| { | |
| "best_metric": 41.11150694691842, | |
| "best_model_checkpoint": "whisper_large_v2_adam_8bit/checkpoint-485", | |
| "epoch": 17.962962962962962, | |
| "global_step": 485, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 8.000000000000001e-07, | |
| "loss": 1.0874, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_loss": 1.0718352794647217, | |
| "eval_runtime": 305.4443, | |
| "eval_samples_per_second": 0.262, | |
| "eval_steps_per_second": 0.131, | |
| "eval_wer": 54.82721767011044, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 1.8000000000000001e-06, | |
| "loss": 1.0169, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_loss": 0.895382285118103, | |
| "eval_runtime": 335.7721, | |
| "eval_samples_per_second": 0.238, | |
| "eval_steps_per_second": 0.119, | |
| "eval_wer": 57.14285714285714, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 2.6e-06, | |
| "loss": 0.8095, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "eval_loss": 0.7291234135627747, | |
| "eval_runtime": 341.1764, | |
| "eval_samples_per_second": 0.234, | |
| "eval_steps_per_second": 0.117, | |
| "eval_wer": 59.707873174207336, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 3.6000000000000003e-06, | |
| "loss": 0.6586, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "eval_loss": 0.5784670114517212, | |
| "eval_runtime": 208.1203, | |
| "eval_samples_per_second": 0.384, | |
| "eval_steps_per_second": 0.192, | |
| "eval_wer": 83.07801923762024, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 4.600000000000001e-06, | |
| "loss": 0.5849, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "eval_loss": 0.5224391222000122, | |
| "eval_runtime": 251.382, | |
| "eval_samples_per_second": 0.318, | |
| "eval_steps_per_second": 0.159, | |
| "eval_wer": 91.73494834342715, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 5.600000000000001e-06, | |
| "loss": 0.4623, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "eval_loss": 0.48158684372901917, | |
| "eval_runtime": 279.2357, | |
| "eval_samples_per_second": 0.286, | |
| "eval_steps_per_second": 0.143, | |
| "eval_wer": 72.99608122550765, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 6.600000000000001e-06, | |
| "loss": 0.4459, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "eval_loss": 0.4286937713623047, | |
| "eval_runtime": 232.3372, | |
| "eval_samples_per_second": 0.344, | |
| "eval_steps_per_second": 0.172, | |
| "eval_wer": 76.0242251514072, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 7.600000000000001e-06, | |
| "loss": 0.3496, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "eval_loss": 0.4074183404445648, | |
| "eval_runtime": 289.554, | |
| "eval_samples_per_second": 0.276, | |
| "eval_steps_per_second": 0.138, | |
| "eval_wer": 48.6284289276808, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 8.6e-06, | |
| "loss": 0.3964, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "eval_loss": 0.3671049177646637, | |
| "eval_runtime": 356.7998, | |
| "eval_samples_per_second": 0.224, | |
| "eval_steps_per_second": 0.112, | |
| "eval_wer": 57.67723548272177, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 9.600000000000001e-06, | |
| "loss": 0.3506, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "eval_loss": 0.37803885340690613, | |
| "eval_runtime": 338.1245, | |
| "eval_samples_per_second": 0.237, | |
| "eval_steps_per_second": 0.118, | |
| "eval_wer": 49.76843605272533, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 9.96842105263158e-06, | |
| "loss": 0.377, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "eval_loss": 0.35833048820495605, | |
| "eval_runtime": 344.9289, | |
| "eval_samples_per_second": 0.232, | |
| "eval_steps_per_second": 0.116, | |
| "eval_wer": 56.21660135375846, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 9.915789473684211e-06, | |
| "loss": 0.2766, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "eval_loss": 0.37928158044815063, | |
| "eval_runtime": 350.8056, | |
| "eval_samples_per_second": 0.228, | |
| "eval_steps_per_second": 0.114, | |
| "eval_wer": 83.0423940149626, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 9.863157894736843e-06, | |
| "loss": 0.2522, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "eval_loss": 0.3648552894592285, | |
| "eval_runtime": 343.605, | |
| "eval_samples_per_second": 0.233, | |
| "eval_steps_per_second": 0.116, | |
| "eval_wer": 69.89668685429284, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 9.810526315789475e-06, | |
| "loss": 0.2793, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "eval_loss": 0.36223381757736206, | |
| "eval_runtime": 325.8454, | |
| "eval_samples_per_second": 0.246, | |
| "eval_steps_per_second": 0.123, | |
| "eval_wer": 48.37905236907731, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 9.757894736842106e-06, | |
| "loss": 0.27, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "eval_loss": 0.38781291246414185, | |
| "eval_runtime": 331.2847, | |
| "eval_samples_per_second": 0.241, | |
| "eval_steps_per_second": 0.121, | |
| "eval_wer": 48.09405058781617, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 9.715789473684212e-06, | |
| "loss": 0.2624, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "eval_loss": 0.38262635469436646, | |
| "eval_runtime": 329.4177, | |
| "eval_samples_per_second": 0.243, | |
| "eval_steps_per_second": 0.121, | |
| "eval_wer": 47.167794798717495, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "learning_rate": 9.663157894736843e-06, | |
| "loss": 0.225, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "eval_loss": 0.38125258684158325, | |
| "eval_runtime": 310.0949, | |
| "eval_samples_per_second": 0.258, | |
| "eval_steps_per_second": 0.129, | |
| "eval_wer": 54.39971499821874, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 9.610526315789475e-06, | |
| "loss": 0.2062, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "eval_loss": 0.3962409198284149, | |
| "eval_runtime": 299.3235, | |
| "eval_samples_per_second": 0.267, | |
| "eval_steps_per_second": 0.134, | |
| "eval_wer": 72.92483078019238, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "learning_rate": 9.557894736842107e-06, | |
| "loss": 0.192, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "eval_loss": 0.3863430619239807, | |
| "eval_runtime": 332.6447, | |
| "eval_samples_per_second": 0.24, | |
| "eval_steps_per_second": 0.12, | |
| "eval_wer": 59.17349483434271, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 9.505263157894738e-06, | |
| "loss": 0.224, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "eval_loss": 0.38708847761154175, | |
| "eval_runtime": 330.6389, | |
| "eval_samples_per_second": 0.242, | |
| "eval_steps_per_second": 0.121, | |
| "eval_wer": 54.39971499821874, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 9.452631578947368e-06, | |
| "loss": 0.2127, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "eval_loss": 0.41648903489112854, | |
| "eval_runtime": 357.2142, | |
| "eval_samples_per_second": 0.224, | |
| "eval_steps_per_second": 0.112, | |
| "eval_wer": 70.07481296758104, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 9.4e-06, | |
| "loss": 0.2033, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "eval_loss": 0.39047971367836, | |
| "eval_runtime": 316.4665, | |
| "eval_samples_per_second": 0.253, | |
| "eval_steps_per_second": 0.126, | |
| "eval_wer": 50.69469184182401, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "learning_rate": 9.347368421052633e-06, | |
| "loss": 0.1426, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "eval_loss": 0.41405850648880005, | |
| "eval_runtime": 326.6943, | |
| "eval_samples_per_second": 0.245, | |
| "eval_steps_per_second": 0.122, | |
| "eval_wer": 44.53152832205201, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 9.294736842105265e-06, | |
| "loss": 0.1737, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "eval_loss": 0.4162759780883789, | |
| "eval_runtime": 334.7907, | |
| "eval_samples_per_second": 0.239, | |
| "eval_steps_per_second": 0.119, | |
| "eval_wer": 47.52404702529391, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.63, | |
| "learning_rate": 9.242105263157896e-06, | |
| "loss": 0.1404, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 4.63, | |
| "eval_loss": 0.4179295599460602, | |
| "eval_runtime": 313.7669, | |
| "eval_samples_per_second": 0.255, | |
| "eval_steps_per_second": 0.127, | |
| "eval_wer": 47.951549697185605, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 4.81, | |
| "learning_rate": 9.189473684210526e-06, | |
| "loss": 0.1738, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 4.81, | |
| "eval_loss": 0.4070858061313629, | |
| "eval_runtime": 322.5875, | |
| "eval_samples_per_second": 0.248, | |
| "eval_steps_per_second": 0.124, | |
| "eval_wer": 43.747773423583894, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 9.136842105263158e-06, | |
| "loss": 0.1852, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.3889687657356262, | |
| "eval_runtime": 331.8197, | |
| "eval_samples_per_second": 0.241, | |
| "eval_steps_per_second": 0.121, | |
| "eval_wer": 65.94228713929462, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 5.19, | |
| "learning_rate": 9.08421052631579e-06, | |
| "loss": 0.1111, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 5.19, | |
| "eval_loss": 0.4467521607875824, | |
| "eval_runtime": 340.6568, | |
| "eval_samples_per_second": 0.235, | |
| "eval_steps_per_second": 0.117, | |
| "eval_wer": 47.5952974706092, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 5.37, | |
| "learning_rate": 9.031578947368423e-06, | |
| "loss": 0.1642, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 5.37, | |
| "eval_loss": 0.47300252318382263, | |
| "eval_runtime": 330.5579, | |
| "eval_samples_per_second": 0.242, | |
| "eval_steps_per_second": 0.121, | |
| "eval_wer": 53.152832205201285, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 5.56, | |
| "learning_rate": 8.978947368421055e-06, | |
| "loss": 0.1552, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 5.56, | |
| "eval_loss": 0.44687384366989136, | |
| "eval_runtime": 317.2285, | |
| "eval_samples_per_second": 0.252, | |
| "eval_steps_per_second": 0.126, | |
| "eval_wer": 65.62166013537585, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 5.74, | |
| "learning_rate": 8.926315789473685e-06, | |
| "loss": 0.1497, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 5.74, | |
| "eval_loss": 0.44148778915405273, | |
| "eval_runtime": 315.1611, | |
| "eval_samples_per_second": 0.254, | |
| "eval_steps_per_second": 0.127, | |
| "eval_wer": 48.05842536515853, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 5.93, | |
| "learning_rate": 8.873684210526316e-06, | |
| "loss": 0.1419, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 5.93, | |
| "eval_loss": 0.4392489492893219, | |
| "eval_runtime": 336.253, | |
| "eval_samples_per_second": 0.238, | |
| "eval_steps_per_second": 0.119, | |
| "eval_wer": 43.9258995368721, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 6.11, | |
| "learning_rate": 8.821052631578948e-06, | |
| "loss": 0.1176, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 6.11, | |
| "eval_loss": 0.44891494512557983, | |
| "eval_runtime": 328.7909, | |
| "eval_samples_per_second": 0.243, | |
| "eval_steps_per_second": 0.122, | |
| "eval_wer": 41.1827573922337, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 6.3, | |
| "learning_rate": 8.76842105263158e-06, | |
| "loss": 0.0996, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 6.3, | |
| "eval_loss": 0.505344569683075, | |
| "eval_runtime": 338.587, | |
| "eval_samples_per_second": 0.236, | |
| "eval_steps_per_second": 0.118, | |
| "eval_wer": 46.81154257214107, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 6.48, | |
| "learning_rate": 8.715789473684211e-06, | |
| "loss": 0.1376, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 6.48, | |
| "eval_loss": 0.48324888944625854, | |
| "eval_runtime": 326.4874, | |
| "eval_samples_per_second": 0.245, | |
| "eval_steps_per_second": 0.123, | |
| "eval_wer": 52.29782686141788, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 8.663157894736843e-06, | |
| "loss": 0.1093, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "eval_loss": 0.4419935643672943, | |
| "eval_runtime": 340.888, | |
| "eval_samples_per_second": 0.235, | |
| "eval_steps_per_second": 0.117, | |
| "eval_wer": 54.39971499821874, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 6.85, | |
| "learning_rate": 8.610526315789474e-06, | |
| "loss": 0.1325, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 6.85, | |
| "eval_loss": 0.4463180899620056, | |
| "eval_runtime": 333.0453, | |
| "eval_samples_per_second": 0.24, | |
| "eval_steps_per_second": 0.12, | |
| "eval_wer": 45.92091200570004, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 8.557894736842106e-06, | |
| "loss": 0.1151, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "eval_loss": 0.4659479558467865, | |
| "eval_runtime": 321.4982, | |
| "eval_samples_per_second": 0.249, | |
| "eval_steps_per_second": 0.124, | |
| "eval_wer": 46.09903811898825, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 7.22, | |
| "learning_rate": 8.505263157894738e-06, | |
| "loss": 0.0892, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 7.22, | |
| "eval_loss": 0.4642786979675293, | |
| "eval_runtime": 304.1157, | |
| "eval_samples_per_second": 0.263, | |
| "eval_steps_per_second": 0.132, | |
| "eval_wer": 61.9878874242964, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 7.41, | |
| "learning_rate": 8.45263157894737e-06, | |
| "loss": 0.0819, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 7.41, | |
| "eval_loss": 0.4606548845767975, | |
| "eval_runtime": 314.9466, | |
| "eval_samples_per_second": 0.254, | |
| "eval_steps_per_second": 0.127, | |
| "eval_wer": 49.162807267545425, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 7.59, | |
| "learning_rate": 8.400000000000001e-06, | |
| "loss": 0.0881, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 7.59, | |
| "eval_loss": 0.49518561363220215, | |
| "eval_runtime": 333.9165, | |
| "eval_samples_per_second": 0.24, | |
| "eval_steps_per_second": 0.12, | |
| "eval_wer": 47.167794798717495, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 7.78, | |
| "learning_rate": 8.347368421052633e-06, | |
| "loss": 0.0902, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 7.78, | |
| "eval_loss": 0.48233914375305176, | |
| "eval_runtime": 323.2853, | |
| "eval_samples_per_second": 0.247, | |
| "eval_steps_per_second": 0.124, | |
| "eval_wer": 47.98717491984325, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 7.96, | |
| "learning_rate": 8.294736842105264e-06, | |
| "loss": 0.1125, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 7.96, | |
| "eval_loss": 0.48526009917259216, | |
| "eval_runtime": 331.099, | |
| "eval_samples_per_second": 0.242, | |
| "eval_steps_per_second": 0.121, | |
| "eval_wer": 59.45849661560385, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 8.15, | |
| "learning_rate": 8.242105263157896e-06, | |
| "loss": 0.0668, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 8.15, | |
| "eval_loss": 0.5275024175643921, | |
| "eval_runtime": 310.2348, | |
| "eval_samples_per_second": 0.258, | |
| "eval_steps_per_second": 0.129, | |
| "eval_wer": 58.42536515853224, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "learning_rate": 8.189473684210527e-06, | |
| "loss": 0.0723, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "eval_loss": 0.532504141330719, | |
| "eval_runtime": 331.9501, | |
| "eval_samples_per_second": 0.241, | |
| "eval_steps_per_second": 0.121, | |
| "eval_wer": 46.775917349483436, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 8.52, | |
| "learning_rate": 8.136842105263159e-06, | |
| "loss": 0.0825, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 8.52, | |
| "eval_loss": 0.5000892877578735, | |
| "eval_runtime": 328.9744, | |
| "eval_samples_per_second": 0.243, | |
| "eval_steps_per_second": 0.122, | |
| "eval_wer": 44.21090131813324, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 8.7, | |
| "learning_rate": 8.08421052631579e-06, | |
| "loss": 0.0859, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 8.7, | |
| "eval_loss": 0.5018569827079773, | |
| "eval_runtime": 321.0271, | |
| "eval_samples_per_second": 0.249, | |
| "eval_steps_per_second": 0.125, | |
| "eval_wer": 55.68222301389384, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 8.89, | |
| "learning_rate": 8.03157894736842e-06, | |
| "loss": 0.0838, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 8.89, | |
| "eval_loss": 0.5195188522338867, | |
| "eval_runtime": 326.5564, | |
| "eval_samples_per_second": 0.245, | |
| "eval_steps_per_second": 0.122, | |
| "eval_wer": 50.65906661916637, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 9.07, | |
| "learning_rate": 7.978947368421052e-06, | |
| "loss": 0.0634, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 9.07, | |
| "eval_loss": 0.5121615529060364, | |
| "eval_runtime": 333.3682, | |
| "eval_samples_per_second": 0.24, | |
| "eval_steps_per_second": 0.12, | |
| "eval_wer": 45.42215888849305, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 9.26, | |
| "learning_rate": 7.926315789473686e-06, | |
| "loss": 0.0611, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 9.26, | |
| "eval_loss": 0.53472900390625, | |
| "eval_runtime": 334.7977, | |
| "eval_samples_per_second": 0.239, | |
| "eval_steps_per_second": 0.119, | |
| "eval_wer": 45.38653366583541, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 9.44, | |
| "learning_rate": 7.873684210526317e-06, | |
| "loss": 0.0742, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 9.44, | |
| "eval_loss": 0.5380967855453491, | |
| "eval_runtime": 341.6365, | |
| "eval_samples_per_second": 0.234, | |
| "eval_steps_per_second": 0.117, | |
| "eval_wer": 46.170288564303526, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 9.63, | |
| "learning_rate": 7.821052631578949e-06, | |
| "loss": 0.0544, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 9.63, | |
| "eval_loss": 0.5325397253036499, | |
| "eval_runtime": 344.44, | |
| "eval_samples_per_second": 0.232, | |
| "eval_steps_per_second": 0.116, | |
| "eval_wer": 52.08407552547203, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 9.81, | |
| "learning_rate": 7.768421052631579e-06, | |
| "loss": 0.0832, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 9.81, | |
| "eval_loss": 0.5298484563827515, | |
| "eval_runtime": 338.7288, | |
| "eval_samples_per_second": 0.236, | |
| "eval_steps_per_second": 0.118, | |
| "eval_wer": 46.81154257214107, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 7.71578947368421e-06, | |
| "loss": 0.079, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.5149779319763184, | |
| "eval_runtime": 322.201, | |
| "eval_samples_per_second": 0.248, | |
| "eval_steps_per_second": 0.124, | |
| "eval_wer": 49.447809048806555, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 10.19, | |
| "learning_rate": 7.663157894736842e-06, | |
| "loss": 0.0692, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 10.19, | |
| "eval_loss": 0.5358998775482178, | |
| "eval_runtime": 304.2081, | |
| "eval_samples_per_second": 0.263, | |
| "eval_steps_per_second": 0.131, | |
| "eval_wer": 50.3384396152476, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 10.37, | |
| "learning_rate": 7.610526315789474e-06, | |
| "loss": 0.0549, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 10.37, | |
| "eval_loss": 0.5207683444023132, | |
| "eval_runtime": 333.629, | |
| "eval_samples_per_second": 0.24, | |
| "eval_steps_per_second": 0.12, | |
| "eval_wer": 49.12718204488778, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 10.56, | |
| "learning_rate": 7.557894736842106e-06, | |
| "loss": 0.0533, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 10.56, | |
| "eval_loss": 0.528590202331543, | |
| "eval_runtime": 332.7917, | |
| "eval_samples_per_second": 0.24, | |
| "eval_steps_per_second": 0.12, | |
| "eval_wer": 47.02529390808692, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 10.74, | |
| "learning_rate": 7.505263157894738e-06, | |
| "loss": 0.0821, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 10.74, | |
| "eval_loss": 0.51960289478302, | |
| "eval_runtime": 332.8933, | |
| "eval_samples_per_second": 0.24, | |
| "eval_steps_per_second": 0.12, | |
| "eval_wer": 46.24153900961881, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 10.93, | |
| "learning_rate": 7.4526315789473695e-06, | |
| "loss": 0.0672, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 10.93, | |
| "eval_loss": 0.5139871835708618, | |
| "eval_runtime": 342.8201, | |
| "eval_samples_per_second": 0.233, | |
| "eval_steps_per_second": 0.117, | |
| "eval_wer": 46.45529034556466, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 11.11, | |
| "learning_rate": 7.4e-06, | |
| "loss": 0.0694, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 11.11, | |
| "eval_loss": 0.5785766243934631, | |
| "eval_runtime": 334.4012, | |
| "eval_samples_per_second": 0.239, | |
| "eval_steps_per_second": 0.12, | |
| "eval_wer": 44.92340577128607, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 11.3, | |
| "learning_rate": 7.347368421052632e-06, | |
| "loss": 0.09, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 11.3, | |
| "eval_loss": 0.584839940071106, | |
| "eval_runtime": 343.2341, | |
| "eval_samples_per_second": 0.233, | |
| "eval_steps_per_second": 0.117, | |
| "eval_wer": 49.59030993943712, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 11.48, | |
| "learning_rate": 7.2947368421052636e-06, | |
| "loss": 0.0824, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 11.48, | |
| "eval_loss": 0.5461788773536682, | |
| "eval_runtime": 330.7912, | |
| "eval_samples_per_second": 0.242, | |
| "eval_steps_per_second": 0.121, | |
| "eval_wer": 45.24403277520484, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 11.67, | |
| "learning_rate": 7.242105263157896e-06, | |
| "loss": 0.0805, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 11.67, | |
| "eval_loss": 0.537391185760498, | |
| "eval_runtime": 352.6332, | |
| "eval_samples_per_second": 0.227, | |
| "eval_steps_per_second": 0.113, | |
| "eval_wer": 54.25721410758817, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 11.85, | |
| "learning_rate": 7.189473684210527e-06, | |
| "loss": 0.0688, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 11.85, | |
| "eval_loss": 0.5511082410812378, | |
| "eval_runtime": 336.5436, | |
| "eval_samples_per_second": 0.238, | |
| "eval_steps_per_second": 0.119, | |
| "eval_wer": 42.8927680798005, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 12.04, | |
| "learning_rate": 7.1368421052631585e-06, | |
| "loss": 0.048, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 12.04, | |
| "eval_loss": 0.5636075139045715, | |
| "eval_runtime": 319.3896, | |
| "eval_samples_per_second": 0.25, | |
| "eval_steps_per_second": 0.125, | |
| "eval_wer": 44.81653010331315, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 12.22, | |
| "learning_rate": 7.08421052631579e-06, | |
| "loss": 0.054, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 12.22, | |
| "eval_loss": 0.5698742866516113, | |
| "eval_runtime": 345.1573, | |
| "eval_samples_per_second": 0.232, | |
| "eval_steps_per_second": 0.116, | |
| "eval_wer": 46.88279301745636, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 12.41, | |
| "learning_rate": 7.031578947368422e-06, | |
| "loss": 0.0535, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 12.41, | |
| "eval_loss": 0.5786362886428833, | |
| "eval_runtime": 331.7186, | |
| "eval_samples_per_second": 0.241, | |
| "eval_steps_per_second": 0.121, | |
| "eval_wer": 42.60776629853937, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 12.59, | |
| "learning_rate": 6.9789473684210525e-06, | |
| "loss": 0.0644, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 12.59, | |
| "eval_loss": 0.5574811697006226, | |
| "eval_runtime": 339.8046, | |
| "eval_samples_per_second": 0.235, | |
| "eval_steps_per_second": 0.118, | |
| "eval_wer": 45.956537228357675, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 12.78, | |
| "learning_rate": 6.926315789473684e-06, | |
| "loss": 0.0618, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 12.78, | |
| "eval_loss": 0.5477833151817322, | |
| "eval_runtime": 337.626, | |
| "eval_samples_per_second": 0.237, | |
| "eval_steps_per_second": 0.118, | |
| "eval_wer": 46.918418240114, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 12.96, | |
| "learning_rate": 6.873684210526317e-06, | |
| "loss": 0.0832, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 12.96, | |
| "eval_loss": 0.6038811802864075, | |
| "eval_runtime": 336.2086, | |
| "eval_samples_per_second": 0.238, | |
| "eval_steps_per_second": 0.119, | |
| "eval_wer": 44.068400427502674, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 13.15, | |
| "learning_rate": 6.821052631578948e-06, | |
| "loss": 0.0592, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 13.15, | |
| "eval_loss": 0.6127667427062988, | |
| "eval_runtime": 331.5309, | |
| "eval_samples_per_second": 0.241, | |
| "eval_steps_per_second": 0.121, | |
| "eval_wer": 42.32276451727823, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 13.33, | |
| "learning_rate": 6.76842105263158e-06, | |
| "loss": 0.0619, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 13.33, | |
| "eval_loss": 0.591279149055481, | |
| "eval_runtime": 332.3235, | |
| "eval_samples_per_second": 0.241, | |
| "eval_steps_per_second": 0.12, | |
| "eval_wer": 45.03028143925899, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 13.52, | |
| "learning_rate": 6.715789473684211e-06, | |
| "loss": 0.0998, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 13.52, | |
| "eval_loss": 0.5318233966827393, | |
| "eval_runtime": 332.2782, | |
| "eval_samples_per_second": 0.241, | |
| "eval_steps_per_second": 0.12, | |
| "eval_wer": 42.75026718916993, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 13.7, | |
| "learning_rate": 6.663157894736842e-06, | |
| "loss": 0.05, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 13.7, | |
| "eval_loss": 0.5774287581443787, | |
| "eval_runtime": 329.9705, | |
| "eval_samples_per_second": 0.242, | |
| "eval_steps_per_second": 0.121, | |
| "eval_wer": 42.82151763448522, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 13.89, | |
| "learning_rate": 6.610526315789474e-06, | |
| "loss": 0.1014, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 13.89, | |
| "eval_loss": 0.5849052667617798, | |
| "eval_runtime": 328.3008, | |
| "eval_samples_per_second": 0.244, | |
| "eval_steps_per_second": 0.122, | |
| "eval_wer": 48.05842536515853, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 14.07, | |
| "learning_rate": 6.557894736842106e-06, | |
| "loss": 0.0494, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 14.07, | |
| "eval_loss": 0.5518860816955566, | |
| "eval_runtime": 318.5974, | |
| "eval_samples_per_second": 0.251, | |
| "eval_steps_per_second": 0.126, | |
| "eval_wer": 48.69967937299608, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 14.26, | |
| "learning_rate": 6.505263157894738e-06, | |
| "loss": 0.0456, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 14.26, | |
| "eval_loss": 0.5957409739494324, | |
| "eval_runtime": 335.9574, | |
| "eval_samples_per_second": 0.238, | |
| "eval_steps_per_second": 0.119, | |
| "eval_wer": 42.75026718916993, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 14.44, | |
| "learning_rate": 6.452631578947369e-06, | |
| "loss": 0.0957, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 14.44, | |
| "eval_loss": 0.5850934982299805, | |
| "eval_runtime": 347.6573, | |
| "eval_samples_per_second": 0.23, | |
| "eval_steps_per_second": 0.115, | |
| "eval_wer": 48.80655504096901, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 14.63, | |
| "learning_rate": 6.4000000000000006e-06, | |
| "loss": 0.0514, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 14.63, | |
| "eval_loss": 0.6157152652740479, | |
| "eval_runtime": 332.3513, | |
| "eval_samples_per_second": 0.241, | |
| "eval_steps_per_second": 0.12, | |
| "eval_wer": 49.661560384752406, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 14.81, | |
| "learning_rate": 6.347368421052632e-06, | |
| "loss": 0.1045, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 14.81, | |
| "eval_loss": 0.5892783999443054, | |
| "eval_runtime": 341.0177, | |
| "eval_samples_per_second": 0.235, | |
| "eval_steps_per_second": 0.117, | |
| "eval_wer": 52.6540790879943, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 6.294736842105264e-06, | |
| "loss": 0.0485, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 0.5818936228752136, | |
| "eval_runtime": 334.6692, | |
| "eval_samples_per_second": 0.239, | |
| "eval_steps_per_second": 0.12, | |
| "eval_wer": 50.9440684004275, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 15.19, | |
| "learning_rate": 6.242105263157895e-06, | |
| "loss": 0.0705, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 15.19, | |
| "eval_loss": 0.5869933366775513, | |
| "eval_runtime": 338.9314, | |
| "eval_samples_per_second": 0.236, | |
| "eval_steps_per_second": 0.118, | |
| "eval_wer": 45.77841111506947, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 15.37, | |
| "learning_rate": 6.189473684210526e-06, | |
| "loss": 0.0354, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 15.37, | |
| "eval_loss": 0.620836615562439, | |
| "eval_runtime": 340.9346, | |
| "eval_samples_per_second": 0.235, | |
| "eval_steps_per_second": 0.117, | |
| "eval_wer": 47.20342002137513, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 15.56, | |
| "learning_rate": 6.136842105263159e-06, | |
| "loss": 0.0366, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 15.56, | |
| "eval_loss": 0.6018834114074707, | |
| "eval_runtime": 339.0784, | |
| "eval_samples_per_second": 0.236, | |
| "eval_steps_per_second": 0.118, | |
| "eval_wer": 46.02778767367296, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 15.74, | |
| "learning_rate": 6.08421052631579e-06, | |
| "loss": 0.0422, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 15.74, | |
| "eval_loss": 0.5831278562545776, | |
| "eval_runtime": 332.3516, | |
| "eval_samples_per_second": 0.241, | |
| "eval_steps_per_second": 0.12, | |
| "eval_wer": 44.92340577128607, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 15.93, | |
| "learning_rate": 6.031578947368422e-06, | |
| "loss": 0.1133, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 15.93, | |
| "eval_loss": 0.6038104295730591, | |
| "eval_runtime": 333.9842, | |
| "eval_samples_per_second": 0.24, | |
| "eval_steps_per_second": 0.12, | |
| "eval_wer": 44.13965087281795, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 16.11, | |
| "learning_rate": 5.978947368421053e-06, | |
| "loss": 0.0515, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 16.11, | |
| "eval_loss": 0.6272233724594116, | |
| "eval_runtime": 331.7109, | |
| "eval_samples_per_second": 0.241, | |
| "eval_steps_per_second": 0.121, | |
| "eval_wer": 43.53402208763805, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 16.3, | |
| "learning_rate": 5.9263157894736844e-06, | |
| "loss": 0.0709, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 16.3, | |
| "eval_loss": 0.6419104337692261, | |
| "eval_runtime": 336.7239, | |
| "eval_samples_per_second": 0.238, | |
| "eval_steps_per_second": 0.119, | |
| "eval_wer": 45.92091200570004, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 16.48, | |
| "learning_rate": 5.873684210526316e-06, | |
| "loss": 0.0978, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 16.48, | |
| "eval_loss": 0.6428846120834351, | |
| "eval_runtime": 334.722, | |
| "eval_samples_per_second": 0.239, | |
| "eval_steps_per_second": 0.12, | |
| "eval_wer": 46.59779123619522, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 16.67, | |
| "learning_rate": 5.8210526315789486e-06, | |
| "loss": 0.052, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 16.67, | |
| "eval_loss": 0.6158877611160278, | |
| "eval_runtime": 340.1528, | |
| "eval_samples_per_second": 0.235, | |
| "eval_steps_per_second": 0.118, | |
| "eval_wer": 45.2084075525472, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 16.85, | |
| "learning_rate": 5.76842105263158e-06, | |
| "loss": 0.0508, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 16.85, | |
| "eval_loss": 0.631538987159729, | |
| "eval_runtime": 334.4872, | |
| "eval_samples_per_second": 0.239, | |
| "eval_steps_per_second": 0.12, | |
| "eval_wer": 45.24403277520484, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 17.04, | |
| "learning_rate": 5.715789473684211e-06, | |
| "loss": 0.0453, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 17.04, | |
| "eval_loss": 0.6536934971809387, | |
| "eval_runtime": 327.9543, | |
| "eval_samples_per_second": 0.244, | |
| "eval_steps_per_second": 0.122, | |
| "eval_wer": 48.87780548628429, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 17.22, | |
| "learning_rate": 5.663157894736843e-06, | |
| "loss": 0.0454, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 17.22, | |
| "eval_loss": 0.6028529405593872, | |
| "eval_runtime": 337.2385, | |
| "eval_samples_per_second": 0.237, | |
| "eval_steps_per_second": 0.119, | |
| "eval_wer": 48.770929818311366, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 17.41, | |
| "learning_rate": 5.610526315789474e-06, | |
| "loss": 0.0662, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 17.41, | |
| "eval_loss": 0.5731960535049438, | |
| "eval_runtime": 331.8614, | |
| "eval_samples_per_second": 0.241, | |
| "eval_steps_per_second": 0.121, | |
| "eval_wer": 45.885286783042396, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 17.59, | |
| "learning_rate": 5.557894736842105e-06, | |
| "loss": 0.0365, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 17.59, | |
| "eval_loss": 0.6006522178649902, | |
| "eval_runtime": 320.1815, | |
| "eval_samples_per_second": 0.25, | |
| "eval_steps_per_second": 0.125, | |
| "eval_wer": 43.89027431421446, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 17.78, | |
| "learning_rate": 5.505263157894737e-06, | |
| "loss": 0.0728, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 17.78, | |
| "eval_loss": 0.6429422497749329, | |
| "eval_runtime": 333.1015, | |
| "eval_samples_per_second": 0.24, | |
| "eval_steps_per_second": 0.12, | |
| "eval_wer": 44.67402921268258, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 17.96, | |
| "learning_rate": 5.452631578947369e-06, | |
| "loss": 0.0334, | |
| "step": 485 | |
| }, | |
| { | |
| "epoch": 17.96, | |
| "eval_loss": 0.648144543170929, | |
| "eval_runtime": 320.2927, | |
| "eval_samples_per_second": 0.25, | |
| "eval_steps_per_second": 0.125, | |
| "eval_wer": 41.11150694691842, | |
| "step": 485 | |
| } | |
| ], | |
| "max_steps": 1000, | |
| "num_train_epochs": 38, | |
| "total_flos": 1.6475807121408e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |