| { | |
| "best_metric": 28.384476534296027, | |
| "best_model_checkpoint": "./checkpoint-1500", | |
| "epoch": 454.0012, | |
| "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": 4.5999999999999995e-08, | |
| "loss": 1.0681, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 9.6e-08, | |
| "loss": 1.0453, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 1.4599999999999998e-07, | |
| "loss": 1.0136, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 1.96e-07, | |
| "loss": 0.9542, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 2.46e-07, | |
| "loss": 0.9039, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 2.9599999999999995e-07, | |
| "loss": 0.8261, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 3.4599999999999995e-07, | |
| "loss": 0.7581, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 3.96e-07, | |
| "loss": 0.712, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 4.46e-07, | |
| "loss": 0.6576, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "learning_rate": 4.96e-07, | |
| "loss": 0.623, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "learning_rate": 5.46e-07, | |
| "loss": 0.621, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "learning_rate": 5.96e-07, | |
| "loss": 0.5673, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "learning_rate": 6.46e-07, | |
| "loss": 0.5371, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "learning_rate": 6.959999999999999e-07, | |
| "loss": 0.5144, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "learning_rate": 7.459999999999999e-07, | |
| "loss": 0.4976, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "learning_rate": 7.96e-07, | |
| "loss": 0.4514, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "learning_rate": 8.459999999999999e-07, | |
| "loss": 0.443, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "learning_rate": 8.96e-07, | |
| "loss": 0.392, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "learning_rate": 9.459999999999999e-07, | |
| "loss": 0.3709, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "learning_rate": 9.959999999999999e-07, | |
| "loss": 0.3503, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_loss": 0.4873710572719574, | |
| "eval_runtime": 132.2348, | |
| "eval_samples_per_second": 5.619, | |
| "eval_steps_per_second": 0.181, | |
| "eval_wer": 29.18170878459687, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "learning_rate": 9.948888888888888e-07, | |
| "loss": 0.3216, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "learning_rate": 9.893333333333332e-07, | |
| "loss": 0.2992, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "learning_rate": 9.837777777777778e-07, | |
| "loss": 0.2763, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "learning_rate": 9.782222222222222e-07, | |
| "loss": 0.2571, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "learning_rate": 9.726666666666666e-07, | |
| "loss": 0.2379, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 59.0, | |
| "learning_rate": 9.67111111111111e-07, | |
| "loss": 0.2156, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 61.0, | |
| "learning_rate": 9.615555555555556e-07, | |
| "loss": 0.2009, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 63.0, | |
| "learning_rate": 9.559999999999998e-07, | |
| "loss": 0.1852, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "learning_rate": 9.504444444444443e-07, | |
| "loss": 0.1737, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "learning_rate": 9.448888888888888e-07, | |
| "loss": 0.1643, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "learning_rate": 9.393333333333334e-07, | |
| "loss": 0.1511, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "learning_rate": 9.337777777777778e-07, | |
| "loss": 0.1367, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 74.0, | |
| "learning_rate": 9.282222222222222e-07, | |
| "loss": 0.1295, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 77.0, | |
| "learning_rate": 9.226666666666666e-07, | |
| "loss": 0.1155, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 79.0, | |
| "learning_rate": 9.171111111111111e-07, | |
| "loss": 0.1073, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 81.0, | |
| "learning_rate": 9.115555555555555e-07, | |
| "loss": 0.0983, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "learning_rate": 9.06e-07, | |
| "loss": 0.0918, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 86.0, | |
| "learning_rate": 9.004444444444445e-07, | |
| "loss": 0.0832, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 88.0, | |
| "learning_rate": 8.948888888888888e-07, | |
| "loss": 0.0779, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "learning_rate": 8.893333333333333e-07, | |
| "loss": 0.0695, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "eval_loss": 0.495993971824646, | |
| "eval_runtime": 93.9292, | |
| "eval_samples_per_second": 7.91, | |
| "eval_steps_per_second": 0.256, | |
| "eval_wer": 28.459687123947052, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 93.0, | |
| "learning_rate": 8.837777777777777e-07, | |
| "loss": 0.0654, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "learning_rate": 8.782222222222222e-07, | |
| "loss": 0.0603, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 97.0, | |
| "learning_rate": 8.726666666666666e-07, | |
| "loss": 0.0561, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 99.0, | |
| "learning_rate": 8.671111111111111e-07, | |
| "loss": 0.0518, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 102.0, | |
| "learning_rate": 8.615555555555555e-07, | |
| "loss": 0.048, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 104.0, | |
| "learning_rate": 8.559999999999999e-07, | |
| "loss": 0.0463, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 106.0, | |
| "learning_rate": 8.504444444444444e-07, | |
| "loss": 0.0425, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 109.0, | |
| "learning_rate": 8.448888888888888e-07, | |
| "loss": 0.0406, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 111.0, | |
| "learning_rate": 8.393333333333334e-07, | |
| "loss": 0.0387, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 113.0, | |
| "learning_rate": 8.337777777777777e-07, | |
| "loss": 0.0367, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 115.0, | |
| "learning_rate": 8.282222222222222e-07, | |
| "loss": 0.0348, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 118.0, | |
| "learning_rate": 8.226666666666666e-07, | |
| "loss": 0.0332, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 120.0, | |
| "learning_rate": 8.171111111111111e-07, | |
| "loss": 0.0321, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 122.0, | |
| "learning_rate": 8.115555555555556e-07, | |
| "loss": 0.0306, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 124.0, | |
| "learning_rate": 8.06e-07, | |
| "loss": 0.0292, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 127.0, | |
| "learning_rate": 8.004444444444444e-07, | |
| "loss": 0.028, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 129.0, | |
| "learning_rate": 7.948888888888888e-07, | |
| "loss": 0.0274, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 131.0, | |
| "learning_rate": 7.893333333333333e-07, | |
| "loss": 0.0262, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 134.0, | |
| "learning_rate": 7.837777777777777e-07, | |
| "loss": 0.0255, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 136.0, | |
| "learning_rate": 7.782222222222222e-07, | |
| "loss": 0.0243, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 136.0, | |
| "eval_loss": 0.5194724202156067, | |
| "eval_runtime": 92.8935, | |
| "eval_samples_per_second": 7.998, | |
| "eval_steps_per_second": 0.258, | |
| "eval_wer": 28.384476534296027, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 138.0, | |
| "learning_rate": 7.726666666666666e-07, | |
| "loss": 0.0235, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 140.0, | |
| "learning_rate": 7.67111111111111e-07, | |
| "loss": 0.0228, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 143.0, | |
| "learning_rate": 7.615555555555555e-07, | |
| "loss": 0.0219, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 145.0, | |
| "learning_rate": 7.559999999999999e-07, | |
| "loss": 0.0217, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 147.0, | |
| "learning_rate": 7.504444444444444e-07, | |
| "loss": 0.0205, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 149.0, | |
| "learning_rate": 7.448888888888889e-07, | |
| "loss": 0.0201, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 152.0, | |
| "learning_rate": 7.393333333333333e-07, | |
| "loss": 0.0197, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 154.0, | |
| "learning_rate": 7.337777777777778e-07, | |
| "loss": 0.0191, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 156.0, | |
| "learning_rate": 7.282222222222222e-07, | |
| "loss": 0.0186, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 159.0, | |
| "learning_rate": 7.226666666666667e-07, | |
| "loss": 0.0179, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 161.0, | |
| "learning_rate": 7.171111111111111e-07, | |
| "loss": 0.0175, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 163.0, | |
| "learning_rate": 7.115555555555556e-07, | |
| "loss": 0.0173, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 165.0, | |
| "learning_rate": 7.059999999999999e-07, | |
| "loss": 0.0168, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 168.0, | |
| "learning_rate": 7.004444444444444e-07, | |
| "loss": 0.0164, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 170.0, | |
| "learning_rate": 6.948888888888889e-07, | |
| "loss": 0.0159, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 172.0, | |
| "learning_rate": 6.893333333333333e-07, | |
| "loss": 0.0156, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 174.0, | |
| "learning_rate": 6.837777777777778e-07, | |
| "loss": 0.0152, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 177.0, | |
| "learning_rate": 6.782222222222221e-07, | |
| "loss": 0.0151, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 179.0, | |
| "learning_rate": 6.726666666666666e-07, | |
| "loss": 0.0145, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 181.0, | |
| "learning_rate": 6.67111111111111e-07, | |
| "loss": 0.0145, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 181.0, | |
| "eval_loss": 0.5334057211875916, | |
| "eval_runtime": 93.7728, | |
| "eval_samples_per_second": 7.923, | |
| "eval_steps_per_second": 0.256, | |
| "eval_wer": 28.64771359807461, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 184.0, | |
| "learning_rate": 6.615555555555555e-07, | |
| "loss": 0.0141, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 186.0, | |
| "learning_rate": 6.56e-07, | |
| "loss": 0.0137, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 188.0, | |
| "learning_rate": 6.504444444444445e-07, | |
| "loss": 0.0137, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 190.0, | |
| "learning_rate": 6.448888888888889e-07, | |
| "loss": 0.0131, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 193.0, | |
| "learning_rate": 6.393333333333333e-07, | |
| "loss": 0.013, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 195.0, | |
| "learning_rate": 6.337777777777778e-07, | |
| "loss": 0.0127, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 197.0, | |
| "learning_rate": 6.282222222222222e-07, | |
| "loss": 0.0125, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 199.0, | |
| "learning_rate": 6.226666666666667e-07, | |
| "loss": 0.0124, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 202.0, | |
| "learning_rate": 6.171111111111111e-07, | |
| "loss": 0.0119, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 204.0, | |
| "learning_rate": 6.115555555555555e-07, | |
| "loss": 0.0118, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 206.0, | |
| "learning_rate": 6.06e-07, | |
| "loss": 0.0117, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 209.0, | |
| "learning_rate": 6.004444444444444e-07, | |
| "loss": 0.0113, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 211.0, | |
| "learning_rate": 5.948888888888889e-07, | |
| "loss": 0.0113, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 213.0, | |
| "learning_rate": 5.893333333333333e-07, | |
| "loss": 0.011, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 215.0, | |
| "learning_rate": 5.837777777777777e-07, | |
| "loss": 0.0109, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 218.0, | |
| "learning_rate": 5.782222222222221e-07, | |
| "loss": 0.0108, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 220.0, | |
| "learning_rate": 5.726666666666666e-07, | |
| "loss": 0.0105, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 222.0, | |
| "learning_rate": 5.671111111111111e-07, | |
| "loss": 0.0102, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 224.0, | |
| "learning_rate": 5.615555555555555e-07, | |
| "loss": 0.0102, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 227.0, | |
| "learning_rate": 5.560000000000001e-07, | |
| "loss": 0.0101, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 227.0, | |
| "eval_loss": 0.5453755259513855, | |
| "eval_runtime": 92.8434, | |
| "eval_samples_per_second": 8.003, | |
| "eval_steps_per_second": 0.258, | |
| "eval_wer": 28.67779783393502, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 229.0, | |
| "learning_rate": 5.504444444444444e-07, | |
| "loss": 0.0098, | |
| "step": 2525 | |
| }, | |
| { | |
| "epoch": 231.0, | |
| "learning_rate": 5.448888888888889e-07, | |
| "loss": 0.0096, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 234.0, | |
| "learning_rate": 5.393333333333333e-07, | |
| "loss": 0.0096, | |
| "step": 2575 | |
| }, | |
| { | |
| "epoch": 236.0, | |
| "learning_rate": 5.337777777777778e-07, | |
| "loss": 0.0093, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 238.0, | |
| "learning_rate": 5.282222222222223e-07, | |
| "loss": 0.0094, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 240.0, | |
| "learning_rate": 5.226666666666666e-07, | |
| "loss": 0.0092, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 243.0, | |
| "learning_rate": 5.171111111111111e-07, | |
| "loss": 0.0091, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 245.0, | |
| "learning_rate": 5.115555555555555e-07, | |
| "loss": 0.0089, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 247.0, | |
| "learning_rate": 5.06e-07, | |
| "loss": 0.0088, | |
| "step": 2725 | |
| }, | |
| { | |
| "epoch": 249.0, | |
| "learning_rate": 5.004444444444444e-07, | |
| "loss": 0.0088, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 252.0, | |
| "learning_rate": 4.948888888888888e-07, | |
| "loss": 0.0085, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 254.0, | |
| "learning_rate": 4.893333333333333e-07, | |
| "loss": 0.0085, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 256.0, | |
| "learning_rate": 4.837777777777777e-07, | |
| "loss": 0.0084, | |
| "step": 2825 | |
| }, | |
| { | |
| "epoch": 259.0, | |
| "learning_rate": 4.782222222222222e-07, | |
| "loss": 0.0081, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 261.0, | |
| "learning_rate": 4.7266666666666664e-07, | |
| "loss": 0.0081, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 263.0, | |
| "learning_rate": 4.6711111111111104e-07, | |
| "loss": 0.0081, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 265.0, | |
| "learning_rate": 4.6155555555555555e-07, | |
| "loss": 0.0079, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 268.0, | |
| "learning_rate": 4.56e-07, | |
| "loss": 0.0079, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 270.0, | |
| "learning_rate": 4.504444444444444e-07, | |
| "loss": 0.0077, | |
| "step": 2975 | |
| }, | |
| { | |
| "epoch": 272.0, | |
| "learning_rate": 4.4488888888888887e-07, | |
| "loss": 0.0077, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 272.0, | |
| "eval_loss": 0.5547551512718201, | |
| "eval_runtime": 93.7417, | |
| "eval_samples_per_second": 7.926, | |
| "eval_steps_per_second": 0.256, | |
| "eval_wer": 28.692839951865224, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 274.0, | |
| "learning_rate": 4.393333333333333e-07, | |
| "loss": 0.0076, | |
| "step": 3025 | |
| }, | |
| { | |
| "epoch": 277.0, | |
| "learning_rate": 4.3377777777777773e-07, | |
| "loss": 0.0075, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 279.0, | |
| "learning_rate": 4.282222222222222e-07, | |
| "loss": 0.0075, | |
| "step": 3075 | |
| }, | |
| { | |
| "epoch": 281.0, | |
| "learning_rate": 4.226666666666667e-07, | |
| "loss": 0.0073, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 284.0, | |
| "learning_rate": 4.171111111111111e-07, | |
| "loss": 0.0072, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 286.0, | |
| "learning_rate": 4.1155555555555555e-07, | |
| "loss": 0.0071, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 288.0, | |
| "learning_rate": 4.06e-07, | |
| "loss": 0.0071, | |
| "step": 3175 | |
| }, | |
| { | |
| "epoch": 290.0, | |
| "learning_rate": 4.004444444444444e-07, | |
| "loss": 0.0071, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 293.0, | |
| "learning_rate": 3.9488888888888887e-07, | |
| "loss": 0.007, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 295.0, | |
| "learning_rate": 3.8933333333333327e-07, | |
| "loss": 0.0069, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 297.0, | |
| "learning_rate": 3.837777777777778e-07, | |
| "loss": 0.0067, | |
| "step": 3275 | |
| }, | |
| { | |
| "epoch": 299.0, | |
| "learning_rate": 3.7822222222222224e-07, | |
| "loss": 0.0067, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 302.0, | |
| "learning_rate": 3.7266666666666664e-07, | |
| "loss": 0.0067, | |
| "step": 3325 | |
| }, | |
| { | |
| "epoch": 304.0, | |
| "learning_rate": 3.671111111111111e-07, | |
| "loss": 0.0066, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 306.0, | |
| "learning_rate": 3.6155555555555555e-07, | |
| "loss": 0.0065, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 309.0, | |
| "learning_rate": 3.5599999999999996e-07, | |
| "loss": 0.0065, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 311.0, | |
| "learning_rate": 3.504444444444444e-07, | |
| "loss": 0.0064, | |
| "step": 3425 | |
| }, | |
| { | |
| "epoch": 313.0, | |
| "learning_rate": 3.4488888888888887e-07, | |
| "loss": 0.0063, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 315.0, | |
| "learning_rate": 3.3933333333333333e-07, | |
| "loss": 0.0062, | |
| "step": 3475 | |
| }, | |
| { | |
| "epoch": 318.0, | |
| "learning_rate": 3.337777777777778e-07, | |
| "loss": 0.0063, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 318.0, | |
| "eval_loss": 0.5624770522117615, | |
| "eval_runtime": 93.8649, | |
| "eval_samples_per_second": 7.916, | |
| "eval_steps_per_second": 0.256, | |
| "eval_wer": 28.707882069795428, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 320.0, | |
| "learning_rate": 3.2822222222222224e-07, | |
| "loss": 0.0061, | |
| "step": 3525 | |
| }, | |
| { | |
| "epoch": 322.0, | |
| "learning_rate": 3.2266666666666664e-07, | |
| "loss": 0.006, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 324.0, | |
| "learning_rate": 3.171111111111111e-07, | |
| "loss": 0.0061, | |
| "step": 3575 | |
| }, | |
| { | |
| "epoch": 327.0, | |
| "learning_rate": 3.115555555555555e-07, | |
| "loss": 0.006, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 329.0, | |
| "learning_rate": 3.0599999999999996e-07, | |
| "loss": 0.006, | |
| "step": 3625 | |
| }, | |
| { | |
| "epoch": 331.0, | |
| "learning_rate": 3.0044444444444447e-07, | |
| "loss": 0.0059, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 334.0, | |
| "learning_rate": 2.948888888888889e-07, | |
| "loss": 0.0059, | |
| "step": 3675 | |
| }, | |
| { | |
| "epoch": 336.0, | |
| "learning_rate": 2.8933333333333333e-07, | |
| "loss": 0.0058, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 338.0, | |
| "learning_rate": 2.837777777777778e-07, | |
| "loss": 0.0058, | |
| "step": 3725 | |
| }, | |
| { | |
| "epoch": 340.0, | |
| "learning_rate": 2.782222222222222e-07, | |
| "loss": 0.0058, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 343.0, | |
| "learning_rate": 2.7266666666666665e-07, | |
| "loss": 0.0057, | |
| "step": 3775 | |
| }, | |
| { | |
| "epoch": 345.0, | |
| "learning_rate": 2.671111111111111e-07, | |
| "loss": 0.0057, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 347.0, | |
| "learning_rate": 2.615555555555555e-07, | |
| "loss": 0.0055, | |
| "step": 3825 | |
| }, | |
| { | |
| "epoch": 349.0, | |
| "learning_rate": 2.56e-07, | |
| "loss": 0.0056, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 352.0, | |
| "learning_rate": 2.5044444444444447e-07, | |
| "loss": 0.0055, | |
| "step": 3875 | |
| }, | |
| { | |
| "epoch": 354.0, | |
| "learning_rate": 2.448888888888889e-07, | |
| "loss": 0.0055, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 356.0, | |
| "learning_rate": 2.3933333333333333e-07, | |
| "loss": 0.0054, | |
| "step": 3925 | |
| }, | |
| { | |
| "epoch": 359.0, | |
| "learning_rate": 2.3377777777777776e-07, | |
| "loss": 0.0055, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 361.0, | |
| "learning_rate": 2.282222222222222e-07, | |
| "loss": 0.0054, | |
| "step": 3975 | |
| }, | |
| { | |
| "epoch": 363.0, | |
| "learning_rate": 2.2266666666666668e-07, | |
| "loss": 0.0054, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 363.0, | |
| "eval_loss": 0.5683514475822449, | |
| "eval_runtime": 96.5765, | |
| "eval_samples_per_second": 7.693, | |
| "eval_steps_per_second": 0.249, | |
| "eval_wer": 29.02376654632972, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 365.0, | |
| "learning_rate": 2.171111111111111e-07, | |
| "loss": 0.0053, | |
| "step": 4025 | |
| }, | |
| { | |
| "epoch": 368.0, | |
| "learning_rate": 2.1155555555555554e-07, | |
| "loss": 0.0053, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 370.0, | |
| "learning_rate": 2.06e-07, | |
| "loss": 0.0052, | |
| "step": 4075 | |
| }, | |
| { | |
| "epoch": 372.0, | |
| "learning_rate": 2.0044444444444445e-07, | |
| "loss": 0.0052, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 374.0, | |
| "learning_rate": 1.9488888888888888e-07, | |
| "loss": 0.0052, | |
| "step": 4125 | |
| }, | |
| { | |
| "epoch": 377.0, | |
| "learning_rate": 1.893333333333333e-07, | |
| "loss": 0.0051, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 379.0, | |
| "learning_rate": 1.837777777777778e-07, | |
| "loss": 0.0052, | |
| "step": 4175 | |
| }, | |
| { | |
| "epoch": 381.0, | |
| "learning_rate": 1.7822222222222222e-07, | |
| "loss": 0.0051, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 384.0, | |
| "learning_rate": 1.7266666666666665e-07, | |
| "loss": 0.0051, | |
| "step": 4225 | |
| }, | |
| { | |
| "epoch": 386.0, | |
| "learning_rate": 1.6711111111111108e-07, | |
| "loss": 0.005, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 388.0, | |
| "learning_rate": 1.6155555555555556e-07, | |
| "loss": 0.005, | |
| "step": 4275 | |
| }, | |
| { | |
| "epoch": 390.0, | |
| "learning_rate": 1.56e-07, | |
| "loss": 0.005, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 393.0, | |
| "learning_rate": 1.5044444444444442e-07, | |
| "loss": 0.005, | |
| "step": 4325 | |
| }, | |
| { | |
| "epoch": 395.0, | |
| "learning_rate": 1.448888888888889e-07, | |
| "loss": 0.0049, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 397.0, | |
| "learning_rate": 1.3933333333333334e-07, | |
| "loss": 0.005, | |
| "step": 4375 | |
| }, | |
| { | |
| "epoch": 399.0, | |
| "learning_rate": 1.3377777777777777e-07, | |
| "loss": 0.0049, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 402.0, | |
| "learning_rate": 1.282222222222222e-07, | |
| "loss": 0.0049, | |
| "step": 4425 | |
| }, | |
| { | |
| "epoch": 404.0, | |
| "learning_rate": 1.2266666666666665e-07, | |
| "loss": 0.0048, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 406.0, | |
| "learning_rate": 1.1711111111111111e-07, | |
| "loss": 0.0048, | |
| "step": 4475 | |
| }, | |
| { | |
| "epoch": 409.0, | |
| "learning_rate": 1.1155555555555555e-07, | |
| "loss": 0.0048, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 409.0, | |
| "eval_loss": 0.5727224946022034, | |
| "eval_runtime": 96.7753, | |
| "eval_samples_per_second": 7.678, | |
| "eval_steps_per_second": 0.248, | |
| "eval_wer": 28.925992779783392, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 411.0, | |
| "learning_rate": 1.06e-07, | |
| "loss": 0.0048, | |
| "step": 4525 | |
| }, | |
| { | |
| "epoch": 413.0, | |
| "learning_rate": 1.0044444444444444e-07, | |
| "loss": 0.0048, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 415.0, | |
| "learning_rate": 9.488888888888888e-08, | |
| "loss": 0.0048, | |
| "step": 4575 | |
| }, | |
| { | |
| "epoch": 418.0, | |
| "learning_rate": 8.933333333333333e-08, | |
| "loss": 0.0048, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 420.0, | |
| "learning_rate": 8.377777777777778e-08, | |
| "loss": 0.0048, | |
| "step": 4625 | |
| }, | |
| { | |
| "epoch": 422.0, | |
| "learning_rate": 7.822222222222221e-08, | |
| "loss": 0.0048, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 424.0, | |
| "learning_rate": 7.266666666666667e-08, | |
| "loss": 0.0047, | |
| "step": 4675 | |
| }, | |
| { | |
| "epoch": 427.0, | |
| "learning_rate": 6.71111111111111e-08, | |
| "loss": 0.0047, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 429.0, | |
| "learning_rate": 6.155555555555556e-08, | |
| "loss": 0.0047, | |
| "step": 4725 | |
| }, | |
| { | |
| "epoch": 431.0, | |
| "learning_rate": 5.6e-08, | |
| "loss": 0.0047, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 434.0, | |
| "learning_rate": 5.044444444444444e-08, | |
| "loss": 0.0047, | |
| "step": 4775 | |
| }, | |
| { | |
| "epoch": 436.0, | |
| "learning_rate": 4.4888888888888885e-08, | |
| "loss": 0.0047, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 438.0, | |
| "learning_rate": 3.933333333333333e-08, | |
| "loss": 0.0047, | |
| "step": 4825 | |
| }, | |
| { | |
| "epoch": 440.0, | |
| "learning_rate": 3.377777777777777e-08, | |
| "loss": 0.0047, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 443.0, | |
| "learning_rate": 2.822222222222222e-08, | |
| "loss": 0.0046, | |
| "step": 4875 | |
| }, | |
| { | |
| "epoch": 445.0, | |
| "learning_rate": 2.2666666666666668e-08, | |
| "loss": 0.0047, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 447.0, | |
| "learning_rate": 1.711111111111111e-08, | |
| "loss": 0.0046, | |
| "step": 4925 | |
| }, | |
| { | |
| "epoch": 449.0, | |
| "learning_rate": 1.1555555555555554e-08, | |
| "loss": 0.0046, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 452.0, | |
| "learning_rate": 6e-09, | |
| "loss": 0.0047, | |
| "step": 4975 | |
| }, | |
| { | |
| "epoch": 454.0, | |
| "learning_rate": 4.4444444444444443e-10, | |
| "loss": 0.0046, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 454.0, | |
| "eval_loss": 0.5742946267127991, | |
| "eval_runtime": 96.7359, | |
| "eval_samples_per_second": 7.681, | |
| "eval_steps_per_second": 0.248, | |
| "eval_wer": 28.925992779783392, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 454.0, | |
| "step": 5000, | |
| "total_flos": 1.990125206175744e+19, | |
| "train_loss": 0.09349360605478287, | |
| "train_runtime": 19634.3192, | |
| "train_samples_per_second": 16.298, | |
| "train_steps_per_second": 0.255 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "total_flos": 1.990125206175744e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |