| { | |
| "best_metric": 20.513027345325483, | |
| "best_model_checkpoint": "gdrive/MyDrive/ScalableML/Lab2/model/checkpoint-2000", | |
| "epoch": 2.5873221216041395, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 4.2000000000000006e-07, | |
| "loss": 4.1984, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 9.200000000000001e-07, | |
| "loss": 3.3699, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 1.42e-06, | |
| "loss": 2.4858, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 1.9200000000000003e-06, | |
| "loss": 1.4969, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 2.42e-06, | |
| "loss": 1.1419, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 2.92e-06, | |
| "loss": 1.0148, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 3.4200000000000007e-06, | |
| "loss": 0.9297, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 3.920000000000001e-06, | |
| "loss": 0.8621, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 4.42e-06, | |
| "loss": 0.712, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 4.92e-06, | |
| "loss": 0.6186, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 5.420000000000001e-06, | |
| "loss": 0.4717, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 5.92e-06, | |
| "loss": 0.3114, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 6.42e-06, | |
| "loss": 0.3384, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 6.92e-06, | |
| "loss": 0.2962, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 7.420000000000001e-06, | |
| "loss": 0.3173, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 7.92e-06, | |
| "loss": 0.3202, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 8.42e-06, | |
| "loss": 0.3133, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 8.920000000000001e-06, | |
| "loss": 0.3045, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 9.42e-06, | |
| "loss": 0.2761, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 9.920000000000002e-06, | |
| "loss": 0.3105, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 9.940000000000001e-06, | |
| "loss": 0.2969, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 9.86857142857143e-06, | |
| "loss": 0.2923, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 9.797142857142858e-06, | |
| "loss": 0.3035, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 9.725714285714287e-06, | |
| "loss": 0.3098, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 9.654285714285716e-06, | |
| "loss": 0.2933, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 9.582857142857143e-06, | |
| "loss": 0.3066, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 9.511428571428572e-06, | |
| "loss": 0.2754, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 9.440000000000001e-06, | |
| "loss": 0.2881, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 9.368571428571428e-06, | |
| "loss": 0.2919, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 9.297142857142857e-06, | |
| "loss": 0.2861, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 9.225714285714286e-06, | |
| "loss": 0.2804, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 9.154285714285715e-06, | |
| "loss": 0.1391, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 9.082857142857143e-06, | |
| "loss": 0.1481, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 9.011428571428572e-06, | |
| "loss": 0.1495, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 8.94e-06, | |
| "loss": 0.1301, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 8.86857142857143e-06, | |
| "loss": 0.1508, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 8.797142857142857e-06, | |
| "loss": 0.1577, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 8.725714285714286e-06, | |
| "loss": 0.1522, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 8.654285714285715e-06, | |
| "loss": 0.1499, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 8.582857142857144e-06, | |
| "loss": 0.1379, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "eval_loss": 0.2958109676837921, | |
| "eval_runtime": 1998.4718, | |
| "eval_samples_per_second": 2.536, | |
| "eval_steps_per_second": 0.317, | |
| "eval_wer": 21.467559355758116, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 9.052727272727273e-06, | |
| "loss": 0.156, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 9.007272727272729e-06, | |
| "loss": 0.1415, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 8.961818181818182e-06, | |
| "loss": 0.1409, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 8.916363636363638e-06, | |
| "loss": 0.1515, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 8.870909090909091e-06, | |
| "loss": 0.1426, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 8.825454545454545e-06, | |
| "loss": 0.1455, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 8.78e-06, | |
| "loss": 0.1405, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 8.734545454545456e-06, | |
| "loss": 0.1561, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 8.68909090909091e-06, | |
| "loss": 0.1412, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 8.643636363636364e-06, | |
| "loss": 0.1349, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 8.598181818181819e-06, | |
| "loss": 0.1449, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 8.552727272727274e-06, | |
| "loss": 0.1322, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 8.507272727272728e-06, | |
| "loss": 0.1461, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 8.461818181818182e-06, | |
| "loss": 0.1297, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 8.416363636363637e-06, | |
| "loss": 0.1562, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 8.370909090909091e-06, | |
| "loss": 0.1411, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 8.325454545454547e-06, | |
| "loss": 0.1323, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 8.28e-06, | |
| "loss": 0.1278, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 8.234545454545456e-06, | |
| "loss": 0.1514, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 8.18909090909091e-06, | |
| "loss": 0.1357, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 8.143636363636365e-06, | |
| "loss": 0.1472, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 8.098181818181819e-06, | |
| "loss": 0.1146, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 8.052727272727272e-06, | |
| "loss": 0.0513, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 8.007272727272728e-06, | |
| "loss": 0.0525, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 7.961818181818183e-06, | |
| "loss": 0.0571, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 7.916363636363637e-06, | |
| "loss": 0.0575, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 7.870909090909091e-06, | |
| "loss": 0.0579, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 7.825454545454546e-06, | |
| "loss": 0.0599, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 7.78e-06, | |
| "loss": 0.052, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 7.734545454545455e-06, | |
| "loss": 0.0522, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 7.68909090909091e-06, | |
| "loss": 0.0553, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 7.643636363636365e-06, | |
| "loss": 0.0494, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 7.598181818181819e-06, | |
| "loss": 0.0559, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 7.552727272727274e-06, | |
| "loss": 0.054, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "learning_rate": 7.507272727272728e-06, | |
| "loss": 0.0554, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 7.461818181818182e-06, | |
| "loss": 0.0524, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 7.416363636363637e-06, | |
| "loss": 0.0548, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 7.370909090909092e-06, | |
| "loss": 0.0482, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 7.325454545454546e-06, | |
| "loss": 0.0515, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 7.280000000000001e-06, | |
| "loss": 0.0504, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "eval_loss": 0.29510194063186646, | |
| "eval_runtime": 2075.6161, | |
| "eval_samples_per_second": 2.442, | |
| "eval_steps_per_second": 0.305, | |
| "eval_wer": 20.513027345325483, | |
| "step": 2000 | |
| } | |
| ], | |
| "max_steps": 6000, | |
| "num_train_epochs": 8, | |
| "total_flos": 9.23011544383488e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |