{
  "best_metric": 38.50746268656716,
  "best_model_checkpoint": "./content/drive/MyDrive/kyrgyz_asr/checkpoint-4000",
  "epoch": 1.893939393939394,
  "eval_steps": 1000,
  "global_step": 4000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.011837121212121212,
      "grad_norm": 38.770484924316406,
      "learning_rate": 4.4e-07,
      "loss": 3.1763,
      "step": 25
    },
    {
      "epoch": 0.023674242424242424,
      "grad_norm": 26.670623779296875,
      "learning_rate": 9.400000000000001e-07,
      "loss": 2.904,
      "step": 50
    },
    {
      "epoch": 0.03551136363636364,
      "grad_norm": 24.12047576904297,
      "learning_rate": 1.44e-06,
      "loss": 2.573,
      "step": 75
    },
    {
      "epoch": 0.04734848484848485,
      "grad_norm": 21.595748901367188,
      "learning_rate": 1.94e-06,
      "loss": 2.3106,
      "step": 100
    },
    {
      "epoch": 0.059185606060606064,
      "grad_norm": 18.133272171020508,
      "learning_rate": 2.4400000000000004e-06,
      "loss": 1.9787,
      "step": 125
    },
    {
      "epoch": 0.07102272727272728,
      "grad_norm": 17.44855499267578,
      "learning_rate": 2.9400000000000002e-06,
      "loss": 1.7705,
      "step": 150
    },
    {
      "epoch": 0.08285984848484848,
      "grad_norm": 16.95909309387207,
      "learning_rate": 3.44e-06,
      "loss": 1.5665,
      "step": 175
    },
    {
      "epoch": 0.0946969696969697,
      "grad_norm": 17.190784454345703,
      "learning_rate": 3.94e-06,
      "loss": 1.4962,
      "step": 200
    },
    {
      "epoch": 0.10653409090909091,
      "grad_norm": 16.988037109375,
      "learning_rate": 4.440000000000001e-06,
      "loss": 1.4281,
      "step": 225
    },
    {
      "epoch": 0.11837121212121213,
      "grad_norm": 14.77700424194336,
      "learning_rate": 4.94e-06,
      "loss": 1.3685,
      "step": 250
    },
    {
      "epoch": 0.13020833333333334,
      "grad_norm": 15.917333602905273,
      "learning_rate": 5.4400000000000004e-06,
      "loss": 1.2807,
      "step": 275
    },
    {
      "epoch": 0.14204545454545456,
      "grad_norm": 14.992497444152832,
      "learning_rate": 5.94e-06,
      "loss": 1.1803,
      "step": 300
    },
    {
      "epoch": 0.15388257575757575,
      "grad_norm": 15.551587104797363,
      "learning_rate": 6.440000000000001e-06,
      "loss": 1.148,
      "step": 325
    },
    {
      "epoch": 0.16571969696969696,
      "grad_norm": 14.418575286865234,
      "learning_rate": 6.9400000000000005e-06,
      "loss": 1.0953,
      "step": 350
    },
    {
      "epoch": 0.17755681818181818,
      "grad_norm": 15.400343894958496,
      "learning_rate": 7.440000000000001e-06,
      "loss": 1.0356,
      "step": 375
    },
    {
      "epoch": 0.1893939393939394,
      "grad_norm": 12.426461219787598,
      "learning_rate": 7.94e-06,
      "loss": 0.9864,
      "step": 400
    },
    {
      "epoch": 0.2012310606060606,
      "grad_norm": 13.831282615661621,
      "learning_rate": 8.44e-06,
      "loss": 0.9653,
      "step": 425
    },
    {
      "epoch": 0.21306818181818182,
      "grad_norm": 15.357656478881836,
      "learning_rate": 8.94e-06,
      "loss": 0.9935,
      "step": 450
    },
    {
      "epoch": 0.22490530303030304,
      "grad_norm": 13.355987548828125,
      "learning_rate": 9.440000000000001e-06,
      "loss": 0.9047,
      "step": 475
    },
    {
      "epoch": 0.23674242424242425,
      "grad_norm": 13.817172050476074,
      "learning_rate": 9.940000000000001e-06,
      "loss": 0.9242,
      "step": 500
    },
    {
      "epoch": 0.24857954545454544,
      "grad_norm": 15.737274169921875,
      "learning_rate": 9.937142857142858e-06,
      "loss": 0.8997,
      "step": 525
    },
    {
      "epoch": 0.2604166666666667,
      "grad_norm": 12.787036895751953,
      "learning_rate": 9.865714285714285e-06,
      "loss": 0.8344,
      "step": 550
    },
    {
      "epoch": 0.2722537878787879,
      "grad_norm": 13.924437522888184,
      "learning_rate": 9.794285714285714e-06,
      "loss": 0.8179,
      "step": 575
    },
    {
      "epoch": 0.2840909090909091,
      "grad_norm": 14.679474830627441,
      "learning_rate": 9.722857142857143e-06,
      "loss": 0.7755,
      "step": 600
    },
    {
      "epoch": 0.2959280303030303,
      "grad_norm": 13.893200874328613,
      "learning_rate": 9.651428571428572e-06,
      "loss": 0.7419,
      "step": 625
    },
    {
      "epoch": 0.3077651515151515,
      "grad_norm": 14.868252754211426,
      "learning_rate": 9.58e-06,
      "loss": 0.7708,
      "step": 650
    },
    {
      "epoch": 0.3196022727272727,
      "grad_norm": 14.574288368225098,
      "learning_rate": 9.508571428571429e-06,
      "loss": 0.7333,
      "step": 675
    },
    {
      "epoch": 0.3314393939393939,
      "grad_norm": 15.690669059753418,
      "learning_rate": 9.437142857142858e-06,
      "loss": 0.7006,
      "step": 700
    },
    {
      "epoch": 0.34327651515151514,
      "grad_norm": 12.136014938354492,
      "learning_rate": 9.365714285714287e-06,
      "loss": 0.705,
      "step": 725
    },
    {
      "epoch": 0.35511363636363635,
      "grad_norm": 13.789889335632324,
      "learning_rate": 9.294285714285714e-06,
      "loss": 0.7195,
      "step": 750
    },
    {
      "epoch": 0.36695075757575757,
      "grad_norm": 11.885604858398438,
      "learning_rate": 9.222857142857143e-06,
      "loss": 0.6753,
      "step": 775
    },
    {
      "epoch": 0.3787878787878788,
      "grad_norm": 12.935856819152832,
      "learning_rate": 9.151428571428572e-06,
      "loss": 0.6545,
      "step": 800
    },
    {
      "epoch": 0.390625,
      "grad_norm": 13.731322288513184,
      "learning_rate": 9.080000000000001e-06,
      "loss": 0.7291,
      "step": 825
    },
    {
      "epoch": 0.4024621212121212,
      "grad_norm": 13.80007553100586,
      "learning_rate": 9.00857142857143e-06,
      "loss": 0.6667,
      "step": 850
    },
    {
      "epoch": 0.41429924242424243,
      "grad_norm": 11.839824676513672,
      "learning_rate": 8.937142857142857e-06,
      "loss": 0.6221,
      "step": 875
    },
    {
      "epoch": 0.42613636363636365,
      "grad_norm": 12.550195693969727,
      "learning_rate": 8.865714285714287e-06,
      "loss": 0.6114,
      "step": 900
    },
    {
      "epoch": 0.43797348484848486,
      "grad_norm": 12.121932029724121,
      "learning_rate": 8.794285714285716e-06,
      "loss": 0.652,
      "step": 925
    },
    {
      "epoch": 0.4498106060606061,
      "grad_norm": 13.047248840332031,
      "learning_rate": 8.722857142857145e-06,
      "loss": 0.6201,
      "step": 950
    },
    {
      "epoch": 0.4616477272727273,
      "grad_norm": 12.148579597473145,
      "learning_rate": 8.651428571428572e-06,
      "loss": 0.622,
      "step": 975
    },
    {
      "epoch": 0.4734848484848485,
      "grad_norm": 14.126455307006836,
      "learning_rate": 8.580000000000001e-06,
      "loss": 0.59,
      "step": 1000
    },
    {
      "epoch": 0.4734848484848485,
      "eval_loss": 0.5916658043861389,
      "eval_runtime": 413.6011,
      "eval_samples_per_second": 3.902,
      "eval_steps_per_second": 0.488,
      "eval_wer": 60.805065581184984,
      "step": 1000
    },
    {
      "epoch": 0.4853219696969697,
      "grad_norm": 11.438024520874023,
      "learning_rate": 8.50857142857143e-06,
      "loss": 0.6024,
      "step": 1025
    },
    {
      "epoch": 0.4971590909090909,
      "grad_norm": 11.665973663330078,
      "learning_rate": 8.437142857142859e-06,
      "loss": 0.5949,
      "step": 1050
    },
    {
      "epoch": 0.5089962121212122,
      "grad_norm": 12.14466381072998,
      "learning_rate": 8.365714285714286e-06,
      "loss": 0.5825,
      "step": 1075
    },
    {
      "epoch": 0.5208333333333334,
      "grad_norm": 11.364850044250488,
      "learning_rate": 8.294285714285715e-06,
      "loss": 0.5962,
      "step": 1100
    },
    {
      "epoch": 0.5326704545454546,
      "grad_norm": 13.55361557006836,
      "learning_rate": 8.222857142857144e-06,
      "loss": 0.5897,
      "step": 1125
    },
    {
      "epoch": 0.5445075757575758,
      "grad_norm": 11.306770324707031,
      "learning_rate": 8.151428571428572e-06,
      "loss": 0.5654,
      "step": 1150
    },
    {
      "epoch": 0.556344696969697,
      "grad_norm": 11.402146339416504,
      "learning_rate": 8.08e-06,
      "loss": 0.5916,
      "step": 1175
    },
    {
      "epoch": 0.5681818181818182,
      "grad_norm": 10.837797164916992,
      "learning_rate": 8.00857142857143e-06,
      "loss": 0.5345,
      "step": 1200
    },
    {
      "epoch": 0.5800189393939394,
      "grad_norm": 12.855684280395508,
      "learning_rate": 7.937142857142857e-06,
      "loss": 0.5263,
      "step": 1225
    },
    {
      "epoch": 0.5918560606060606,
      "grad_norm": 11.619951248168945,
      "learning_rate": 7.865714285714286e-06,
      "loss": 0.5461,
      "step": 1250
    },
    {
      "epoch": 0.6036931818181818,
      "grad_norm": 11.1388578414917,
      "learning_rate": 7.794285714285715e-06,
      "loss": 0.5625,
      "step": 1275
    },
    {
      "epoch": 0.615530303030303,
      "grad_norm": 11.090073585510254,
      "learning_rate": 7.722857142857142e-06,
      "loss": 0.5178,
      "step": 1300
    },
    {
      "epoch": 0.6273674242424242,
      "grad_norm": 10.146711349487305,
      "learning_rate": 7.651428571428571e-06,
      "loss": 0.5201,
      "step": 1325
    },
    {
      "epoch": 0.6392045454545454,
      "grad_norm": 16.155235290527344,
      "learning_rate": 7.58e-06,
      "loss": 0.5247,
      "step": 1350
    },
    {
      "epoch": 0.6510416666666666,
      "grad_norm": 12.52512264251709,
      "learning_rate": 7.508571428571429e-06,
      "loss": 0.5237,
      "step": 1375
    },
    {
      "epoch": 0.6628787878787878,
      "grad_norm": 9.786041259765625,
      "learning_rate": 7.4371428571428575e-06,
      "loss": 0.483,
      "step": 1400
    },
    {
      "epoch": 0.6747159090909091,
      "grad_norm": 11.963520050048828,
      "learning_rate": 7.365714285714286e-06,
      "loss": 0.5099,
      "step": 1425
    },
    {
      "epoch": 0.6865530303030303,
      "grad_norm": 11.08193302154541,
      "learning_rate": 7.294285714285715e-06,
      "loss": 0.5417,
      "step": 1450
    },
    {
      "epoch": 0.6983901515151515,
      "grad_norm": 10.529409408569336,
      "learning_rate": 7.222857142857144e-06,
      "loss": 0.4837,
      "step": 1475
    },
    {
      "epoch": 0.7102272727272727,
      "grad_norm": 10.106945037841797,
      "learning_rate": 7.151428571428573e-06,
      "loss": 0.5158,
      "step": 1500
    },
    {
      "epoch": 0.7220643939393939,
      "grad_norm": 13.151567459106445,
      "learning_rate": 7.08e-06,
      "loss": 0.5181,
      "step": 1525
    },
    {
      "epoch": 0.7339015151515151,
      "grad_norm": 13.37302303314209,
      "learning_rate": 7.008571428571429e-06,
      "loss": 0.5059,
      "step": 1550
    },
    {
      "epoch": 0.7457386363636364,
      "grad_norm": 11.584297180175781,
      "learning_rate": 6.937142857142858e-06,
      "loss": 0.5098,
      "step": 1575
    },
    {
      "epoch": 0.7575757575757576,
      "grad_norm": 9.750558853149414,
      "learning_rate": 6.865714285714287e-06,
      "loss": 0.4852,
      "step": 1600
    },
    {
      "epoch": 0.7694128787878788,
      "grad_norm": 13.944685935974121,
      "learning_rate": 6.794285714285714e-06,
      "loss": 0.5334,
      "step": 1625
    },
    {
      "epoch": 0.78125,
      "grad_norm": 11.0306396484375,
      "learning_rate": 6.722857142857143e-06,
      "loss": 0.5418,
      "step": 1650
    },
    {
      "epoch": 0.7930871212121212,
      "grad_norm": 10.839930534362793,
      "learning_rate": 6.651428571428572e-06,
      "loss": 0.4968,
      "step": 1675
    },
    {
      "epoch": 0.8049242424242424,
      "grad_norm": 10.40273380279541,
      "learning_rate": 6.5800000000000005e-06,
      "loss": 0.4605,
      "step": 1700
    },
    {
      "epoch": 0.8167613636363636,
      "grad_norm": 12.725444793701172,
      "learning_rate": 6.5085714285714295e-06,
      "loss": 0.4701,
      "step": 1725
    },
    {
      "epoch": 0.8285984848484849,
      "grad_norm": 10.695274353027344,
      "learning_rate": 6.437142857142858e-06,
      "loss": 0.464,
      "step": 1750
    },
    {
      "epoch": 0.8404356060606061,
      "grad_norm": 11.814519882202148,
      "learning_rate": 6.365714285714286e-06,
      "loss": 0.4919,
      "step": 1775
    },
    {
      "epoch": 0.8522727272727273,
      "grad_norm": 8.929429054260254,
      "learning_rate": 6.294285714285715e-06,
      "loss": 0.4433,
      "step": 1800
    },
    {
      "epoch": 0.8641098484848485,
      "grad_norm": 12.6130952835083,
      "learning_rate": 6.222857142857144e-06,
      "loss": 0.49,
      "step": 1825
    },
    {
      "epoch": 0.8759469696969697,
      "grad_norm": 9.853967666625977,
      "learning_rate": 6.151428571428571e-06,
      "loss": 0.4424,
      "step": 1850
    },
    {
      "epoch": 0.8877840909090909,
      "grad_norm": 11.585264205932617,
      "learning_rate": 6.08e-06,
      "loss": 0.457,
      "step": 1875
    },
    {
      "epoch": 0.8996212121212122,
      "grad_norm": 10.458426475524902,
      "learning_rate": 6.008571428571429e-06,
      "loss": 0.4771,
      "step": 1900
    },
    {
      "epoch": 0.9114583333333334,
      "grad_norm": 9.979692459106445,
      "learning_rate": 5.937142857142858e-06,
      "loss": 0.4393,
      "step": 1925
    },
    {
      "epoch": 0.9232954545454546,
      "grad_norm": 10.795998573303223,
      "learning_rate": 5.865714285714286e-06,
      "loss": 0.4375,
      "step": 1950
    },
    {
      "epoch": 0.9351325757575758,
      "grad_norm": 11.389334678649902,
      "learning_rate": 5.794285714285715e-06,
      "loss": 0.4289,
      "step": 1975
    },
    {
      "epoch": 0.946969696969697,
      "grad_norm": 11.161763191223145,
      "learning_rate": 5.722857142857144e-06,
      "loss": 0.4987,
      "step": 2000
    },
    {
      "epoch": 0.946969696969697,
      "eval_loss": 0.41947832703590393,
      "eval_runtime": 407.8391,
      "eval_samples_per_second": 3.957,
      "eval_steps_per_second": 0.495,
      "eval_wer": 47.85165083672546,
      "step": 2000
    },
    {
      "epoch": 0.9588068181818182,
      "grad_norm": 8.887962341308594,
      "learning_rate": 5.651428571428572e-06,
      "loss": 0.4188,
      "step": 2025
    },
    {
      "epoch": 0.9706439393939394,
      "grad_norm": 10.415024757385254,
      "learning_rate": 5.580000000000001e-06,
      "loss": 0.4313,
      "step": 2050
    },
    {
      "epoch": 0.9824810606060606,
      "grad_norm": 11.072915077209473,
      "learning_rate": 5.508571428571429e-06,
      "loss": 0.4364,
      "step": 2075
    },
    {
      "epoch": 0.9943181818181818,
      "grad_norm": 13.284914016723633,
      "learning_rate": 5.437142857142857e-06,
      "loss": 0.4868,
      "step": 2100
    },
    {
      "epoch": 1.006155303030303,
      "grad_norm": 9.159514427185059,
      "learning_rate": 5.365714285714286e-06,
      "loss": 0.3972,
      "step": 2125
    },
    {
      "epoch": 1.0179924242424243,
      "grad_norm": 9.100288391113281,
      "learning_rate": 5.294285714285715e-06,
      "loss": 0.4142,
      "step": 2150
    },
    {
      "epoch": 1.0298295454545454,
      "grad_norm": 8.876397132873535,
      "learning_rate": 5.2228571428571425e-06,
      "loss": 0.3867,
      "step": 2175
    },
    {
      "epoch": 1.0416666666666667,
      "grad_norm": 12.151432037353516,
      "learning_rate": 5.1514285714285715e-06,
      "loss": 0.3663,
      "step": 2200
    },
    {
      "epoch": 1.0535037878787878,
      "grad_norm": 9.571785926818848,
      "learning_rate": 5.0800000000000005e-06,
      "loss": 0.3979,
      "step": 2225
    },
    {
      "epoch": 1.0653409090909092,
      "grad_norm": 11.223183631896973,
      "learning_rate": 5.0085714285714295e-06,
      "loss": 0.3899,
      "step": 2250
    },
    {
      "epoch": 1.0771780303030303,
      "grad_norm": 8.434731483459473,
      "learning_rate": 4.937142857142858e-06,
      "loss": 0.3484,
      "step": 2275
    },
    {
      "epoch": 1.0890151515151516,
      "grad_norm": 10.766022682189941,
      "learning_rate": 4.865714285714287e-06,
      "loss": 0.4095,
      "step": 2300
    },
    {
      "epoch": 1.1008522727272727,
      "grad_norm": 8.537741661071777,
      "learning_rate": 4.794285714285715e-06,
      "loss": 0.3688,
      "step": 2325
    },
    {
      "epoch": 1.112689393939394,
      "grad_norm": 10.699164390563965,
      "learning_rate": 4.722857142857144e-06,
      "loss": 0.3559,
      "step": 2350
    },
    {
      "epoch": 1.1245265151515151,
      "grad_norm": 9.601336479187012,
      "learning_rate": 4.651428571428572e-06,
      "loss": 0.3711,
      "step": 2375
    },
    {
      "epoch": 1.1363636363636362,
      "grad_norm": 12.841407775878906,
      "learning_rate": 4.58e-06,
      "loss": 0.3541,
      "step": 2400
    },
    {
      "epoch": 1.1482007575757576,
      "grad_norm": 11.84900951385498,
      "learning_rate": 4.508571428571429e-06,
      "loss": 0.3548,
      "step": 2425
    },
    {
      "epoch": 1.160037878787879,
      "grad_norm": 9.47382926940918,
      "learning_rate": 4.437142857142857e-06,
      "loss": 0.3659,
      "step": 2450
    },
    {
      "epoch": 1.171875,
      "grad_norm": 11.142972946166992,
      "learning_rate": 4.3657142857142855e-06,
      "loss": 0.3473,
      "step": 2475
    },
    {
      "epoch": 1.183712121212121,
      "grad_norm": 9.656496047973633,
      "learning_rate": 4.2942857142857146e-06,
      "loss": 0.3634,
      "step": 2500
    },
    {
      "epoch": 1.1955492424242424,
      "grad_norm": 11.295799255371094,
      "learning_rate": 4.222857142857143e-06,
      "loss": 0.3573,
      "step": 2525
    },
    {
      "epoch": 1.2073863636363638,
      "grad_norm": 11.13376522064209,
      "learning_rate": 4.151428571428572e-06,
      "loss": 0.38,
      "step": 2550
    },
    {
      "epoch": 1.2192234848484849,
      "grad_norm": 10.365365028381348,
      "learning_rate": 4.08e-06,
      "loss": 0.3749,
      "step": 2575
    },
    {
      "epoch": 1.231060606060606,
      "grad_norm": 9.453851699829102,
      "learning_rate": 4.008571428571429e-06,
      "loss": 0.3256,
      "step": 2600
    },
    {
      "epoch": 1.2428977272727273,
      "grad_norm": 8.543546676635742,
      "learning_rate": 3.937142857142858e-06,
      "loss": 0.3744,
      "step": 2625
    },
    {
      "epoch": 1.2547348484848486,
      "grad_norm": 11.191537857055664,
      "learning_rate": 3.865714285714286e-06,
      "loss": 0.3493,
      "step": 2650
    },
    {
      "epoch": 1.2665719696969697,
      "grad_norm": 7.995920181274414,
      "learning_rate": 3.7942857142857147e-06,
      "loss": 0.3684,
      "step": 2675
    },
    {
      "epoch": 1.2784090909090908,
      "grad_norm": 10.559860229492188,
      "learning_rate": 3.722857142857143e-06,
      "loss": 0.3364,
      "step": 2700
    },
    {
      "epoch": 1.2902462121212122,
      "grad_norm": 10.414945602416992,
      "learning_rate": 3.651428571428572e-06,
      "loss": 0.3629,
      "step": 2725
    },
    {
      "epoch": 1.3020833333333333,
      "grad_norm": 8.4398775100708,
      "learning_rate": 3.58e-06,
      "loss": 0.3464,
      "step": 2750
    },
    {
      "epoch": 1.3139204545454546,
      "grad_norm": 11.154573440551758,
      "learning_rate": 3.508571428571429e-06,
      "loss": 0.3394,
      "step": 2775
    },
    {
      "epoch": 1.3257575757575757,
      "grad_norm": 9.244209289550781,
      "learning_rate": 3.437142857142857e-06,
      "loss": 0.402,
      "step": 2800
    },
    {
      "epoch": 1.337594696969697,
      "grad_norm": 9.690404891967773,
      "learning_rate": 3.3657142857142862e-06,
      "loss": 0.3552,
      "step": 2825
    },
    {
      "epoch": 1.3494318181818181,
      "grad_norm": 9.71345043182373,
      "learning_rate": 3.2942857142857144e-06,
      "loss": 0.3483,
      "step": 2850
    },
    {
      "epoch": 1.3612689393939394,
      "grad_norm": 11.922121047973633,
      "learning_rate": 3.222857142857143e-06,
      "loss": 0.3582,
      "step": 2875
    },
    {
      "epoch": 1.3731060606060606,
      "grad_norm": 9.053750038146973,
      "learning_rate": 3.151428571428572e-06,
      "loss": 0.3409,
      "step": 2900
    },
    {
      "epoch": 1.3849431818181819,
      "grad_norm": 11.320926666259766,
      "learning_rate": 3.08e-06,
      "loss": 0.3467,
      "step": 2925
    },
    {
      "epoch": 1.396780303030303,
      "grad_norm": 10.648012161254883,
      "learning_rate": 3.008571428571429e-06,
      "loss": 0.3465,
      "step": 2950
    },
    {
      "epoch": 1.4086174242424243,
      "grad_norm": 9.58975601196289,
      "learning_rate": 2.9371428571428573e-06,
      "loss": 0.3686,
      "step": 2975
    },
    {
      "epoch": 1.4204545454545454,
      "grad_norm": 10.39020824432373,
      "learning_rate": 2.865714285714286e-06,
      "loss": 0.3932,
      "step": 3000
    },
    {
      "epoch": 1.4204545454545454,
      "eval_loss": 0.35611891746520996,
      "eval_runtime": 410.66,
      "eval_samples_per_second": 3.93,
      "eval_steps_per_second": 0.492,
      "eval_wer": 42.66847580280416,
      "step": 3000
    },
    {
      "epoch": 1.4322916666666667,
      "grad_norm": 10.914188385009766,
      "learning_rate": 2.7942857142857145e-06,
      "loss": 0.3501,
      "step": 3025
    },
    {
      "epoch": 1.4441287878787878,
      "grad_norm": 8.137165069580078,
      "learning_rate": 2.722857142857143e-06,
      "loss": 0.339,
      "step": 3050
    },
    {
      "epoch": 1.4559659090909092,
      "grad_norm": 10.744149208068848,
      "learning_rate": 2.6514285714285713e-06,
      "loss": 0.3442,
      "step": 3075
    },
    {
      "epoch": 1.4678030303030303,
      "grad_norm": 10.14416790008545,
      "learning_rate": 2.5800000000000003e-06,
      "loss": 0.3455,
      "step": 3100
    },
    {
      "epoch": 1.4796401515151514,
      "grad_norm": 11.480497360229492,
      "learning_rate": 2.5085714285714285e-06,
      "loss": 0.3431,
      "step": 3125
    },
    {
      "epoch": 1.4914772727272727,
      "grad_norm": 14.735957145690918,
      "learning_rate": 2.4371428571428575e-06,
      "loss": 0.3572,
      "step": 3150
    },
    {
      "epoch": 1.503314393939394,
      "grad_norm": 10.636829376220703,
      "learning_rate": 2.365714285714286e-06,
      "loss": 0.3334,
      "step": 3175
    },
    {
      "epoch": 1.5151515151515151,
      "grad_norm": 8.89909553527832,
      "learning_rate": 2.2942857142857146e-06,
      "loss": 0.3294,
      "step": 3200
    },
    {
      "epoch": 1.5269886363636362,
      "grad_norm": 9.316367149353027,
      "learning_rate": 2.222857142857143e-06,
      "loss": 0.3459,
      "step": 3225
    },
    {
      "epoch": 1.5388257575757576,
      "grad_norm": 9.63460922241211,
      "learning_rate": 2.1514285714285714e-06,
      "loss": 0.3403,
      "step": 3250
    },
    {
      "epoch": 1.550662878787879,
      "grad_norm": 10.554502487182617,
      "learning_rate": 2.08e-06,
      "loss": 0.3656,
      "step": 3275
    },
    {
      "epoch": 1.5625,
      "grad_norm": 9.131353378295898,
      "learning_rate": 2.0085714285714286e-06,
      "loss": 0.3313,
      "step": 3300
    },
    {
      "epoch": 1.574337121212121,
      "grad_norm": 11.400104522705078,
      "learning_rate": 1.9371428571428576e-06,
      "loss": 0.3547,
      "step": 3325
    },
    {
      "epoch": 1.5861742424242424,
      "grad_norm": 10.741024017333984,
      "learning_rate": 1.865714285714286e-06,
      "loss": 0.3639,
      "step": 3350
    },
    {
      "epoch": 1.5980113636363638,
      "grad_norm": 7.954362869262695,
      "learning_rate": 1.7942857142857146e-06,
      "loss": 0.3232,
      "step": 3375
    },
    {
      "epoch": 1.6098484848484849,
      "grad_norm": 8.994126319885254,
      "learning_rate": 1.7228571428571432e-06,
      "loss": 0.3328,
      "step": 3400
    },
    {
      "epoch": 1.621685606060606,
      "grad_norm": 9.671642303466797,
      "learning_rate": 1.6514285714285715e-06,
      "loss": 0.3249,
      "step": 3425
    },
    {
      "epoch": 1.6335227272727273,
      "grad_norm": 12.057549476623535,
      "learning_rate": 1.5800000000000001e-06,
      "loss": 0.3126,
      "step": 3450
    },
    {
      "epoch": 1.6453598484848486,
      "grad_norm": 8.958949089050293,
      "learning_rate": 1.5085714285714287e-06,
      "loss": 0.3185,
      "step": 3475
    },
    {
      "epoch": 1.6571969696969697,
      "grad_norm": 10.565046310424805,
      "learning_rate": 1.4371428571428573e-06,
      "loss": 0.3424,
      "step": 3500
    },
    {
      "epoch": 1.6690340909090908,
      "grad_norm": 9.329957962036133,
      "learning_rate": 1.3657142857142857e-06,
      "loss": 0.337,
      "step": 3525
    },
    {
      "epoch": 1.6808712121212122,
      "grad_norm": 8.723917007446289,
      "learning_rate": 1.2942857142857143e-06,
      "loss": 0.3487,
      "step": 3550
    },
    {
      "epoch": 1.6927083333333335,
      "grad_norm": 9.308070182800293,
      "learning_rate": 1.222857142857143e-06,
      "loss": 0.3356,
      "step": 3575
    },
    {
      "epoch": 1.7045454545454546,
      "grad_norm": 8.200910568237305,
      "learning_rate": 1.1514285714285714e-06,
      "loss": 0.3008,
      "step": 3600
    },
    {
      "epoch": 1.7163825757575757,
      "grad_norm": 12.48343276977539,
      "learning_rate": 1.08e-06,
      "loss": 0.3621,
      "step": 3625
    },
    {
      "epoch": 1.728219696969697,
      "grad_norm": 9.072750091552734,
      "learning_rate": 1.0085714285714286e-06,
      "loss": 0.2966,
      "step": 3650
    },
    {
      "epoch": 1.7400568181818183,
      "grad_norm": 9.534562110900879,
      "learning_rate": 9.371428571428571e-07,
      "loss": 0.3444,
      "step": 3675
    },
    {
      "epoch": 1.7518939393939394,
      "grad_norm": 8.556685447692871,
      "learning_rate": 8.657142857142858e-07,
      "loss": 0.323,
      "step": 3700
    },
    {
      "epoch": 1.7637310606060606,
      "grad_norm": 9.425592422485352,
      "learning_rate": 7.942857142857144e-07,
      "loss": 0.3433,
      "step": 3725
    },
    {
      "epoch": 1.7755681818181817,
      "grad_norm": 10.603422164916992,
      "learning_rate": 7.228571428571429e-07,
      "loss": 0.3376,
      "step": 3750
    },
    {
      "epoch": 1.787405303030303,
      "grad_norm": 8.562864303588867,
      "learning_rate": 6.514285714285715e-07,
      "loss": 0.3175,
      "step": 3775
    },
    {
      "epoch": 1.7992424242424243,
      "grad_norm": 10.139692306518555,
      "learning_rate": 5.800000000000001e-07,
      "loss": 0.3453,
      "step": 3800
    },
    {
      "epoch": 1.8110795454545454,
      "grad_norm": 10.980659484863281,
      "learning_rate": 5.085714285714286e-07,
      "loss": 0.3448,
      "step": 3825
    },
    {
      "epoch": 1.8229166666666665,
      "grad_norm": 9.530597686767578,
      "learning_rate": 4.371428571428572e-07,
      "loss": 0.3303,
      "step": 3850
    },
    {
      "epoch": 1.8347537878787878,
      "grad_norm": 7.428104877471924,
      "learning_rate": 3.657142857142858e-07,
      "loss": 0.3111,
      "step": 3875
    },
    {
      "epoch": 1.8465909090909092,
      "grad_norm": 9.596365928649902,
      "learning_rate": 2.942857142857143e-07,
      "loss": 0.3421,
      "step": 3900
    },
    {
      "epoch": 1.8584280303030303,
      "grad_norm": 10.390746116638184,
      "learning_rate": 2.228571428571429e-07,
      "loss": 0.3443,
      "step": 3925
    },
    {
      "epoch": 1.8702651515151514,
      "grad_norm": 10.99935245513916,
      "learning_rate": 1.5142857142857144e-07,
      "loss": 0.3591,
      "step": 3950
    },
    {
      "epoch": 1.8821022727272727,
      "grad_norm": 11.292437553405762,
      "learning_rate": 8e-08,
      "loss": 0.3492,
      "step": 3975
    },
    {
      "epoch": 1.893939393939394,
      "grad_norm": 10.099928855895996,
      "learning_rate": 8.571428571428572e-09,
      "loss": 0.3441,
      "step": 4000
    },
    {
      "epoch": 1.893939393939394,
      "eval_loss": 0.33238619565963745,
      "eval_runtime": 435.0715,
      "eval_samples_per_second": 3.71,
      "eval_steps_per_second": 0.464,
      "eval_wer": 38.50746268656716,
      "step": 4000
    }
  ],
  "logging_steps": 25,
  "max_steps": 4000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.5752393644032e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}