{
  "best_metric": 67.00100704934542,
  "best_model_checkpoint": "./whisper-small-finetune_egyptian/checkpoint-3000",
  "epoch": 2.8200789622109417,
  "eval_steps": 1000,
  "global_step": 5000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "grad_norm": 27.050291061401367,
      "learning_rate": 5.000000000000001e-07,
      "loss": 1.6686,
      "step": 25
    },
    {
      "epoch": 0.03,
      "grad_norm": 30.311138153076172,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 1.5534,
      "step": 50
    },
    {
      "epoch": 0.04,
      "grad_norm": 28.575551986694336,
      "learning_rate": 1.5e-06,
      "loss": 1.5919,
      "step": 75
    },
    {
      "epoch": 0.06,
      "grad_norm": 33.246788024902344,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 1.6118,
      "step": 100
    },
    {
      "epoch": 0.07,
      "grad_norm": 22.33919334411621,
      "learning_rate": 2.5e-06,
      "loss": 1.3478,
      "step": 125
    },
    {
      "epoch": 0.08,
      "grad_norm": 23.980566024780273,
      "learning_rate": 3e-06,
      "loss": 1.4199,
      "step": 150
    },
    {
      "epoch": 0.1,
      "grad_norm": 24.947845458984375,
      "learning_rate": 3.5e-06,
      "loss": 1.3642,
      "step": 175
    },
    {
      "epoch": 0.11,
      "grad_norm": 19.380922317504883,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.3299,
      "step": 200
    },
    {
      "epoch": 0.13,
      "grad_norm": 23.713611602783203,
      "learning_rate": 4.5e-06,
      "loss": 1.4656,
      "step": 225
    },
    {
      "epoch": 0.14,
      "grad_norm": 24.127946853637695,
      "learning_rate": 5e-06,
      "loss": 1.3392,
      "step": 250
    },
    {
      "epoch": 0.16,
      "grad_norm": 19.620635986328125,
      "learning_rate": 5.500000000000001e-06,
      "loss": 1.2386,
      "step": 275
    },
    {
      "epoch": 0.17,
      "grad_norm": 30.834217071533203,
      "learning_rate": 6e-06,
      "loss": 1.2566,
      "step": 300
    },
    {
      "epoch": 0.18,
      "grad_norm": 27.610654830932617,
      "learning_rate": 6.5000000000000004e-06,
      "loss": 1.3781,
      "step": 325
    },
    {
      "epoch": 0.2,
      "grad_norm": 22.86842155456543,
      "learning_rate": 7e-06,
      "loss": 1.2213,
      "step": 350
    },
    {
      "epoch": 0.21,
      "grad_norm": 17.04463768005371,
      "learning_rate": 7.500000000000001e-06,
      "loss": 1.1576,
      "step": 375
    },
    {
      "epoch": 0.23,
      "grad_norm": 21.666824340820312,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.234,
      "step": 400
    },
    {
      "epoch": 0.24,
      "grad_norm": 22.232830047607422,
      "learning_rate": 8.5e-06,
      "loss": 1.1243,
      "step": 425
    },
    {
      "epoch": 0.25,
      "grad_norm": 22.612762451171875,
      "learning_rate": 9e-06,
      "loss": 1.1452,
      "step": 450
    },
    {
      "epoch": 0.27,
      "grad_norm": 15.275314331054688,
      "learning_rate": 9.5e-06,
      "loss": 1.1841,
      "step": 475
    },
    {
      "epoch": 0.28,
      "grad_norm": 15.876676559448242,
      "learning_rate": 1e-05,
      "loss": 1.2719,
      "step": 500
    },
    {
      "epoch": 0.3,
      "grad_norm": 21.2716007232666,
      "learning_rate": 9.944444444444445e-06,
      "loss": 1.2149,
      "step": 525
    },
    {
      "epoch": 0.31,
      "grad_norm": 22.795671463012695,
      "learning_rate": 9.88888888888889e-06,
      "loss": 1.2219,
      "step": 550
    },
    {
      "epoch": 0.32,
      "grad_norm": 19.169240951538086,
      "learning_rate": 9.833333333333333e-06,
      "loss": 1.2376,
      "step": 575
    },
    {
      "epoch": 0.34,
      "grad_norm": 15.571391105651855,
      "learning_rate": 9.777777777777779e-06,
      "loss": 1.1932,
      "step": 600
    },
    {
      "epoch": 0.35,
      "grad_norm": 16.36895751953125,
      "learning_rate": 9.722222222222223e-06,
      "loss": 1.2936,
      "step": 625
    },
    {
      "epoch": 0.37,
      "grad_norm": 21.461872100830078,
      "learning_rate": 9.666666666666667e-06,
      "loss": 1.2132,
      "step": 650
    },
    {
      "epoch": 0.38,
      "grad_norm": 19.754638671875,
      "learning_rate": 9.611111111111112e-06,
      "loss": 1.182,
      "step": 675
    },
    {
      "epoch": 0.39,
      "grad_norm": 15.755151748657227,
      "learning_rate": 9.555555555555556e-06,
      "loss": 1.1941,
      "step": 700
    },
    {
      "epoch": 0.41,
      "grad_norm": 22.760698318481445,
      "learning_rate": 9.5e-06,
      "loss": 1.2489,
      "step": 725
    },
    {
      "epoch": 0.42,
      "grad_norm": 18.82946014404297,
      "learning_rate": 9.444444444444445e-06,
      "loss": 1.1129,
      "step": 750
    },
    {
      "epoch": 0.44,
      "grad_norm": 18.776079177856445,
      "learning_rate": 9.38888888888889e-06,
      "loss": 1.1889,
      "step": 775
    },
    {
      "epoch": 0.45,
      "grad_norm": 19.27667999267578,
      "learning_rate": 9.333333333333334e-06,
      "loss": 1.1342,
      "step": 800
    },
    {
      "epoch": 0.47,
      "grad_norm": 17.593551635742188,
      "learning_rate": 9.277777777777778e-06,
      "loss": 1.2009,
      "step": 825
    },
    {
      "epoch": 0.48,
      "grad_norm": 19.147933959960938,
      "learning_rate": 9.222222222222224e-06,
      "loss": 1.1845,
      "step": 850
    },
    {
      "epoch": 0.49,
      "grad_norm": 22.792325973510742,
      "learning_rate": 9.166666666666666e-06,
      "loss": 1.2317,
      "step": 875
    },
    {
      "epoch": 0.51,
      "grad_norm": 18.69807243347168,
      "learning_rate": 9.111111111111112e-06,
      "loss": 1.0044,
      "step": 900
    },
    {
      "epoch": 0.52,
      "grad_norm": 28.471355438232422,
      "learning_rate": 9.055555555555556e-06,
      "loss": 1.3067,
      "step": 925
    },
    {
      "epoch": 0.54,
      "grad_norm": 18.05466079711914,
      "learning_rate": 9e-06,
      "loss": 1.1757,
      "step": 950
    },
    {
      "epoch": 0.55,
      "grad_norm": 21.076562881469727,
      "learning_rate": 8.944444444444446e-06,
      "loss": 1.1317,
      "step": 975
    },
    {
      "epoch": 0.56,
      "grad_norm": 19.519039154052734,
      "learning_rate": 8.888888888888888e-06,
      "loss": 1.1256,
      "step": 1000
    },
    {
      "epoch": 0.56,
      "eval_cer": 42.770119634502464,
      "eval_loss": 1.1109758615493774,
      "eval_runtime": 907.6046,
      "eval_samples_per_second": 3.907,
      "eval_steps_per_second": 0.489,
      "eval_wer": 71.8147029204431,
      "step": 1000
    },
    {
      "epoch": 0.58,
      "grad_norm": 15.736879348754883,
      "learning_rate": 8.833333333333334e-06,
      "loss": 1.0697,
      "step": 1025
    },
    {
      "epoch": 0.59,
      "grad_norm": 17.32091522216797,
      "learning_rate": 8.777777777777778e-06,
      "loss": 1.0977,
      "step": 1050
    },
    {
      "epoch": 0.61,
      "grad_norm": 19.34945297241211,
      "learning_rate": 8.722222222222224e-06,
      "loss": 1.1967,
      "step": 1075
    },
    {
      "epoch": 0.62,
      "grad_norm": 20.485715866088867,
      "learning_rate": 8.666666666666668e-06,
      "loss": 1.1102,
      "step": 1100
    },
    {
      "epoch": 0.63,
      "grad_norm": 23.4635009765625,
      "learning_rate": 8.611111111111112e-06,
      "loss": 1.1413,
      "step": 1125
    },
    {
      "epoch": 0.65,
      "grad_norm": 19.843730926513672,
      "learning_rate": 8.555555555555556e-06,
      "loss": 1.157,
      "step": 1150
    },
    {
      "epoch": 0.66,
      "grad_norm": 21.161502838134766,
      "learning_rate": 8.5e-06,
      "loss": 1.0979,
      "step": 1175
    },
    {
      "epoch": 0.68,
      "grad_norm": 20.150407791137695,
      "learning_rate": 8.444444444444446e-06,
      "loss": 1.2714,
      "step": 1200
    },
    {
      "epoch": 0.69,
      "grad_norm": 17.049306869506836,
      "learning_rate": 8.38888888888889e-06,
      "loss": 1.0982,
      "step": 1225
    },
    {
      "epoch": 0.71,
      "grad_norm": 19.597333908081055,
      "learning_rate": 8.333333333333334e-06,
      "loss": 1.0803,
      "step": 1250
    },
    {
      "epoch": 0.72,
      "grad_norm": 17.46756935119629,
      "learning_rate": 8.277777777777778e-06,
      "loss": 1.0542,
      "step": 1275
    },
    {
      "epoch": 0.73,
      "grad_norm": 16.63570213317871,
      "learning_rate": 8.222222222222222e-06,
      "loss": 1.0432,
      "step": 1300
    },
    {
      "epoch": 0.75,
      "grad_norm": 16.54294204711914,
      "learning_rate": 8.166666666666668e-06,
      "loss": 0.9944,
      "step": 1325
    },
    {
      "epoch": 0.76,
      "grad_norm": 10.231407165527344,
      "learning_rate": 8.111111111111112e-06,
      "loss": 1.1438,
      "step": 1350
    },
    {
      "epoch": 0.78,
      "grad_norm": 17.762981414794922,
      "learning_rate": 8.055555555555557e-06,
      "loss": 1.1539,
      "step": 1375
    },
    {
      "epoch": 0.79,
      "grad_norm": 13.66717529296875,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.0529,
      "step": 1400
    },
    {
      "epoch": 0.8,
      "grad_norm": 17.681964874267578,
      "learning_rate": 7.944444444444445e-06,
      "loss": 1.158,
      "step": 1425
    },
    {
      "epoch": 0.82,
      "grad_norm": 17.317996978759766,
      "learning_rate": 7.88888888888889e-06,
      "loss": 1.0955,
      "step": 1450
    },
    {
      "epoch": 0.83,
      "grad_norm": 15.947030067443848,
      "learning_rate": 7.833333333333333e-06,
      "loss": 1.1302,
      "step": 1475
    },
    {
      "epoch": 0.85,
      "grad_norm": 17.65252113342285,
      "learning_rate": 7.77777777777778e-06,
      "loss": 1.0589,
      "step": 1500
    },
    {
      "epoch": 0.86,
      "grad_norm": 17.65298843383789,
      "learning_rate": 7.722222222222223e-06,
      "loss": 1.1063,
      "step": 1525
    },
    {
      "epoch": 0.87,
      "grad_norm": 14.979933738708496,
      "learning_rate": 7.666666666666667e-06,
      "loss": 1.1141,
      "step": 1550
    },
    {
      "epoch": 0.89,
      "grad_norm": 16.299774169921875,
      "learning_rate": 7.611111111111111e-06,
      "loss": 1.0243,
      "step": 1575
    },
    {
      "epoch": 0.9,
      "grad_norm": 23.286108016967773,
      "learning_rate": 7.555555555555556e-06,
      "loss": 1.0325,
      "step": 1600
    },
    {
      "epoch": 0.92,
      "grad_norm": 20.537403106689453,
      "learning_rate": 7.500000000000001e-06,
      "loss": 1.0279,
      "step": 1625
    },
    {
      "epoch": 0.93,
      "grad_norm": 19.857568740844727,
      "learning_rate": 7.444444444444445e-06,
      "loss": 1.0283,
      "step": 1650
    },
    {
      "epoch": 0.94,
      "grad_norm": 18.773427963256836,
      "learning_rate": 7.38888888888889e-06,
      "loss": 1.1405,
      "step": 1675
    },
    {
      "epoch": 0.96,
      "grad_norm": 17.24260139465332,
      "learning_rate": 7.333333333333333e-06,
      "loss": 1.0614,
      "step": 1700
    },
    {
      "epoch": 0.97,
      "grad_norm": 21.773202896118164,
      "learning_rate": 7.277777777777778e-06,
      "loss": 1.0629,
      "step": 1725
    },
    {
      "epoch": 0.99,
      "grad_norm": 19.747053146362305,
      "learning_rate": 7.222222222222223e-06,
      "loss": 1.0446,
      "step": 1750
    },
    {
      "epoch": 1.0,
      "grad_norm": 11.02785587310791,
      "learning_rate": 7.166666666666667e-06,
      "loss": 1.0765,
      "step": 1775
    },
    {
      "epoch": 1.02,
      "grad_norm": 11.050880432128906,
      "learning_rate": 7.111111111111112e-06,
      "loss": 0.9379,
      "step": 1800
    },
    {
      "epoch": 1.03,
      "grad_norm": 16.35251808166504,
      "learning_rate": 7.055555555555557e-06,
      "loss": 0.9255,
      "step": 1825
    },
    {
      "epoch": 1.04,
      "grad_norm": 14.105949401855469,
      "learning_rate": 7e-06,
      "loss": 0.7552,
      "step": 1850
    },
    {
      "epoch": 1.06,
      "grad_norm": 12.735761642456055,
      "learning_rate": 6.944444444444445e-06,
      "loss": 0.8187,
      "step": 1875
    },
    {
      "epoch": 1.07,
      "grad_norm": 15.418607711791992,
      "learning_rate": 6.88888888888889e-06,
      "loss": 0.8395,
      "step": 1900
    },
    {
      "epoch": 1.09,
      "grad_norm": 14.72042179107666,
      "learning_rate": 6.833333333333334e-06,
      "loss": 0.8058,
      "step": 1925
    },
    {
      "epoch": 1.1,
      "grad_norm": 14.438790321350098,
      "learning_rate": 6.777777777777779e-06,
      "loss": 0.7582,
      "step": 1950
    },
    {
      "epoch": 1.11,
      "grad_norm": 17.728883743286133,
      "learning_rate": 6.7222222222222235e-06,
      "loss": 0.7984,
      "step": 1975
    },
    {
      "epoch": 1.13,
      "grad_norm": 19.346345901489258,
      "learning_rate": 6.666666666666667e-06,
      "loss": 0.7331,
      "step": 2000
    },
    {
      "epoch": 1.13,
      "eval_cer": 51.86124281721983,
      "eval_loss": 1.032954454421997,
      "eval_runtime": 959.1561,
      "eval_samples_per_second": 3.697,
      "eval_steps_per_second": 0.463,
      "eval_wer": 82.26384692849949,
      "step": 2000
    },
    {
      "epoch": 1.14,
      "grad_norm": 14.71066665649414,
      "learning_rate": 6.6111111111111115e-06,
      "loss": 0.7684,
      "step": 2025
    },
    {
      "epoch": 1.16,
      "grad_norm": 13.412025451660156,
      "learning_rate": 6.555555555555556e-06,
      "loss": 0.7742,
      "step": 2050
    },
    {
      "epoch": 1.17,
      "grad_norm": 17.332744598388672,
      "learning_rate": 6.5000000000000004e-06,
      "loss": 0.8515,
      "step": 2075
    },
    {
      "epoch": 1.18,
      "grad_norm": 14.944128036499023,
      "learning_rate": 6.444444444444445e-06,
      "loss": 0.8012,
      "step": 2100
    },
    {
      "epoch": 1.2,
      "grad_norm": 18.193222045898438,
      "learning_rate": 6.3888888888888885e-06,
      "loss": 0.8217,
      "step": 2125
    },
    {
      "epoch": 1.21,
      "grad_norm": 18.6146240234375,
      "learning_rate": 6.333333333333333e-06,
      "loss": 0.8732,
      "step": 2150
    },
    {
      "epoch": 1.23,
      "grad_norm": 17.37205696105957,
      "learning_rate": 6.277777777777778e-06,
      "loss": 0.8824,
      "step": 2175
    },
    {
      "epoch": 1.24,
      "grad_norm": 18.284181594848633,
      "learning_rate": 6.222222222222223e-06,
      "loss": 0.7562,
      "step": 2200
    },
    {
      "epoch": 1.25,
      "grad_norm": 13.632092475891113,
      "learning_rate": 6.166666666666667e-06,
      "loss": 0.7502,
      "step": 2225
    },
    {
      "epoch": 1.27,
      "grad_norm": 16.052021026611328,
      "learning_rate": 6.111111111111112e-06,
      "loss": 0.8167,
      "step": 2250
    },
    {
      "epoch": 1.28,
      "grad_norm": 17.9954833984375,
      "learning_rate": 6.055555555555555e-06,
      "loss": 0.7396,
      "step": 2275
    },
    {
      "epoch": 1.3,
      "grad_norm": 19.523056030273438,
      "learning_rate": 6e-06,
      "loss": 0.6644,
      "step": 2300
    },
    {
      "epoch": 1.31,
      "grad_norm": 15.319243431091309,
      "learning_rate": 5.944444444444445e-06,
      "loss": 0.7699,
      "step": 2325
    },
    {
      "epoch": 1.33,
      "grad_norm": 13.561873435974121,
      "learning_rate": 5.88888888888889e-06,
      "loss": 0.7416,
      "step": 2350
    },
    {
      "epoch": 1.34,
      "grad_norm": 13.100536346435547,
      "learning_rate": 5.833333333333334e-06,
      "loss": 0.8497,
      "step": 2375
    },
    {
      "epoch": 1.35,
      "grad_norm": 15.060547828674316,
      "learning_rate": 5.777777777777778e-06,
      "loss": 0.772,
      "step": 2400
    },
    {
      "epoch": 1.37,
      "grad_norm": 15.338216781616211,
      "learning_rate": 5.722222222222222e-06,
      "loss": 0.7752,
      "step": 2425
    },
    {
      "epoch": 1.38,
      "grad_norm": 12.978151321411133,
      "learning_rate": 5.666666666666667e-06,
      "loss": 0.8158,
      "step": 2450
    },
    {
      "epoch": 1.4,
      "grad_norm": 16.481496810913086,
      "learning_rate": 5.611111111111112e-06,
      "loss": 0.7345,
      "step": 2475
    },
    {
      "epoch": 1.41,
      "grad_norm": 15.723489761352539,
      "learning_rate": 5.555555555555557e-06,
      "loss": 0.8427,
      "step": 2500
    },
    {
      "epoch": 1.42,
      "grad_norm": 15.024873733520508,
      "learning_rate": 5.500000000000001e-06,
      "loss": 0.6852,
      "step": 2525
    },
    {
      "epoch": 1.44,
      "grad_norm": 14.445104598999023,
      "learning_rate": 5.444444444444445e-06,
      "loss": 0.8407,
      "step": 2550
    },
    {
      "epoch": 1.45,
      "grad_norm": 18.344518661499023,
      "learning_rate": 5.388888888888889e-06,
      "loss": 0.7858,
      "step": 2575
    },
    {
      "epoch": 1.47,
      "grad_norm": 15.053083419799805,
      "learning_rate": 5.333333333333334e-06,
      "loss": 0.8178,
      "step": 2600
    },
    {
      "epoch": 1.48,
      "grad_norm": 15.452193260192871,
      "learning_rate": 5.2777777777777785e-06,
      "loss": 0.6985,
      "step": 2625
    },
    {
      "epoch": 1.49,
      "grad_norm": 20.083093643188477,
      "learning_rate": 5.2222222222222226e-06,
      "loss": 0.7759,
      "step": 2650
    },
    {
      "epoch": 1.51,
      "grad_norm": 17.425199508666992,
      "learning_rate": 5.1666666666666675e-06,
      "loss": 0.7553,
      "step": 2675
    },
    {
      "epoch": 1.52,
      "grad_norm": 11.294227600097656,
      "learning_rate": 5.1111111111111115e-06,
      "loss": 0.7384,
      "step": 2700
    },
    {
      "epoch": 1.54,
      "grad_norm": 11.263272285461426,
      "learning_rate": 5.0555555555555555e-06,
      "loss": 0.7835,
      "step": 2725
    },
    {
      "epoch": 1.55,
      "grad_norm": 11.892974853515625,
      "learning_rate": 5e-06,
      "loss": 0.7622,
      "step": 2750
    },
    {
      "epoch": 1.57,
      "grad_norm": 11.682631492614746,
      "learning_rate": 4.944444444444445e-06,
      "loss": 0.6707,
      "step": 2775
    },
    {
      "epoch": 1.58,
      "grad_norm": 13.932768821716309,
      "learning_rate": 4.888888888888889e-06,
      "loss": 0.7527,
      "step": 2800
    },
    {
      "epoch": 1.59,
      "grad_norm": 14.168807983398438,
      "learning_rate": 4.833333333333333e-06,
      "loss": 0.7444,
      "step": 2825
    },
    {
      "epoch": 1.61,
      "grad_norm": 8.4957857131958,
      "learning_rate": 4.777777777777778e-06,
      "loss": 0.8083,
      "step": 2850
    },
    {
      "epoch": 1.62,
      "grad_norm": 17.012508392333984,
      "learning_rate": 4.722222222222222e-06,
      "loss": 0.8356,
      "step": 2875
    },
    {
      "epoch": 1.64,
      "grad_norm": 14.761796951293945,
      "learning_rate": 4.666666666666667e-06,
      "loss": 0.8432,
      "step": 2900
    },
    {
      "epoch": 1.65,
      "grad_norm": 16.764440536499023,
      "learning_rate": 4.611111111111112e-06,
      "loss": 0.8869,
      "step": 2925
    },
    {
      "epoch": 1.66,
      "grad_norm": 12.401867866516113,
      "learning_rate": 4.555555555555556e-06,
      "loss": 0.7753,
      "step": 2950
    },
    {
      "epoch": 1.68,
      "grad_norm": 10.731365203857422,
      "learning_rate": 4.5e-06,
      "loss": 0.7484,
      "step": 2975
    },
    {
      "epoch": 1.69,
      "grad_norm": 12.973823547363281,
      "learning_rate": 4.444444444444444e-06,
      "loss": 0.7332,
      "step": 3000
    },
    {
      "epoch": 1.69,
      "eval_cer": 39.18814959022828,
      "eval_loss": 0.994914174079895,
      "eval_runtime": 889.7059,
      "eval_samples_per_second": 3.986,
      "eval_steps_per_second": 0.499,
      "eval_wer": 67.00100704934542,
      "step": 3000
    },
    {
      "epoch": 1.71,
      "grad_norm": 12.647448539733887,
      "learning_rate": 4.388888888888889e-06,
      "loss": 0.7481,
      "step": 3025
    },
    {
      "epoch": 1.72,
      "grad_norm": 16.329065322875977,
      "learning_rate": 4.333333333333334e-06,
      "loss": 0.7734,
      "step": 3050
    },
    {
      "epoch": 1.73,
      "grad_norm": 13.567760467529297,
      "learning_rate": 4.277777777777778e-06,
      "loss": 0.7122,
      "step": 3075
    },
    {
      "epoch": 1.75,
      "grad_norm": 12.22362232208252,
      "learning_rate": 4.222222222222223e-06,
      "loss": 0.7515,
      "step": 3100
    },
    {
      "epoch": 1.76,
      "grad_norm": 18.0850772857666,
      "learning_rate": 4.166666666666667e-06,
      "loss": 0.6836,
      "step": 3125
    },
    {
      "epoch": 1.78,
      "grad_norm": 24.46388816833496,
      "learning_rate": 4.111111111111111e-06,
      "loss": 0.8578,
      "step": 3150
    },
    {
      "epoch": 1.79,
      "grad_norm": 15.071797370910645,
      "learning_rate": 4.055555555555556e-06,
      "loss": 0.7447,
      "step": 3175
    },
    {
      "epoch": 1.8,
      "grad_norm": 12.878135681152344,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.7035,
      "step": 3200
    },
    {
      "epoch": 1.82,
      "grad_norm": 14.950927734375,
      "learning_rate": 3.944444444444445e-06,
      "loss": 0.7605,
      "step": 3225
    },
    {
      "epoch": 1.83,
      "grad_norm": 15.3999605178833,
      "learning_rate": 3.88888888888889e-06,
      "loss": 0.7413,
      "step": 3250
    },
    {
      "epoch": 1.85,
      "grad_norm": 14.097206115722656,
      "learning_rate": 3.833333333333334e-06,
      "loss": 0.7473,
      "step": 3275
    },
    {
      "epoch": 1.86,
      "grad_norm": 15.387040138244629,
      "learning_rate": 3.777777777777778e-06,
      "loss": 0.716,
      "step": 3300
    },
    {
      "epoch": 1.88,
      "grad_norm": 20.502208709716797,
      "learning_rate": 3.7222222222222225e-06,
      "loss": 0.7577,
      "step": 3325
    },
    {
      "epoch": 1.89,
      "grad_norm": 16.519433975219727,
      "learning_rate": 3.6666666666666666e-06,
      "loss": 0.6927,
      "step": 3350
    },
    {
      "epoch": 1.9,
      "grad_norm": 15.588461875915527,
      "learning_rate": 3.6111111111111115e-06,
      "loss": 0.7123,
      "step": 3375
    },
    {
      "epoch": 1.92,
      "grad_norm": 12.752546310424805,
      "learning_rate": 3.555555555555556e-06,
      "loss": 0.8407,
      "step": 3400
    },
    {
      "epoch": 1.93,
      "grad_norm": 15.805379867553711,
      "learning_rate": 3.5e-06,
      "loss": 0.751,
      "step": 3425
    },
    {
      "epoch": 1.95,
      "grad_norm": 13.27651596069336,
      "learning_rate": 3.444444444444445e-06,
      "loss": 0.8211,
      "step": 3450
    },
    {
      "epoch": 1.96,
      "grad_norm": 11.968923568725586,
      "learning_rate": 3.3888888888888893e-06,
      "loss": 0.6975,
      "step": 3475
    },
    {
      "epoch": 1.97,
      "grad_norm": 13.906292915344238,
      "learning_rate": 3.3333333333333333e-06,
      "loss": 0.8429,
      "step": 3500
    },
    {
      "epoch": 1.99,
      "grad_norm": 13.810843467712402,
      "learning_rate": 3.277777777777778e-06,
      "loss": 0.7043,
      "step": 3525
    },
    {
      "epoch": 2.0,
      "grad_norm": 14.037764549255371,
      "learning_rate": 3.2222222222222227e-06,
      "loss": 0.6782,
      "step": 3550
    },
    {
      "epoch": 2.02,
      "grad_norm": 8.811666488647461,
      "learning_rate": 3.1666666666666667e-06,
      "loss": 0.5169,
      "step": 3575
    },
    {
      "epoch": 2.03,
      "grad_norm": 12.727725982666016,
      "learning_rate": 3.1111111111111116e-06,
      "loss": 0.5087,
      "step": 3600
    },
    {
      "epoch": 2.04,
      "grad_norm": 13.598219871520996,
      "learning_rate": 3.055555555555556e-06,
      "loss": 0.6104,
      "step": 3625
    },
    {
      "epoch": 2.06,
      "grad_norm": 12.1933012008667,
      "learning_rate": 3e-06,
      "loss": 0.5486,
      "step": 3650
    },
    {
      "epoch": 2.07,
      "grad_norm": 7.250199317932129,
      "learning_rate": 2.944444444444445e-06,
      "loss": 0.473,
      "step": 3675
    },
    {
      "epoch": 2.09,
      "grad_norm": 11.98234748840332,
      "learning_rate": 2.888888888888889e-06,
      "loss": 0.5483,
      "step": 3700
    },
    {
      "epoch": 2.1,
      "grad_norm": 15.183174133300781,
      "learning_rate": 2.8333333333333335e-06,
      "loss": 0.5781,
      "step": 3725
    },
    {
      "epoch": 2.12,
      "grad_norm": 14.038616180419922,
      "learning_rate": 2.7777777777777783e-06,
      "loss": 0.5383,
      "step": 3750
    },
    {
      "epoch": 2.13,
      "grad_norm": 11.775838851928711,
      "learning_rate": 2.7222222222222224e-06,
      "loss": 0.6216,
      "step": 3775
    },
    {
      "epoch": 2.14,
      "grad_norm": 11.929536819458008,
      "learning_rate": 2.666666666666667e-06,
      "loss": 0.4929,
      "step": 3800
    },
    {
      "epoch": 2.16,
      "grad_norm": 13.312314987182617,
      "learning_rate": 2.6111111111111113e-06,
      "loss": 0.4658,
      "step": 3825
    },
    {
      "epoch": 2.17,
      "grad_norm": 15.442246437072754,
      "learning_rate": 2.5555555555555557e-06,
      "loss": 0.498,
      "step": 3850
    },
    {
      "epoch": 2.19,
      "grad_norm": 10.368426322937012,
      "learning_rate": 2.5e-06,
      "loss": 0.5588,
      "step": 3875
    },
    {
      "epoch": 2.2,
      "grad_norm": 9.702683448791504,
      "learning_rate": 2.4444444444444447e-06,
      "loss": 0.5221,
      "step": 3900
    },
    {
      "epoch": 2.21,
      "grad_norm": 12.07742977142334,
      "learning_rate": 2.388888888888889e-06,
      "loss": 0.5033,
      "step": 3925
    },
    {
      "epoch": 2.23,
      "grad_norm": 15.718260765075684,
      "learning_rate": 2.3333333333333336e-06,
      "loss": 0.5913,
      "step": 3950
    },
    {
      "epoch": 2.24,
      "grad_norm": 10.115168571472168,
      "learning_rate": 2.277777777777778e-06,
      "loss": 0.5117,
      "step": 3975
    },
    {
      "epoch": 2.26,
      "grad_norm": 9.728811264038086,
      "learning_rate": 2.222222222222222e-06,
      "loss": 0.5493,
      "step": 4000
    },
    {
      "epoch": 2.26,
      "eval_cer": 41.269978333909,
      "eval_loss": 1.0036817789077759,
      "eval_runtime": 911.3291,
      "eval_samples_per_second": 3.891,
      "eval_steps_per_second": 0.487,
      "eval_wer": 68.10473313192347,
      "step": 4000
    },
    {
      "epoch": 2.27,
      "grad_norm": 12.741899490356445,
      "learning_rate": 2.166666666666667e-06,
      "loss": 0.4817,
      "step": 4025
    },
    {
      "epoch": 2.28,
      "grad_norm": 14.738693237304688,
      "learning_rate": 2.1111111111111114e-06,
      "loss": 0.5172,
      "step": 4050
    },
    {
      "epoch": 2.3,
      "grad_norm": 10.202011108398438,
      "learning_rate": 2.0555555555555555e-06,
      "loss": 0.4856,
      "step": 4075
    },
    {
      "epoch": 2.31,
      "grad_norm": 11.9408597946167,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.527,
      "step": 4100
    },
    {
      "epoch": 2.33,
      "grad_norm": 8.329051971435547,
      "learning_rate": 1.944444444444445e-06,
      "loss": 0.432,
      "step": 4125
    },
    {
      "epoch": 2.34,
      "grad_norm": 12.512899398803711,
      "learning_rate": 1.888888888888889e-06,
      "loss": 0.4836,
      "step": 4150
    },
    {
      "epoch": 2.35,
      "grad_norm": 8.79201889038086,
      "learning_rate": 1.8333333333333333e-06,
      "loss": 0.52,
      "step": 4175
    },
    {
      "epoch": 2.37,
      "grad_norm": 15.417551040649414,
      "learning_rate": 1.777777777777778e-06,
      "loss": 0.5239,
      "step": 4200
    },
    {
      "epoch": 2.38,
      "grad_norm": 10.516593933105469,
      "learning_rate": 1.7222222222222224e-06,
      "loss": 0.5497,
      "step": 4225
    },
    {
      "epoch": 2.4,
      "grad_norm": 11.659682273864746,
      "learning_rate": 1.6666666666666667e-06,
      "loss": 0.4883,
      "step": 4250
    },
    {
      "epoch": 2.41,
      "grad_norm": 15.743680000305176,
      "learning_rate": 1.6111111111111113e-06,
      "loss": 0.5215,
      "step": 4275
    },
    {
      "epoch": 2.43,
      "grad_norm": 15.303753852844238,
      "learning_rate": 1.5555555555555558e-06,
      "loss": 0.5714,
      "step": 4300
    },
    {
      "epoch": 2.44,
      "grad_norm": 13.083234786987305,
      "learning_rate": 1.5e-06,
      "loss": 0.479,
      "step": 4325
    },
    {
      "epoch": 2.45,
      "grad_norm": 7.123412132263184,
      "learning_rate": 1.4444444444444445e-06,
      "loss": 0.4791,
      "step": 4350
    },
    {
      "epoch": 2.47,
      "grad_norm": 14.159947395324707,
      "learning_rate": 1.3888888888888892e-06,
      "loss": 0.6012,
      "step": 4375
    },
    {
      "epoch": 2.48,
      "grad_norm": 16.015972137451172,
      "learning_rate": 1.3333333333333334e-06,
      "loss": 0.4877,
      "step": 4400
    },
    {
      "epoch": 2.5,
      "grad_norm": 14.357626914978027,
      "learning_rate": 1.2777777777777779e-06,
      "loss": 0.5159,
      "step": 4425
    },
    {
      "epoch": 2.51,
      "grad_norm": 7.9363274574279785,
      "learning_rate": 1.2222222222222223e-06,
      "loss": 0.523,
      "step": 4450
    },
    {
      "epoch": 2.52,
      "grad_norm": 15.333666801452637,
      "learning_rate": 1.1666666666666668e-06,
      "loss": 0.5083,
      "step": 4475
    },
    {
      "epoch": 2.54,
      "grad_norm": 19.310766220092773,
      "learning_rate": 1.111111111111111e-06,
      "loss": 0.5721,
      "step": 4500
    },
    {
      "epoch": 2.55,
      "grad_norm": 16.54596710205078,
      "learning_rate": 1.0555555555555557e-06,
      "loss": 0.6224,
      "step": 4525
    },
    {
      "epoch": 2.57,
      "grad_norm": 10.166070938110352,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.5366,
      "step": 4550
    },
    {
      "epoch": 2.58,
      "grad_norm": 12.599578857421875,
      "learning_rate": 9.444444444444445e-07,
      "loss": 0.4959,
      "step": 4575
    },
    {
      "epoch": 2.59,
      "grad_norm": 7.448824405670166,
      "learning_rate": 8.88888888888889e-07,
      "loss": 0.4986,
      "step": 4600
    },
    {
      "epoch": 2.61,
      "grad_norm": 11.634490013122559,
      "learning_rate": 8.333333333333333e-07,
      "loss": 0.4859,
      "step": 4625
    },
    {
      "epoch": 2.62,
      "grad_norm": 10.882704734802246,
      "learning_rate": 7.777777777777779e-07,
      "loss": 0.5328,
      "step": 4650
    },
    {
      "epoch": 2.64,
      "grad_norm": 9.251214027404785,
      "learning_rate": 7.222222222222222e-07,
      "loss": 0.5354,
      "step": 4675
    },
    {
      "epoch": 2.65,
      "grad_norm": 15.835794448852539,
      "learning_rate": 6.666666666666667e-07,
      "loss": 0.5906,
      "step": 4700
    },
    {
      "epoch": 2.66,
      "grad_norm": 11.154616355895996,
      "learning_rate": 6.111111111111112e-07,
      "loss": 0.5786,
      "step": 4725
    },
    {
      "epoch": 2.68,
      "grad_norm": 13.227851867675781,
      "learning_rate": 5.555555555555555e-07,
      "loss": 0.4945,
      "step": 4750
    },
    {
      "epoch": 2.69,
      "grad_norm": 13.796296119689941,
      "learning_rate": 5.000000000000001e-07,
      "loss": 0.5045,
      "step": 4775
    },
    {
      "epoch": 2.71,
      "grad_norm": 13.530144691467285,
      "learning_rate": 4.444444444444445e-07,
      "loss": 0.5801,
      "step": 4800
    },
    {
      "epoch": 2.72,
      "grad_norm": 7.212181091308594,
      "learning_rate": 3.8888888888888895e-07,
      "loss": 0.4827,
      "step": 4825
    },
    {
      "epoch": 2.74,
      "grad_norm": 15.471014022827148,
      "learning_rate": 3.3333333333333335e-07,
      "loss": 0.5754,
      "step": 4850
    },
    {
      "epoch": 2.75,
      "grad_norm": 14.308059692382812,
      "learning_rate": 2.7777777777777776e-07,
      "loss": 0.5391,
      "step": 4875
    },
    {
      "epoch": 2.76,
      "grad_norm": 10.918014526367188,
      "learning_rate": 2.2222222222222224e-07,
      "loss": 0.6082,
      "step": 4900
    },
    {
      "epoch": 2.78,
      "grad_norm": 7.789730072021484,
      "learning_rate": 1.6666666666666668e-07,
      "loss": 0.4707,
      "step": 4925
    },
    {
      "epoch": 2.79,
      "grad_norm": 12.3082275390625,
      "learning_rate": 1.1111111111111112e-07,
      "loss": 0.552,
      "step": 4950
    },
    {
      "epoch": 2.81,
      "grad_norm": 15.524524688720703,
      "learning_rate": 5.555555555555556e-08,
      "loss": 0.4998,
      "step": 4975
    },
    {
      "epoch": 2.82,
      "grad_norm": 11.517424583435059,
      "learning_rate": 0.0,
      "loss": 0.4667,
      "step": 5000
    },
    {
      "epoch": 2.82,
      "eval_cer": 42.82114484880836,
      "eval_loss": 0.9976443648338318,
      "eval_runtime": 928.1561,
      "eval_samples_per_second": 3.82,
      "eval_steps_per_second": 0.478,
      "eval_wer": 70.0221550855992,
      "step": 5000
    }
  ],
  "logging_steps": 25,
  "max_steps": 5000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 1000,
  "total_flos": 1.154110732959744e+19,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}