 | |
| }, | |
| { | |
| "epoch": 303.0, | |
| "learning_rate": 2.3355555555555552e-07, | |
| "loss": 0.0072, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 305.0, | |
| "learning_rate": 2.28e-07, | |
| "loss": 0.0071, | |
| "step": 3975 | |
| }, | |
| { | |
| "epoch": 307.0, | |
| "learning_rate": 2.2244444444444443e-07, | |
| "loss": 0.0071, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 307.0, | |
| "eval_loss": 0.7629747986793518, | |
| "eval_runtime": 153.9358, | |
| "eval_samples_per_second": 6.262, | |
| "eval_steps_per_second": 0.201, | |
| "eval_wer": 31.987436615454477, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 309.0, | |
| "learning_rate": 2.1688888888888886e-07, | |
| "loss": 0.0071, | |
| "step": 4025 | |
| }, | |
| { | |
| "epoch": 311.0, | |
| "learning_rate": 2.1133333333333335e-07, | |
| "loss": 0.007, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 313.0, | |
| "learning_rate": 2.0577777777777778e-07, | |
| "loss": 0.0069, | |
| "step": 4075 | |
| }, | |
| { | |
| "epoch": 315.0, | |
| "learning_rate": 2.002222222222222e-07, | |
| "loss": 0.0069, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 317.0, | |
| "learning_rate": 1.9466666666666664e-07, | |
| "loss": 0.0069, | |
| "step": 4125 | |
| }, | |
| { | |
| "epoch": 319.0, | |
| "learning_rate": 1.8911111111111112e-07, | |
| "loss": 0.0068, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 321.0, | |
| "learning_rate": 1.8355555555555555e-07, | |
| "loss": 0.0067, | |
| "step": 4175 | |
| }, | |
| { | |
| "epoch": 323.0, | |
| "learning_rate": 1.7799999999999998e-07, | |
| "loss": 0.0067, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 324.0, | |
| "learning_rate": 1.7244444444444444e-07, | |
| "loss": 0.0067, | |
| "step": 4225 | |
| }, | |
| { | |
| "epoch": 326.0, | |
| "learning_rate": 1.668888888888889e-07, | |
| "loss": 0.0066, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 328.0, | |
| "learning_rate": 1.6133333333333332e-07, | |
| "loss": 0.0066, | |
| "step": 4275 | |
| }, | |
| { | |
| "epoch": 330.0, | |
| "learning_rate": 1.5577777777777775e-07, | |
| "loss": 0.0066, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 332.0, | |
| "learning_rate": 1.5022222222222224e-07, | |
| "loss": 0.0065, | |
| "step": 4325 | |
| }, | |
| { | |
| "epoch": 334.0, | |
| "learning_rate": 1.4466666666666667e-07, | |
| "loss": 0.0065, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 336.0, | |
| "learning_rate": 1.391111111111111e-07, | |
| "loss": 0.0065, | |
| "step": 4375 | |
| }, | |
| { | |
| "epoch": 338.0, | |
| "learning_rate": 1.3355555555555555e-07, | |
| "loss": 0.0064, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 340.0, | |
| "learning_rate": 1.28e-07, | |
| "loss": 0.0065, | |
| "step": 4425 | |
| }, | |
| { | |
| "epoch": 342.0, | |
| "learning_rate": 1.2244444444444444e-07, | |
| "loss": 0.0064, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 344.0, | |
| "learning_rate": 1.1688888888888888e-07, | |
| "loss": 0.0063, | |
| "step": 4475 | |
| }, | |
| { | |
| "epoch": 346.0, | |
| "learning_rate": 1.1133333333333334e-07, | |
| "loss": 0.0064, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 346.0, | |
| "eval_loss": 0.7674962878227234, | |
| "eval_runtime": 153.8688, | |
| "eval_samples_per_second": 6.265, | |
| "eval_steps_per_second": 0.201, | |
| "eval_wer": 32.00635737531219, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 348.0, | |
| "learning_rate": 1.0577777777777777e-07, | |
| "loss": 0.0063, | |
| "step": 4525 | |
| }, | |
| { | |
| "epoch": 349.0, | |
| "learning_rate": 1.0022222222222222e-07, | |
| "loss": 0.0063, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 351.0, | |
| "learning_rate": 9.466666666666665e-08, | |
| "loss": 0.0062, | |
| "step": 4575 | |
| }, | |
| { | |
| "epoch": 353.0, | |
| "learning_rate": 8.911111111111111e-08, | |
| "loss": 0.0062, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 355.0, | |
| "learning_rate": 8.355555555555554e-08, | |
| "loss": 0.0062, | |
| "step": 4625 | |
| }, | |
| { | |
| "epoch": 357.0, | |
| "learning_rate": 7.8e-08, | |
| "loss": 0.0062, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 359.0, | |
| "learning_rate": 7.244444444444445e-08, | |
| "loss": 0.0062, | |
| "step": 4675 | |
| }, | |
| { | |
| "epoch": 361.0, | |
| "learning_rate": 6.688888888888888e-08, | |
| "loss": 0.0061, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 363.0, | |
| "learning_rate": 6.133333333333333e-08, | |
| "loss": 0.0061, | |
| "step": 4725 | |
| }, | |
| { | |
| "epoch": 365.0, | |
| "learning_rate": 5.5777777777777777e-08, | |
| "loss": 0.0062, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 367.0, | |
| "learning_rate": 5.022222222222222e-08, | |
| "loss": 0.0061, | |
| "step": 4775 | |
| }, | |
| { | |
| "epoch": 369.0, | |
| "learning_rate": 4.466666666666666e-08, | |
| "loss": 0.0061, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 371.0, | |
| "learning_rate": 3.9111111111111106e-08, | |
| "loss": 0.0061, | |
| "step": 4825 | |
| }, | |
| { | |
| "epoch": 373.0, | |
| "learning_rate": 3.355555555555555e-08, | |
| "loss": 0.0061, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 374.0, | |
| "learning_rate": 2.8e-08, | |
| "loss": 0.0061, | |
| "step": 4875 | |
| }, | |
| { | |
| "epoch": 376.0, | |
| "learning_rate": 2.2444444444444442e-08, | |
| "loss": 0.006, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 378.0, | |
| "learning_rate": 1.6888888888888886e-08, | |
| "loss": 0.0061, | |
| "step": 4925 | |
| }, | |
| { | |
| "epoch": 380.0, | |
| "learning_rate": 1.1333333333333334e-08, | |
| "loss": 0.006, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 382.0, | |
| "learning_rate": 5.777777777777777e-09, | |
| "loss": 0.006, | |
| "step": 4975 | |
| }, | |
| { | |
| "epoch": 384.0, | |
| "learning_rate": 2.2222222222222221e-10, | |
| "loss": 0.0061, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 384.0, | |
| "eval_loss": 0.7692466974258423, | |
| "eval_runtime": 152.4767, | |
| "eval_samples_per_second": 6.322, | |
| "eval_steps_per_second": 0.203, | |
| "eval_wer": 32.066903806856885, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 384.0, | |
| "step": 5000, | |
| "total_flos": 2.028198023921664e+19, | |
| "train_loss": 0.1594879190891981, | |
| "train_runtime": 22066.1664, | |
| "train_samples_per_second": 14.502, | |
| "train_steps_per_second": 0.227 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "total_flos": 2.028198023921664e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |