| { | |
| "best_metric": 15.021535719590078, | |
| "best_model_checkpoint": "../../outdir/2022-12-08_00-59-06/checkpoint-2500", | |
| "epoch": 29.4093567251462, | |
| "global_step": 2500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 5.000000000000001e-07, | |
| "loss": 1.001, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 0.8166, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 1.5e-06, | |
| "loss": 0.6237, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 0.5312, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.4587, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 3e-06, | |
| "loss": 0.425, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 3.5e-06, | |
| "loss": 0.4055, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.3455, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 4.5e-06, | |
| "loss": 0.3192, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 5e-06, | |
| "loss": 0.3146, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 5.500000000000001e-06, | |
| "loss": 0.2732, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 6e-06, | |
| "loss": 0.2073, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "learning_rate": 6.5000000000000004e-06, | |
| "loss": 0.185, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "learning_rate": 7e-06, | |
| "loss": 0.1786, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 4.41, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.1354, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 4.7, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.1369, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "learning_rate": 8.5e-06, | |
| "loss": 0.1298, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 5.29, | |
| "learning_rate": 9e-06, | |
| "loss": 0.0926, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 5.58, | |
| "learning_rate": 9.5e-06, | |
| "loss": 0.0898, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0882, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "eval_loss": 0.267423152923584, | |
| "eval_runtime": 1165.7105, | |
| "eval_samples_per_second": 0.507, | |
| "eval_steps_per_second": 0.016, | |
| "eval_wer": 16.735481954552206, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.18, | |
| "learning_rate": 9.944444444444445e-06, | |
| "loss": 0.067, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 6.47, | |
| "learning_rate": 9.88888888888889e-06, | |
| "loss": 0.0524, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 6.76, | |
| "learning_rate": 9.833333333333333e-06, | |
| "loss": 0.049, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "learning_rate": 9.777777777777779e-06, | |
| "loss": 0.0497, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 7.35, | |
| "learning_rate": 9.722222222222223e-06, | |
| "loss": 0.026, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 7.64, | |
| "learning_rate": 9.666666666666667e-06, | |
| "loss": 0.0286, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 7.94, | |
| "learning_rate": 9.611111111111112e-06, | |
| "loss": 0.0279, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 8.23, | |
| "learning_rate": 9.555555555555556e-06, | |
| "loss": 0.0189, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 8.53, | |
| "learning_rate": 9.5e-06, | |
| "loss": 0.0161, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 8.82, | |
| "learning_rate": 9.444444444444445e-06, | |
| "loss": 0.0131, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 9.12, | |
| "learning_rate": 9.38888888888889e-06, | |
| "loss": 0.0124, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 9.41, | |
| "learning_rate": 9.333333333333334e-06, | |
| "loss": 0.0089, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 9.7, | |
| "learning_rate": 9.277777777777778e-06, | |
| "loss": 0.0087, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 9.99, | |
| "learning_rate": 9.222222222222224e-06, | |
| "loss": 0.0081, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 10.29, | |
| "learning_rate": 9.166666666666666e-06, | |
| "loss": 0.0059, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 10.58, | |
| "learning_rate": 9.111111111111112e-06, | |
| "loss": 0.0059, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 10.88, | |
| "learning_rate": 9.055555555555556e-06, | |
| "loss": 0.0047, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 11.18, | |
| "learning_rate": 9e-06, | |
| "loss": 0.0036, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 11.47, | |
| "learning_rate": 8.944444444444446e-06, | |
| "loss": 0.0027, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 11.76, | |
| "learning_rate": 8.888888888888888e-06, | |
| "loss": 0.0026, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 11.76, | |
| "eval_loss": 0.3508181571960449, | |
| "eval_runtime": 1146.9726, | |
| "eval_samples_per_second": 0.515, | |
| "eval_steps_per_second": 0.017, | |
| "eval_wer": 15.372048121194117, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 12.06, | |
| "learning_rate": 8.833333333333334e-06, | |
| "loss": 0.0028, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 12.35, | |
| "learning_rate": 8.777777777777778e-06, | |
| "loss": 0.0015, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 12.64, | |
| "learning_rate": 8.722222222222224e-06, | |
| "loss": 0.0017, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 12.94, | |
| "learning_rate": 8.666666666666668e-06, | |
| "loss": 0.0019, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 13.23, | |
| "learning_rate": 8.611111111111112e-06, | |
| "loss": 0.0011, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 13.53, | |
| "learning_rate": 8.555555555555556e-06, | |
| "loss": 0.0011, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 13.82, | |
| "learning_rate": 8.5e-06, | |
| "loss": 0.0012, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 14.12, | |
| "learning_rate": 8.444444444444446e-06, | |
| "loss": 0.0012, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 14.41, | |
| "learning_rate": 8.38888888888889e-06, | |
| "loss": 0.0012, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 14.7, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.001, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 14.99, | |
| "learning_rate": 8.277777777777778e-06, | |
| "loss": 0.0017, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 15.29, | |
| "learning_rate": 8.222222222222222e-06, | |
| "loss": 0.002, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 15.58, | |
| "learning_rate": 8.166666666666668e-06, | |
| "loss": 0.0019, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 15.88, | |
| "learning_rate": 8.111111111111112e-06, | |
| "loss": 0.0015, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 16.18, | |
| "learning_rate": 8.055555555555557e-06, | |
| "loss": 0.0016, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 16.47, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.0013, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 16.76, | |
| "learning_rate": 7.944444444444445e-06, | |
| "loss": 0.0016, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 17.06, | |
| "learning_rate": 7.88888888888889e-06, | |
| "loss": 0.0013, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 17.35, | |
| "learning_rate": 7.833333333333333e-06, | |
| "loss": 0.0013, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 17.64, | |
| "learning_rate": 7.77777777777778e-06, | |
| "loss": 0.0012, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 17.64, | |
| "eval_loss": 0.3920586407184601, | |
| "eval_runtime": 1148.0874, | |
| "eval_samples_per_second": 0.515, | |
| "eval_steps_per_second": 0.017, | |
| "eval_wer": 15.615624535868111, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 17.94, | |
| "learning_rate": 7.722222222222223e-06, | |
| "loss": 0.0015, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 18.23, | |
| "learning_rate": 7.666666666666667e-06, | |
| "loss": 0.0018, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 18.53, | |
| "learning_rate": 7.611111111111111e-06, | |
| "loss": 0.0018, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 18.82, | |
| "learning_rate": 7.555555555555556e-06, | |
| "loss": 0.0016, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 19.12, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.0015, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 19.41, | |
| "learning_rate": 7.444444444444445e-06, | |
| "loss": 0.0017, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 19.7, | |
| "learning_rate": 7.38888888888889e-06, | |
| "loss": 0.0014, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 19.99, | |
| "learning_rate": 7.335555555555556e-06, | |
| "loss": 0.0021, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 20.29, | |
| "learning_rate": 7.280000000000001e-06, | |
| "loss": 0.0024, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 20.58, | |
| "learning_rate": 7.224444444444445e-06, | |
| "loss": 0.0024, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 20.88, | |
| "learning_rate": 7.1688888888888895e-06, | |
| "loss": 0.0024, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 21.18, | |
| "learning_rate": 7.113333333333334e-06, | |
| "loss": 0.0023, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 21.47, | |
| "learning_rate": 7.057777777777778e-06, | |
| "loss": 0.002, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 21.76, | |
| "learning_rate": 7.0022222222222225e-06, | |
| "loss": 0.0027, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 22.06, | |
| "learning_rate": 6.946666666666667e-06, | |
| "loss": 0.0024, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 22.35, | |
| "learning_rate": 6.891111111111111e-06, | |
| "loss": 0.0015, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 22.64, | |
| "learning_rate": 6.835555555555556e-06, | |
| "loss": 0.002, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 22.94, | |
| "learning_rate": 6.780000000000001e-06, | |
| "loss": 0.0014, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 23.23, | |
| "learning_rate": 6.724444444444444e-06, | |
| "loss": 0.001, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 23.53, | |
| "learning_rate": 6.668888888888889e-06, | |
| "loss": 0.0009, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 23.53, | |
| "eval_loss": 0.4076804518699646, | |
| "eval_runtime": 1148.5839, | |
| "eval_samples_per_second": 0.515, | |
| "eval_steps_per_second": 0.017, | |
| "eval_wer": 15.42848655874053, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 23.82, | |
| "learning_rate": 6.613333333333334e-06, | |
| "loss": 0.0008, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 24.12, | |
| "learning_rate": 6.557777777777778e-06, | |
| "loss": 0.001, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 24.41, | |
| "learning_rate": 6.502222222222223e-06, | |
| "loss": 0.0009, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 24.7, | |
| "learning_rate": 6.446666666666668e-06, | |
| "loss": 0.0005, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 24.99, | |
| "learning_rate": 6.391111111111111e-06, | |
| "loss": 0.0004, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 25.29, | |
| "learning_rate": 6.335555555555556e-06, | |
| "loss": 0.0003, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 25.58, | |
| "learning_rate": 6.280000000000001e-06, | |
| "loss": 0.0003, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 25.88, | |
| "learning_rate": 6.224444444444445e-06, | |
| "loss": 0.0003, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 26.18, | |
| "learning_rate": 6.16888888888889e-06, | |
| "loss": 0.0003, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 26.47, | |
| "learning_rate": 6.113333333333333e-06, | |
| "loss": 0.0002, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 26.76, | |
| "learning_rate": 6.057777777777778e-06, | |
| "loss": 0.0002, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 27.06, | |
| "learning_rate": 6.002222222222223e-06, | |
| "loss": 0.0002, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 27.35, | |
| "learning_rate": 5.946666666666668e-06, | |
| "loss": 0.0002, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 27.64, | |
| "learning_rate": 5.891111111111112e-06, | |
| "loss": 0.0002, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 27.94, | |
| "learning_rate": 5.8355555555555565e-06, | |
| "loss": 0.0002, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 28.23, | |
| "learning_rate": 5.78e-06, | |
| "loss": 0.0002, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 28.53, | |
| "learning_rate": 5.724444444444445e-06, | |
| "loss": 0.0002, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 28.82, | |
| "learning_rate": 5.6688888888888895e-06, | |
| "loss": 0.0002, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 29.12, | |
| "learning_rate": 5.613333333333334e-06, | |
| "loss": 0.0002, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 29.41, | |
| "learning_rate": 5.557777777777778e-06, | |
| "loss": 0.0002, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 29.41, | |
| "eval_loss": 0.42684832215309143, | |
| "eval_runtime": 1155.7652, | |
| "eval_samples_per_second": 0.511, | |
| "eval_steps_per_second": 0.016, | |
| "eval_wer": 15.021535719590078, | |
| "step": 2500 | |
| } | |
| ], | |
| "max_steps": 5000, | |
| "num_train_epochs": 59, | |
| "total_flos": 2.31286769086464e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |