| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 19.2, | |
| "eval_steps": 720, | |
| "global_step": 14400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0013333333333333333, | |
| "grad_norm": 16.625, | |
| "learning_rate": 2e-06, | |
| "loss": 2.3594, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.13333333333333333, | |
| "grad_norm": 0.2490234375, | |
| "learning_rate": 0.0002, | |
| "loss": 0.9537, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.26666666666666666, | |
| "grad_norm": 0.263671875, | |
| "learning_rate": 0.0004, | |
| "loss": 0.5927, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.2890625, | |
| "learning_rate": 0.0006, | |
| "loss": 0.5721, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5333333333333333, | |
| "grad_norm": 0.546875, | |
| "learning_rate": 0.0008, | |
| "loss": 0.4882, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "grad_norm": 0.42578125, | |
| "learning_rate": 0.001, | |
| "loss": 0.3812, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.474609375, | |
| "learning_rate": 0.0012, | |
| "loss": 0.353, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.9333333333333333, | |
| "grad_norm": 0.333984375, | |
| "learning_rate": 0.0014, | |
| "loss": 0.3347, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_commonvoice-bn-transcription_loss": 0.6287540793418884, | |
| "eval_commonvoice-bn-transcription_model_preparation_time": 0.0074, | |
| "eval_commonvoice-bn-transcription_runtime": 11.4061, | |
| "eval_commonvoice-bn-transcription_samples_per_second": 5.611, | |
| "eval_commonvoice-bn-transcription_steps_per_second": 0.701, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.0666666666666667, | |
| "grad_norm": 0.275390625, | |
| "learning_rate": 0.0016, | |
| "loss": 0.2786, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 0.234375, | |
| "learning_rate": 0.0018000000000000002, | |
| "loss": 0.282, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "grad_norm": 0.1884765625, | |
| "learning_rate": 0.002, | |
| "loss": 0.2787, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.4666666666666668, | |
| "grad_norm": 0.279296875, | |
| "learning_rate": 0.0019997526665988343, | |
| "loss": 0.2676, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.19921875, | |
| "learning_rate": 0.00199901080233714, | |
| "loss": 0.2267, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.7333333333333334, | |
| "grad_norm": 0.208984375, | |
| "learning_rate": 0.0019977748149656092, | |
| "loss": 0.2247, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.8666666666666667, | |
| "grad_norm": 0.2021484375, | |
| "learning_rate": 0.0019960453838197083, | |
| "loss": 0.2223, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "eval_commonvoice-bn-transcription_loss": 0.5989859104156494, | |
| "eval_commonvoice-bn-transcription_model_preparation_time": 0.0074, | |
| "eval_commonvoice-bn-transcription_runtime": 13.0069, | |
| "eval_commonvoice-bn-transcription_samples_per_second": 4.92, | |
| "eval_commonvoice-bn-transcription_steps_per_second": 0.615, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.1337890625, | |
| "learning_rate": 0.0019938234594462988, | |
| "loss": 0.1983, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.1333333333333333, | |
| "grad_norm": 0.203125, | |
| "learning_rate": 0.0019911102630811857, | |
| "loss": 0.1747, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.2666666666666666, | |
| "grad_norm": 0.2060546875, | |
| "learning_rate": 0.001987907285977892, | |
| "loss": 0.1814, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.1689453125, | |
| "learning_rate": 0.0019842162885880185, | |
| "loss": 0.181, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.533333333333333, | |
| "grad_norm": 0.1630859375, | |
| "learning_rate": 0.0019800392995936505, | |
| "loss": 0.1567, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.6666666666666665, | |
| "grad_norm": 0.1357421875, | |
| "learning_rate": 0.0019753786147923315, | |
| "loss": 0.1517, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 0.11083984375, | |
| "learning_rate": 0.0019702367958352256, | |
| "loss": 0.1467, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "eval_commonvoice-bn-transcription_loss": 0.5943728685379028, | |
| "eval_commonvoice-bn-transcription_model_preparation_time": 0.0074, | |
| "eval_commonvoice-bn-transcription_runtime": 12.3498, | |
| "eval_commonvoice-bn-transcription_samples_per_second": 5.182, | |
| "eval_commonvoice-bn-transcription_steps_per_second": 0.648, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 2.9333333333333336, | |
| "grad_norm": 0.1845703125, | |
| "learning_rate": 0.001964616668819156, | |
| "loss": 0.1458, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.066666666666667, | |
| "grad_norm": 0.10595703125, | |
| "learning_rate": 0.001958521322733301, | |
| "loss": 0.116, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 0.177734375, | |
| "learning_rate": 0.0019519541077613908, | |
| "loss": 0.1273, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.3333333333333335, | |
| "grad_norm": 0.1435546875, | |
| "learning_rate": 0.0019449186334403478, | |
| "loss": 0.1232, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.466666666666667, | |
| "grad_norm": 0.1201171875, | |
| "learning_rate": 0.001937418766676378, | |
| "loss": 0.1197, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "grad_norm": 0.11572265625, | |
| "learning_rate": 0.0019294586296196034, | |
| "loss": 0.0999, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 3.7333333333333334, | |
| "grad_norm": 0.1943359375, | |
| "learning_rate": 0.0019210425973984072, | |
| "loss": 0.1029, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "eval_commonvoice-bn-transcription_loss": 0.6720843315124512, | |
| "eval_commonvoice-bn-transcription_model_preparation_time": 0.0074, | |
| "eval_commonvoice-bn-transcription_runtime": 11.4916, | |
| "eval_commonvoice-bn-transcription_samples_per_second": 5.569, | |
| "eval_commonvoice-bn-transcription_steps_per_second": 0.696, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 3.8666666666666667, | |
| "grad_norm": 0.30859375, | |
| "learning_rate": 0.0019121752957147297, | |
| "loss": 0.106, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.0634765625, | |
| "learning_rate": 0.0019028615983016476, | |
| "loss": 0.0966, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.133333333333334, | |
| "grad_norm": 0.275390625, | |
| "learning_rate": 0.0018931066242446255, | |
| "loss": 0.0845, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 4.266666666666667, | |
| "grad_norm": 0.09765625, | |
| "learning_rate": 0.0018829157351679113, | |
| "loss": 0.0889, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "grad_norm": 0.2177734375, | |
| "learning_rate": 0.0018722945322876298, | |
| "loss": 0.0915, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 4.533333333333333, | |
| "grad_norm": 0.091796875, | |
| "learning_rate": 0.0018612488533331878, | |
| "loss": 0.0824, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 4.666666666666667, | |
| "grad_norm": 0.07763671875, | |
| "learning_rate": 0.0018497847693386823, | |
| "loss": 0.0767, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.09130859375, | |
| "learning_rate": 0.0018379085813060821, | |
| "loss": 0.0775, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "eval_commonvoice-bn-transcription_loss": 0.6684668064117432, | |
| "eval_commonvoice-bn-transcription_model_preparation_time": 0.0074, | |
| "eval_commonvoice-bn-transcription_runtime": 11.4545, | |
| "eval_commonvoice-bn-transcription_samples_per_second": 5.587, | |
| "eval_commonvoice-bn-transcription_steps_per_second": 0.698, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 4.933333333333334, | |
| "grad_norm": 0.1220703125, | |
| "learning_rate": 0.0018256268167420066, | |
| "loss": 0.0778, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 5.066666666666666, | |
| "grad_norm": 0.060546875, | |
| "learning_rate": 0.0018129462260700162, | |
| "loss": 0.0638, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 5.2, | |
| "grad_norm": 0.068359375, | |
| "learning_rate": 0.001799873778920377, | |
| "loss": 0.0672, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 5.333333333333333, | |
| "grad_norm": 0.1376953125, | |
| "learning_rate": 0.0017864166602993437, | |
| "loss": 0.0671, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.466666666666667, | |
| "grad_norm": 0.080078125, | |
| "learning_rate": 0.0017725822666400696, | |
| "loss": 0.0654, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "grad_norm": 0.0732421875, | |
| "learning_rate": 0.001758378201737302, | |
| "loss": 0.0574, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 5.733333333333333, | |
| "grad_norm": 0.06591796875, | |
| "learning_rate": 0.001743812272568115, | |
| "loss": 0.0593, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "eval_commonvoice-bn-transcription_loss": 0.6580834984779358, | |
| "eval_commonvoice-bn-transcription_model_preparation_time": 0.0074, | |
| "eval_commonvoice-bn-transcription_runtime": 11.8027, | |
| "eval_commonvoice-bn-transcription_samples_per_second": 5.422, | |
| "eval_commonvoice-bn-transcription_steps_per_second": 0.678, | |
| "step": 4320 | |
| }, | |
| { | |
| "epoch": 5.866666666666667, | |
| "grad_norm": 0.038330078125, | |
| "learning_rate": 0.0017288924850009575, | |
| "loss": 0.0578, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.09423828125, | |
| "learning_rate": 0.0017136270393953942, | |
| "loss": 0.0545, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 6.133333333333334, | |
| "grad_norm": 0.09521484375, | |
| "learning_rate": 0.0016980243260949395, | |
| "loss": 0.0505, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 6.266666666666667, | |
| "grad_norm": 0.037841796875, | |
| "learning_rate": 0.0016820929208154786, | |
| "loss": 0.0521, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "grad_norm": 0.037109375, | |
| "learning_rate": 0.0016658415799317965, | |
| "loss": 0.0513, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 6.533333333333333, | |
| "grad_norm": 0.02783203125, | |
| "learning_rate": 0.001649279235664813, | |
| "loss": 0.0495, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "grad_norm": 0.031494140625, | |
| "learning_rate": 0.0016324149911721704, | |
| "loss": 0.0479, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "eval_commonvoice-bn-transcription_loss": 0.6727238893508911, | |
| "eval_commonvoice-bn-transcription_model_preparation_time": 0.0074, | |
| "eval_commonvoice-bn-transcription_runtime": 14.0248, | |
| "eval_commonvoice-bn-transcription_samples_per_second": 4.563, | |
| "eval_commonvoice-bn-transcription_steps_per_second": 0.57, | |
| "step": 5040 | |
| }, | |
| { | |
| "epoch": 6.8, | |
| "grad_norm": 0.034423828125, | |
| "learning_rate": 0.0016152581155448648, | |
| "loss": 0.0473, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 6.933333333333334, | |
| "grad_norm": 0.061279296875, | |
| "learning_rate": 0.0015978180387126795, | |
| "loss": 0.0474, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 7.066666666666666, | |
| "grad_norm": 0.033447265625, | |
| "learning_rate": 0.0015801043462612132, | |
| "loss": 0.0445, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "grad_norm": 0.041748046875, | |
| "learning_rate": 0.0015621267741633578, | |
| "loss": 0.045, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 7.333333333333333, | |
| "grad_norm": 0.0234375, | |
| "learning_rate": 0.0015438952034281166, | |
| "loss": 0.0444, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 7.466666666666667, | |
| "grad_norm": 0.033447265625, | |
| "learning_rate": 0.0015254196546697087, | |
| "loss": 0.0444, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 7.6, | |
| "grad_norm": 0.03271484375, | |
| "learning_rate": 0.0015067102825999403, | |
| "loss": 0.0426, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 7.68, | |
| "eval_commonvoice-bn-transcription_loss": 0.6831041574478149, | |
| "eval_commonvoice-bn-transcription_model_preparation_time": 0.0074, | |
| "eval_commonvoice-bn-transcription_runtime": 16.8726, | |
| "eval_commonvoice-bn-transcription_samples_per_second": 3.793, | |
| "eval_commonvoice-bn-transcription_steps_per_second": 0.474, | |
| "step": 5760 | |
| }, | |
| { | |
| "epoch": 7.733333333333333, | |
| "grad_norm": 0.05029296875, | |
| "learning_rate": 0.0014877773704468733, | |
| "loss": 0.0426, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 7.866666666666667, | |
| "grad_norm": 0.0262451171875, | |
| "learning_rate": 0.001468631324302856, | |
| "loss": 0.0426, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.0283203125, | |
| "learning_rate": 0.0014492826674050248, | |
| "loss": 0.0416, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 8.133333333333333, | |
| "grad_norm": 0.0224609375, | |
| "learning_rate": 0.0014297420343514216, | |
| "loss": 0.0409, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 8.266666666666667, | |
| "grad_norm": 0.0213623046875, | |
| "learning_rate": 0.0014100201652558998, | |
| "loss": 0.0411, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 8.4, | |
| "grad_norm": 0.036865234375, | |
| "learning_rate": 0.0013901278998450384, | |
| "loss": 0.0415, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 8.533333333333333, | |
| "grad_norm": 0.022705078125, | |
| "learning_rate": 0.0013700761715003068, | |
| "loss": 0.04, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 8.64, | |
| "eval_commonvoice-bn-transcription_loss": 0.6834076642990112, | |
| "eval_commonvoice-bn-transcription_model_preparation_time": 0.0074, | |
| "eval_commonvoice-bn-transcription_runtime": 12.4556, | |
| "eval_commonvoice-bn-transcription_samples_per_second": 5.138, | |
| "eval_commonvoice-bn-transcription_steps_per_second": 0.642, | |
| "step": 6480 | |
| }, | |
| { | |
| "epoch": 8.666666666666666, | |
| "grad_norm": 0.0191650390625, | |
| "learning_rate": 0.0013498760012487503, | |
| "loss": 0.0402, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 8.8, | |
| "grad_norm": 0.03515625, | |
| "learning_rate": 0.001329538491705509, | |
| "loss": 0.04, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 8.933333333333334, | |
| "grad_norm": 0.043701171875, | |
| "learning_rate": 0.001309074820971485, | |
| "loss": 0.0405, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 9.066666666666666, | |
| "grad_norm": 0.020751953125, | |
| "learning_rate": 0.0012884962364895303, | |
| "loss": 0.0375, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 9.2, | |
| "grad_norm": 0.022216796875, | |
| "learning_rate": 0.0012678140488625131, | |
| "loss": 0.0391, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 9.333333333333334, | |
| "grad_norm": 0.03515625, | |
| "learning_rate": 0.0012470396256366771, | |
| "loss": 0.0393, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 9.466666666666667, | |
| "grad_norm": 0.0213623046875, | |
| "learning_rate": 0.0012261843850536976, | |
| "loss": 0.0396, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "grad_norm": 0.0194091796875, | |
| "learning_rate": 0.0012052597897748746, | |
| "loss": 0.0375, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "eval_commonvoice-bn-transcription_loss": 0.6838600039482117, | |
| "eval_commonvoice-bn-transcription_model_preparation_time": 0.0074, | |
| "eval_commonvoice-bn-transcription_runtime": 13.9275, | |
| "eval_commonvoice-bn-transcription_samples_per_second": 4.595, | |
| "eval_commonvoice-bn-transcription_steps_per_second": 0.574, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 9.733333333333333, | |
| "grad_norm": 0.04052734375, | |
| "learning_rate": 0.001184277340580916, | |
| "loss": 0.0385, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 9.866666666666667, | |
| "grad_norm": 0.02880859375, | |
| "learning_rate": 0.0011632485700507637, | |
| "loss": 0.0389, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.0234375, | |
| "learning_rate": 0.001142185036222946, | |
| "loss": 0.0377, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 10.133333333333333, | |
| "grad_norm": 0.01708984375, | |
| "learning_rate": 0.0011210983162429347, | |
| "loss": 0.037, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 10.266666666666667, | |
| "grad_norm": 0.023193359375, | |
| "learning_rate": 0.0011, | |
| "loss": 0.0378, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 10.4, | |
| "grad_norm": 0.03662109375, | |
| "learning_rate": 0.0010789016837570657, | |
| "loss": 0.0384, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 10.533333333333333, | |
| "grad_norm": 0.0201416015625, | |
| "learning_rate": 0.001057814963777054, | |
| "loss": 0.0371, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 10.56, | |
| "eval_commonvoice-bn-transcription_loss": 0.6807686686515808, | |
| "eval_commonvoice-bn-transcription_model_preparation_time": 0.0074, | |
| "eval_commonvoice-bn-transcription_runtime": 13.6962, | |
| "eval_commonvoice-bn-transcription_samples_per_second": 4.673, | |
| "eval_commonvoice-bn-transcription_steps_per_second": 0.584, | |
| "step": 7920 | |
| }, | |
| { | |
| "epoch": 10.666666666666666, | |
| "grad_norm": 0.01806640625, | |
| "learning_rate": 0.0010367514299492366, | |
| "loss": 0.0374, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 10.8, | |
| "grad_norm": 0.032470703125, | |
| "learning_rate": 0.0010157226594190844, | |
| "loss": 0.0375, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 10.933333333333334, | |
| "grad_norm": 0.043212890625, | |
| "learning_rate": 0.0009947402102251258, | |
| "loss": 0.0381, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 11.066666666666666, | |
| "grad_norm": 0.0260009765625, | |
| "learning_rate": 0.0009738156149463029, | |
| "loss": 0.0355, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 11.2, | |
| "grad_norm": 0.0235595703125, | |
| "learning_rate": 0.0009529603743633229, | |
| "loss": 0.0371, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 11.333333333333334, | |
| "grad_norm": 0.03271484375, | |
| "learning_rate": 0.000932185951137487, | |
| "loss": 0.0373, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 11.466666666666667, | |
| "grad_norm": 0.02392578125, | |
| "learning_rate": 0.0009115037635104702, | |
| "loss": 0.0376, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 11.52, | |
| "eval_commonvoice-bn-transcription_loss": 0.6774536967277527, | |
| "eval_commonvoice-bn-transcription_model_preparation_time": 0.0074, | |
| "eval_commonvoice-bn-transcription_runtime": 12.1429, | |
| "eval_commonvoice-bn-transcription_samples_per_second": 5.271, | |
| "eval_commonvoice-bn-transcription_steps_per_second": 0.659, | |
| "step": 8640 | |
| }, | |
| { | |
| "epoch": 11.6, | |
| "grad_norm": 0.0205078125, | |
| "learning_rate": 0.0008909251790285153, | |
| "loss": 0.0359, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 11.733333333333333, | |
| "grad_norm": 0.0198974609375, | |
| "learning_rate": 0.0008704615082944913, | |
| "loss": 0.0368, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 11.866666666666667, | |
| "grad_norm": 0.0234375, | |
| "learning_rate": 0.0008501239987512495, | |
| "loss": 0.0373, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.0238037109375, | |
| "learning_rate": 0.0008299238284996935, | |
| "loss": 0.0363, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 12.133333333333333, | |
| "grad_norm": 0.0164794921875, | |
| "learning_rate": 0.0008098721001549618, | |
| "loss": 0.0358, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 12.266666666666667, | |
| "grad_norm": 0.018798828125, | |
| "learning_rate": 0.0007899798347441006, | |
| "loss": 0.0366, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 12.4, | |
| "grad_norm": 0.048583984375, | |
| "learning_rate": 0.0007702579656485785, | |
| "loss": 0.0371, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 12.48, | |
| "eval_commonvoice-bn-transcription_loss": 0.683639645576477, | |
| "eval_commonvoice-bn-transcription_model_preparation_time": 0.0074, | |
| "eval_commonvoice-bn-transcription_runtime": 13.4595, | |
| "eval_commonvoice-bn-transcription_samples_per_second": 4.755, | |
| "eval_commonvoice-bn-transcription_steps_per_second": 0.594, | |
| "step": 9360 | |
| }, | |
| { | |
| "epoch": 12.533333333333333, | |
| "grad_norm": 0.0189208984375, | |
| "learning_rate": 0.0007507173325949752, | |
| "loss": 0.036, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 12.666666666666666, | |
| "grad_norm": 0.0272216796875, | |
| "learning_rate": 0.0007313686756971443, | |
| "loss": 0.0363, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 12.8, | |
| "grad_norm": 0.0306396484375, | |
| "learning_rate": 0.0007122226295531266, | |
| "loss": 0.0365, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 12.933333333333334, | |
| "grad_norm": 0.035400390625, | |
| "learning_rate": 0.0006932897174000596, | |
| "loss": 0.037, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 13.066666666666666, | |
| "grad_norm": 0.02197265625, | |
| "learning_rate": 0.0006745803453302912, | |
| "loss": 0.0347, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 13.2, | |
| "grad_norm": 0.0277099609375, | |
| "learning_rate": 0.0006561047965718835, | |
| "loss": 0.0363, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 13.333333333333334, | |
| "grad_norm": 0.017333984375, | |
| "learning_rate": 0.0006378732258366421, | |
| "loss": 0.0365, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 13.44, | |
| "eval_commonvoice-bn-transcription_loss": 0.6834643483161926, | |
| "eval_commonvoice-bn-transcription_model_preparation_time": 0.0074, | |
| "eval_commonvoice-bn-transcription_runtime": 13.0019, | |
| "eval_commonvoice-bn-transcription_samples_per_second": 4.922, | |
| "eval_commonvoice-bn-transcription_steps_per_second": 0.615, | |
| "step": 10080 | |
| }, | |
| { | |
| "epoch": 13.466666666666667, | |
| "grad_norm": 0.0230712890625, | |
| "learning_rate": 0.0006198956537387869, | |
| "loss": 0.0369, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 13.6, | |
| "grad_norm": 0.0281982421875, | |
| "learning_rate": 0.0006021819612873205, | |
| "loss": 0.0352, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 13.733333333333333, | |
| "grad_norm": 0.018310546875, | |
| "learning_rate": 0.0005847418844551355, | |
| "loss": 0.0362, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 13.866666666666667, | |
| "grad_norm": 0.0194091796875, | |
| "learning_rate": 0.0005675850088278299, | |
| "loss": 0.0367, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 0.03271484375, | |
| "learning_rate": 0.0005507207643351873, | |
| "loss": 0.0358, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 14.133333333333333, | |
| "grad_norm": 0.0166015625, | |
| "learning_rate": 0.0005341584200682039, | |
| "loss": 0.0353, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 14.266666666666667, | |
| "grad_norm": 0.017578125, | |
| "learning_rate": 0.0005179070791845213, | |
| "loss": 0.0361, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 14.4, | |
| "grad_norm": 0.04833984375, | |
| "learning_rate": 0.0005019756739050605, | |
| "loss": 0.0367, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 14.4, | |
| "eval_commonvoice-bn-transcription_loss": 0.6827465295791626, | |
| "eval_commonvoice-bn-transcription_model_preparation_time": 0.0074, | |
| "eval_commonvoice-bn-transcription_runtime": 13.7129, | |
| "eval_commonvoice-bn-transcription_samples_per_second": 4.667, | |
| "eval_commonvoice-bn-transcription_steps_per_second": 0.583, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 14.533333333333333, | |
| "grad_norm": 0.018798828125, | |
| "learning_rate": 0.00048637296060460595, | |
| "loss": 0.0357, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 14.666666666666666, | |
| "grad_norm": 0.04296875, | |
| "learning_rate": 0.0004711075149990425, | |
| "loss": 0.0359, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 14.8, | |
| "grad_norm": 0.0255126953125, | |
| "learning_rate": 0.0004561877274318854, | |
| "loss": 0.0361, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 14.933333333333334, | |
| "grad_norm": 0.03515625, | |
| "learning_rate": 0.0004416217982626981, | |
| "loss": 0.0367, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 15.066666666666666, | |
| "grad_norm": 0.018310546875, | |
| "learning_rate": 0.0004274177333599306, | |
| "loss": 0.0345, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 15.2, | |
| "grad_norm": 0.0194091796875, | |
| "learning_rate": 0.00041358333970065636, | |
| "loss": 0.036, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 15.333333333333334, | |
| "grad_norm": 0.01806640625, | |
| "learning_rate": 0.00040012622107962314, | |
| "loss": 0.0363, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 15.36, | |
| "eval_commonvoice-bn-transcription_loss": 0.6828598380088806, | |
| "eval_commonvoice-bn-transcription_model_preparation_time": 0.0074, | |
| "eval_commonvoice-bn-transcription_runtime": 13.9421, | |
| "eval_commonvoice-bn-transcription_samples_per_second": 4.59, | |
| "eval_commonvoice-bn-transcription_steps_per_second": 0.574, | |
| "step": 11520 | |
| }, | |
| { | |
| "epoch": 15.466666666666667, | |
| "grad_norm": 0.034423828125, | |
| "learning_rate": 0.0003870537739299836, | |
| "loss": 0.0366, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 15.6, | |
| "grad_norm": 0.02197265625, | |
| "learning_rate": 0.0003743731832579933, | |
| "loss": 0.035, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 15.733333333333333, | |
| "grad_norm": 0.017822265625, | |
| "learning_rate": 0.00036209141869391796, | |
| "loss": 0.036, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 15.866666666666667, | |
| "grad_norm": 0.0179443359375, | |
| "learning_rate": 0.00035021523066131776, | |
| "loss": 0.0365, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.0252685546875, | |
| "learning_rate": 0.00033875114666681236, | |
| "loss": 0.0356, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 16.133333333333333, | |
| "grad_norm": 0.0284423828125, | |
| "learning_rate": 0.0003277054677123703, | |
| "loss": 0.0351, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 16.266666666666666, | |
| "grad_norm": 0.01708984375, | |
| "learning_rate": 0.00031708426483208885, | |
| "loss": 0.0359, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 16.32, | |
| "eval_commonvoice-bn-transcription_loss": 0.6832164525985718, | |
| "eval_commonvoice-bn-transcription_model_preparation_time": 0.0074, | |
| "eval_commonvoice-bn-transcription_runtime": 16.9968, | |
| "eval_commonvoice-bn-transcription_samples_per_second": 3.765, | |
| "eval_commonvoice-bn-transcription_steps_per_second": 0.471, | |
| "step": 12240 | |
| }, | |
| { | |
| "epoch": 16.4, | |
| "grad_norm": 0.038818359375, | |
| "learning_rate": 0.00030689337575537455, | |
| "loss": 0.0365, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 16.533333333333335, | |
| "grad_norm": 0.019775390625, | |
| "learning_rate": 0.00029713840169835217, | |
| "loss": 0.0355, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 16.666666666666668, | |
| "grad_norm": 0.03369140625, | |
| "learning_rate": 0.0002878247042852705, | |
| "loss": 0.0358, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 16.8, | |
| "grad_norm": 0.0283203125, | |
| "learning_rate": 0.0002789574026015931, | |
| "loss": 0.036, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 16.933333333333334, | |
| "grad_norm": 0.035888671875, | |
| "learning_rate": 0.0002705413703803964, | |
| "loss": 0.0366, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 17.066666666666666, | |
| "grad_norm": 0.0174560546875, | |
| "learning_rate": 0.00026258123332362215, | |
| "loss": 0.0344, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 17.2, | |
| "grad_norm": 0.016845703125, | |
| "learning_rate": 0.0002550813665596523, | |
| "loss": 0.0359, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 17.28, | |
| "eval_commonvoice-bn-transcription_loss": 0.6824997067451477, | |
| "eval_commonvoice-bn-transcription_model_preparation_time": 0.0074, | |
| "eval_commonvoice-bn-transcription_runtime": 13.9582, | |
| "eval_commonvoice-bn-transcription_samples_per_second": 4.585, | |
| "eval_commonvoice-bn-transcription_steps_per_second": 0.573, | |
| "step": 12960 | |
| }, | |
| { | |
| "epoch": 17.333333333333332, | |
| "grad_norm": 0.0177001953125, | |
| "learning_rate": 0.00024804589223860934, | |
| "loss": 0.0362, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 17.466666666666665, | |
| "grad_norm": 0.02734375, | |
| "learning_rate": 0.00024147867726669888, | |
| "loss": 0.0365, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 17.6, | |
| "grad_norm": 0.0191650390625, | |
| "learning_rate": 0.00023538333118084396, | |
| "loss": 0.035, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 17.733333333333334, | |
| "grad_norm": 0.026611328125, | |
| "learning_rate": 0.00022976320416477446, | |
| "loss": 0.0359, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 17.866666666666667, | |
| "grad_norm": 0.0189208984375, | |
| "learning_rate": 0.00022462138520766832, | |
| "loss": 0.0364, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 0.02294921875, | |
| "learning_rate": 0.0002199607004063494, | |
| "loss": 0.0356, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 18.133333333333333, | |
| "grad_norm": 0.0390625, | |
| "learning_rate": 0.00021578371141198153, | |
| "loss": 0.0352, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 18.24, | |
| "eval_commonvoice-bn-transcription_loss": 0.6843233704566956, | |
| "eval_commonvoice-bn-transcription_model_preparation_time": 0.0074, | |
| "eval_commonvoice-bn-transcription_runtime": 13.7375, | |
| "eval_commonvoice-bn-transcription_samples_per_second": 4.659, | |
| "eval_commonvoice-bn-transcription_steps_per_second": 0.582, | |
| "step": 13680 | |
| }, | |
| { | |
| "epoch": 18.266666666666666, | |
| "grad_norm": 0.0228271484375, | |
| "learning_rate": 0.00021209271402210822, | |
| "loss": 0.0359, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 18.4, | |
| "grad_norm": 0.044921875, | |
| "learning_rate": 0.00020888973691881416, | |
| "loss": 0.0365, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 18.533333333333335, | |
| "grad_norm": 0.0262451171875, | |
| "learning_rate": 0.0002061765405537013, | |
| "loss": 0.0355, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 18.666666666666668, | |
| "grad_norm": 0.026611328125, | |
| "learning_rate": 0.00020395461618029175, | |
| "loss": 0.0358, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 18.8, | |
| "grad_norm": 0.0201416015625, | |
| "learning_rate": 0.0002022251850343909, | |
| "loss": 0.036, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 18.933333333333334, | |
| "grad_norm": 0.0439453125, | |
| "learning_rate": 0.0002009891976628598, | |
| "loss": 0.0365, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 19.066666666666666, | |
| "grad_norm": 0.021728515625, | |
| "learning_rate": 0.00020024733340116572, | |
| "loss": 0.0344, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 19.2, | |
| "grad_norm": 0.0177001953125, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0358, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 19.2, | |
| "eval_commonvoice-bn-transcription_loss": 0.6850356459617615, | |
| "eval_commonvoice-bn-transcription_model_preparation_time": 0.0074, | |
| "eval_commonvoice-bn-transcription_runtime": 15.339, | |
| "eval_commonvoice-bn-transcription_samples_per_second": 4.172, | |
| "eval_commonvoice-bn-transcription_steps_per_second": 0.522, | |
| "step": 14400 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 14400, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 3600, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6.421884240590438e+17, | |
| "train_batch_size": 24, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |