{
  "best_metric": 68.70090634441087,
  "best_model_checkpoint": "./whisper-small-dialect_egyptian/checkpoint-5000",
  "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": 122.04756164550781,
      "learning_rate": 5.000000000000001e-07,
      "loss": 4.3768,
      "step": 25
    },
    {
      "epoch": 0.03,
      "grad_norm": 40.10433578491211,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 3.742,
      "step": 50
    },
    {
      "epoch": 0.04,
      "grad_norm": 35.64628982543945,
      "learning_rate": 1.5e-06,
      "loss": 3.1278,
      "step": 75
    },
    {
      "epoch": 0.06,
      "grad_norm": 36.580284118652344,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 2.4904,
      "step": 100
    },
    {
      "epoch": 0.07,
      "grad_norm": 23.064411163330078,
      "learning_rate": 2.5e-06,
      "loss": 1.9741,
      "step": 125
    },
    {
      "epoch": 0.08,
      "grad_norm": 25.443105697631836,
      "learning_rate": 3e-06,
      "loss": 1.993,
      "step": 150
    },
    {
      "epoch": 0.1,
      "grad_norm": 25.980615615844727,
      "learning_rate": 3.5e-06,
      "loss": 1.8217,
      "step": 175
    },
    {
      "epoch": 0.11,
      "grad_norm": 20.749223709106445,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.7756,
      "step": 200
    },
    {
      "epoch": 0.13,
      "grad_norm": 26.173999786376953,
      "learning_rate": 4.5e-06,
      "loss": 1.8824,
      "step": 225
    },
    {
      "epoch": 0.14,
      "grad_norm": 26.512161254882812,
      "learning_rate": 5e-06,
      "loss": 1.7371,
      "step": 250
    },
    {
      "epoch": 0.16,
      "grad_norm": 23.805395126342773,
      "learning_rate": 5.500000000000001e-06,
      "loss": 1.5593,
      "step": 275
    },
    {
      "epoch": 0.17,
      "grad_norm": 33.373779296875,
      "learning_rate": 6e-06,
      "loss": 1.4868,
      "step": 300
    },
    {
      "epoch": 0.18,
      "grad_norm": 27.45977783203125,
      "learning_rate": 6.5000000000000004e-06,
      "loss": 1.5585,
      "step": 325
    },
    {
      "epoch": 0.2,
      "grad_norm": 24.94574737548828,
      "learning_rate": 7e-06,
      "loss": 1.3193,
      "step": 350
    },
    {
      "epoch": 0.21,
      "grad_norm": 18.338777542114258,
      "learning_rate": 7.500000000000001e-06,
      "loss": 1.1767,
      "step": 375
    },
    {
      "epoch": 0.23,
      "grad_norm": 26.012649536132812,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.2317,
      "step": 400
    },
    {
      "epoch": 0.24,
      "grad_norm": 23.6628360748291,
      "learning_rate": 8.5e-06,
      "loss": 1.1083,
      "step": 425
    },
    {
      "epoch": 0.25,
      "grad_norm": 24.645048141479492,
      "learning_rate": 9e-06,
      "loss": 1.1444,
      "step": 450
    },
    {
      "epoch": 0.27,
      "grad_norm": 15.89114761352539,
      "learning_rate": 9.5e-06,
      "loss": 1.1598,
      "step": 475
    },
    {
      "epoch": 0.28,
      "grad_norm": 20.403839111328125,
      "learning_rate": 1e-05,
      "loss": 1.28,
      "step": 500
    },
    {
      "epoch": 0.3,
      "grad_norm": 23.043336868286133,
      "learning_rate": 9.944444444444445e-06,
      "loss": 1.2277,
      "step": 525
    },
    {
      "epoch": 0.31,
      "grad_norm": 25.324621200561523,
      "learning_rate": 9.88888888888889e-06,
      "loss": 1.2154,
      "step": 550
    },
    {
      "epoch": 0.32,
      "grad_norm": 21.331308364868164,
      "learning_rate": 9.833333333333333e-06,
      "loss": 1.2405,
      "step": 575
    },
    {
      "epoch": 0.34,
      "grad_norm": 17.898706436157227,
      "learning_rate": 9.777777777777779e-06,
      "loss": 1.1903,
      "step": 600
    },
    {
      "epoch": 0.35,
      "grad_norm": 18.209653854370117,
      "learning_rate": 9.722222222222223e-06,
      "loss": 1.2725,
      "step": 625
    },
    {
      "epoch": 0.37,
      "grad_norm": 24.40468406677246,
      "learning_rate": 9.666666666666667e-06,
      "loss": 1.226,
      "step": 650
    },
    {
      "epoch": 0.38,
      "grad_norm": 22.115428924560547,
      "learning_rate": 9.611111111111112e-06,
      "loss": 1.1932,
      "step": 675
    },
    {
      "epoch": 0.39,
      "grad_norm": 18.02147674560547,
      "learning_rate": 9.555555555555556e-06,
      "loss": 1.1863,
      "step": 700
    },
    {
      "epoch": 0.41,
      "grad_norm": 26.181509017944336,
      "learning_rate": 9.5e-06,
      "loss": 1.2406,
      "step": 725
    },
    {
      "epoch": 0.42,
      "grad_norm": 21.838193893432617,
      "learning_rate": 9.444444444444445e-06,
      "loss": 1.1168,
      "step": 750
    },
    {
      "epoch": 0.44,
      "grad_norm": 21.846904754638672,
      "learning_rate": 9.38888888888889e-06,
      "loss": 1.1986,
      "step": 775
    },
    {
      "epoch": 0.45,
      "grad_norm": 20.488187789916992,
      "learning_rate": 9.333333333333334e-06,
      "loss": 1.1278,
      "step": 800
    },
    {
      "epoch": 0.47,
      "grad_norm": 20.44611167907715,
      "learning_rate": 9.277777777777778e-06,
      "loss": 1.2049,
      "step": 825
    },
    {
      "epoch": 0.48,
      "grad_norm": 22.499937057495117,
      "learning_rate": 9.222222222222224e-06,
      "loss": 1.1704,
      "step": 850
    },
    {
      "epoch": 0.49,
      "grad_norm": 26.488370895385742,
      "learning_rate": 9.166666666666666e-06,
      "loss": 1.2335,
      "step": 875
    },
    {
      "epoch": 0.51,
      "grad_norm": 21.434371948242188,
      "learning_rate": 9.111111111111112e-06,
      "loss": 1.0156,
      "step": 900
    },
    {
      "epoch": 0.52,
      "grad_norm": 34.55289840698242,
      "learning_rate": 9.055555555555556e-06,
      "loss": 1.3202,
      "step": 925
    },
    {
      "epoch": 0.54,
      "grad_norm": 20.672531127929688,
      "learning_rate": 9e-06,
      "loss": 1.1801,
      "step": 950
    },
    {
      "epoch": 0.55,
      "grad_norm": 23.088294982910156,
      "learning_rate": 8.944444444444446e-06,
      "loss": 1.1447,
      "step": 975
    },
    {
      "epoch": 0.56,
      "grad_norm": 20.51502799987793,
      "learning_rate": 8.888888888888888e-06,
      "loss": 1.1284,
      "step": 1000
    },
    {
      "epoch": 0.56,
      "eval_cer": 44.417841554934526,
      "eval_loss": 1.1110167503356934,
      "eval_runtime": 911.3479,
      "eval_samples_per_second": 3.891,
      "eval_steps_per_second": 0.487,
      "eval_wer": 76.0281973816717,
      "step": 1000
    },
    {
      "epoch": 0.58,
      "grad_norm": 17.626625061035156,
      "learning_rate": 8.833333333333334e-06,
      "loss": 1.0659,
      "step": 1025
    },
    {
      "epoch": 0.59,
      "grad_norm": 20.585145950317383,
      "learning_rate": 8.777777777777778e-06,
      "loss": 1.1016,
      "step": 1050
    },
    {
      "epoch": 0.61,
      "grad_norm": 21.167909622192383,
      "learning_rate": 8.722222222222224e-06,
      "loss": 1.205,
      "step": 1075
    },
    {
      "epoch": 0.62,
      "grad_norm": 23.734207153320312,
      "learning_rate": 8.666666666666668e-06,
      "loss": 1.1123,
      "step": 1100
    },
    {
      "epoch": 0.63,
      "grad_norm": 26.98702621459961,
      "learning_rate": 8.611111111111112e-06,
      "loss": 1.1399,
      "step": 1125
    },
    {
      "epoch": 0.65,
      "grad_norm": 22.44954490661621,
      "learning_rate": 8.555555555555556e-06,
      "loss": 1.1662,
      "step": 1150
    },
    {
      "epoch": 0.66,
      "grad_norm": 24.248973846435547,
      "learning_rate": 8.5e-06,
      "loss": 1.1006,
      "step": 1175
    },
    {
      "epoch": 0.68,
      "grad_norm": 21.817211151123047,
      "learning_rate": 8.444444444444446e-06,
      "loss": 1.2818,
      "step": 1200
    },
    {
      "epoch": 0.69,
      "grad_norm": 19.826051712036133,
      "learning_rate": 8.38888888888889e-06,
      "loss": 1.1063,
      "step": 1225
    },
    {
      "epoch": 0.71,
      "grad_norm": 22.592252731323242,
      "learning_rate": 8.333333333333334e-06,
      "loss": 1.0954,
      "step": 1250
    },
    {
      "epoch": 0.72,
      "grad_norm": 19.33829689025879,
      "learning_rate": 8.277777777777778e-06,
      "loss": 1.0621,
      "step": 1275
    },
    {
      "epoch": 0.73,
      "grad_norm": 18.287986755371094,
      "learning_rate": 8.222222222222222e-06,
      "loss": 1.0543,
      "step": 1300
    },
    {
      "epoch": 0.75,
      "grad_norm": 18.611087799072266,
      "learning_rate": 8.166666666666668e-06,
      "loss": 0.9963,
      "step": 1325
    },
    {
      "epoch": 0.76,
      "grad_norm": 10.573747634887695,
      "learning_rate": 8.111111111111112e-06,
      "loss": 1.1583,
      "step": 1350
    },
    {
      "epoch": 0.78,
      "grad_norm": 20.298009872436523,
      "learning_rate": 8.055555555555557e-06,
      "loss": 1.1522,
      "step": 1375
    },
    {
      "epoch": 0.79,
      "grad_norm": 15.819123268127441,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.051,
      "step": 1400
    },
    {
      "epoch": 0.8,
      "grad_norm": 19.158708572387695,
      "learning_rate": 7.944444444444445e-06,
      "loss": 1.159,
      "step": 1425
    },
    {
      "epoch": 0.82,
      "grad_norm": 20.468978881835938,
      "learning_rate": 7.88888888888889e-06,
      "loss": 1.1176,
      "step": 1450
    },
    {
      "epoch": 0.83,
      "grad_norm": 18.97748565673828,
      "learning_rate": 7.833333333333333e-06,
      "loss": 1.1303,
      "step": 1475
    },
    {
      "epoch": 0.85,
      "grad_norm": 20.76470947265625,
      "learning_rate": 7.77777777777778e-06,
      "loss": 1.0628,
      "step": 1500
    },
    {
      "epoch": 0.86,
      "grad_norm": 20.96468162536621,
      "learning_rate": 7.722222222222223e-06,
      "loss": 1.116,
      "step": 1525
    },
    {
      "epoch": 0.87,
      "grad_norm": 16.63492202758789,
      "learning_rate": 7.666666666666667e-06,
      "loss": 1.119,
      "step": 1550
    },
    {
      "epoch": 0.89,
      "grad_norm": 19.015464782714844,
      "learning_rate": 7.611111111111111e-06,
      "loss": 1.0257,
      "step": 1575
    },
    {
      "epoch": 0.9,
      "grad_norm": 26.927936553955078,
      "learning_rate": 7.555555555555556e-06,
      "loss": 1.0344,
      "step": 1600
    },
    {
      "epoch": 0.92,
      "grad_norm": 24.408445358276367,
      "learning_rate": 7.500000000000001e-06,
      "loss": 1.0531,
      "step": 1625
    },
    {
      "epoch": 0.93,
      "grad_norm": 23.072290420532227,
      "learning_rate": 7.444444444444445e-06,
      "loss": 1.035,
      "step": 1650
    },
    {
      "epoch": 0.94,
      "grad_norm": 22.863815307617188,
      "learning_rate": 7.38888888888889e-06,
      "loss": 1.1428,
      "step": 1675
    },
    {
      "epoch": 0.96,
      "grad_norm": 20.69190788269043,
      "learning_rate": 7.333333333333333e-06,
      "loss": 1.0685,
      "step": 1700
    },
    {
      "epoch": 0.97,
      "grad_norm": 22.434356689453125,
      "learning_rate": 7.277777777777778e-06,
      "loss": 1.0722,
      "step": 1725
    },
    {
      "epoch": 0.99,
      "grad_norm": 23.94415283203125,
      "learning_rate": 7.222222222222223e-06,
      "loss": 1.045,
      "step": 1750
    },
    {
      "epoch": 1.0,
      "grad_norm": 13.114087104797363,
      "learning_rate": 7.166666666666667e-06,
      "loss": 1.0877,
      "step": 1775
    },
    {
      "epoch": 1.02,
      "grad_norm": 12.52387523651123,
      "learning_rate": 7.111111111111112e-06,
      "loss": 0.9332,
      "step": 1800
    },
    {
      "epoch": 1.03,
      "grad_norm": 17.861310958862305,
      "learning_rate": 7.055555555555557e-06,
      "loss": 0.9197,
      "step": 1825
    },
    {
      "epoch": 1.04,
      "grad_norm": 15.940573692321777,
      "learning_rate": 7e-06,
      "loss": 0.7525,
      "step": 1850
    },
    {
      "epoch": 1.06,
      "grad_norm": 13.879676818847656,
      "learning_rate": 6.944444444444445e-06,
      "loss": 0.8269,
      "step": 1875
    },
    {
      "epoch": 1.07,
      "grad_norm": 18.230445861816406,
      "learning_rate": 6.88888888888889e-06,
      "loss": 0.8411,
      "step": 1900
    },
    {
      "epoch": 1.09,
      "grad_norm": 15.949620246887207,
      "learning_rate": 6.833333333333334e-06,
      "loss": 0.8183,
      "step": 1925
    },
    {
      "epoch": 1.1,
      "grad_norm": 16.57105827331543,
      "learning_rate": 6.777777777777779e-06,
      "loss": 0.7672,
      "step": 1950
    },
    {
      "epoch": 1.11,
      "grad_norm": 20.67877769470215,
      "learning_rate": 6.7222222222222235e-06,
      "loss": 0.8034,
      "step": 1975
    },
    {
      "epoch": 1.13,
      "grad_norm": 20.689613342285156,
      "learning_rate": 6.666666666666667e-06,
      "loss": 0.7338,
      "step": 2000
    },
    {
      "epoch": 1.13,
      "eval_cer": 45.29547524099601,
      "eval_loss": 1.0379834175109863,
      "eval_runtime": 916.1939,
      "eval_samples_per_second": 3.87,
      "eval_steps_per_second": 0.485,
      "eval_wer": 73.40584088620342,
      "step": 2000
    },
    {
      "epoch": 1.14,
      "grad_norm": 17.04854965209961,
      "learning_rate": 6.6111111111111115e-06,
      "loss": 0.7875,
      "step": 2025
    },
    {
      "epoch": 1.16,
      "grad_norm": 15.2536039352417,
      "learning_rate": 6.555555555555556e-06,
      "loss": 0.7838,
      "step": 2050
    },
    {
      "epoch": 1.17,
      "grad_norm": 19.358837127685547,
      "learning_rate": 6.5000000000000004e-06,
      "loss": 0.8523,
      "step": 2075
    },
    {
      "epoch": 1.18,
      "grad_norm": 16.287504196166992,
      "learning_rate": 6.444444444444445e-06,
      "loss": 0.798,
      "step": 2100
    },
    {
      "epoch": 1.2,
      "grad_norm": 19.646032333374023,
      "learning_rate": 6.3888888888888885e-06,
      "loss": 0.8304,
      "step": 2125
    },
    {
      "epoch": 1.21,
      "grad_norm": 21.761016845703125,
      "learning_rate": 6.333333333333333e-06,
      "loss": 0.8689,
      "step": 2150
    },
    {
      "epoch": 1.23,
      "grad_norm": 20.19859504699707,
      "learning_rate": 6.277777777777778e-06,
      "loss": 0.8861,
      "step": 2175
    },
    {
      "epoch": 1.24,
      "grad_norm": 21.027082443237305,
      "learning_rate": 6.222222222222223e-06,
      "loss": 0.7618,
      "step": 2200
    },
    {
      "epoch": 1.25,
      "grad_norm": 16.476125717163086,
      "learning_rate": 6.166666666666667e-06,
      "loss": 0.7588,
      "step": 2225
    },
    {
      "epoch": 1.27,
      "grad_norm": 19.24110984802246,
      "learning_rate": 6.111111111111112e-06,
      "loss": 0.8258,
      "step": 2250
    },
    {
      "epoch": 1.28,
      "grad_norm": 19.605239868164062,
      "learning_rate": 6.055555555555555e-06,
      "loss": 0.7461,
      "step": 2275
    },
    {
      "epoch": 1.3,
      "grad_norm": 20.91330909729004,
      "learning_rate": 6e-06,
      "loss": 0.6665,
      "step": 2300
    },
    {
      "epoch": 1.31,
      "grad_norm": 16.637998580932617,
      "learning_rate": 5.944444444444445e-06,
      "loss": 0.773,
      "step": 2325
    },
    {
      "epoch": 1.33,
      "grad_norm": 15.381094932556152,
      "learning_rate": 5.88888888888889e-06,
      "loss": 0.7393,
      "step": 2350
    },
    {
      "epoch": 1.34,
      "grad_norm": 15.25363826751709,
      "learning_rate": 5.833333333333334e-06,
      "loss": 0.8684,
      "step": 2375
    },
    {
      "epoch": 1.35,
      "grad_norm": 19.22905921936035,
      "learning_rate": 5.777777777777778e-06,
      "loss": 0.7679,
      "step": 2400
    },
    {
      "epoch": 1.37,
      "grad_norm": 17.750835418701172,
      "learning_rate": 5.722222222222222e-06,
      "loss": 0.7728,
      "step": 2425
    },
    {
      "epoch": 1.38,
      "grad_norm": 14.001325607299805,
      "learning_rate": 5.666666666666667e-06,
      "loss": 0.8133,
      "step": 2450
    },
    {
      "epoch": 1.4,
      "grad_norm": 17.449642181396484,
      "learning_rate": 5.611111111111112e-06,
      "loss": 0.7363,
      "step": 2475
    },
    {
      "epoch": 1.41,
      "grad_norm": 17.825336456298828,
      "learning_rate": 5.555555555555557e-06,
      "loss": 0.8418,
      "step": 2500
    },
    {
      "epoch": 1.42,
      "grad_norm": 17.27322006225586,
      "learning_rate": 5.500000000000001e-06,
      "loss": 0.6961,
      "step": 2525
    },
    {
      "epoch": 1.44,
      "grad_norm": 14.609389305114746,
      "learning_rate": 5.444444444444445e-06,
      "loss": 0.8457,
      "step": 2550
    },
    {
      "epoch": 1.45,
      "grad_norm": 19.48899269104004,
      "learning_rate": 5.388888888888889e-06,
      "loss": 0.7922,
      "step": 2575
    },
    {
      "epoch": 1.47,
      "grad_norm": 16.829326629638672,
      "learning_rate": 5.333333333333334e-06,
      "loss": 0.8122,
      "step": 2600
    },
    {
      "epoch": 1.48,
      "grad_norm": 17.410572052001953,
      "learning_rate": 5.2777777777777785e-06,
      "loss": 0.7086,
      "step": 2625
    },
    {
      "epoch": 1.49,
      "grad_norm": 20.42573356628418,
      "learning_rate": 5.2222222222222226e-06,
      "loss": 0.7816,
      "step": 2650
    },
    {
      "epoch": 1.51,
      "grad_norm": 20.022361755371094,
      "learning_rate": 5.1666666666666675e-06,
      "loss": 0.7642,
      "step": 2675
    },
    {
      "epoch": 1.52,
      "grad_norm": 12.024133682250977,
      "learning_rate": 5.1111111111111115e-06,
      "loss": 0.7466,
      "step": 2700
    },
    {
      "epoch": 1.54,
      "grad_norm": 12.811111450195312,
      "learning_rate": 5.0555555555555555e-06,
      "loss": 0.786,
      "step": 2725
    },
    {
      "epoch": 1.55,
      "grad_norm": 12.076204299926758,
      "learning_rate": 5e-06,
      "loss": 0.7625,
      "step": 2750
    },
    {
      "epoch": 1.57,
      "grad_norm": 12.905007362365723,
      "learning_rate": 4.944444444444445e-06,
      "loss": 0.6815,
      "step": 2775
    },
    {
      "epoch": 1.58,
      "grad_norm": 15.16416072845459,
      "learning_rate": 4.888888888888889e-06,
      "loss": 0.7714,
      "step": 2800
    },
    {
      "epoch": 1.59,
      "grad_norm": 15.578303337097168,
      "learning_rate": 4.833333333333333e-06,
      "loss": 0.7348,
      "step": 2825
    },
    {
      "epoch": 1.61,
      "grad_norm": 9.60951042175293,
      "learning_rate": 4.777777777777778e-06,
      "loss": 0.8109,
      "step": 2850
    },
    {
      "epoch": 1.62,
      "grad_norm": 19.289772033691406,
      "learning_rate": 4.722222222222222e-06,
      "loss": 0.8359,
      "step": 2875
    },
    {
      "epoch": 1.64,
      "grad_norm": 15.233059883117676,
      "learning_rate": 4.666666666666667e-06,
      "loss": 0.8516,
      "step": 2900
    },
    {
      "epoch": 1.65,
      "grad_norm": 18.676319122314453,
      "learning_rate": 4.611111111111112e-06,
      "loss": 0.8884,
      "step": 2925
    },
    {
      "epoch": 1.66,
      "grad_norm": 12.788491249084473,
      "learning_rate": 4.555555555555556e-06,
      "loss": 0.7779,
      "step": 2950
    },
    {
      "epoch": 1.68,
      "grad_norm": 11.790907859802246,
      "learning_rate": 4.5e-06,
      "loss": 0.739,
      "step": 2975
    },
    {
      "epoch": 1.69,
      "grad_norm": 14.475822448730469,
      "learning_rate": 4.444444444444444e-06,
      "loss": 0.7393,
      "step": 3000
    },
    {
      "epoch": 1.69,
      "eval_cer": 41.74961534838446,
      "eval_loss": 1.0024151802062988,
      "eval_runtime": 903.7019,
      "eval_samples_per_second": 3.924,
      "eval_steps_per_second": 0.491,
      "eval_wer": 69.41389728096676,
      "step": 3000
    },
    {
      "epoch": 1.71,
      "grad_norm": 13.670875549316406,
      "learning_rate": 4.388888888888889e-06,
      "loss": 0.7552,
      "step": 3025
    },
    {
      "epoch": 1.72,
      "grad_norm": 17.736740112304688,
      "learning_rate": 4.333333333333334e-06,
      "loss": 0.7793,
      "step": 3050
    },
    {
      "epoch": 1.73,
      "grad_norm": 15.000802040100098,
      "learning_rate": 4.277777777777778e-06,
      "loss": 0.7238,
      "step": 3075
    },
    {
      "epoch": 1.75,
      "grad_norm": 13.509275436401367,
      "learning_rate": 4.222222222222223e-06,
      "loss": 0.7613,
      "step": 3100
    },
    {
      "epoch": 1.76,
      "grad_norm": 19.606420516967773,
      "learning_rate": 4.166666666666667e-06,
      "loss": 0.6867,
      "step": 3125
    },
    {
      "epoch": 1.78,
      "grad_norm": 20.64481544494629,
      "learning_rate": 4.111111111111111e-06,
      "loss": 0.8513,
      "step": 3150
    },
    {
      "epoch": 1.79,
      "grad_norm": 16.68463706970215,
      "learning_rate": 4.055555555555556e-06,
      "loss": 0.754,
      "step": 3175
    },
    {
      "epoch": 1.8,
      "grad_norm": 13.40403938293457,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.7143,
      "step": 3200
    },
    {
      "epoch": 1.82,
      "grad_norm": 15.898687362670898,
      "learning_rate": 3.944444444444445e-06,
      "loss": 0.7647,
      "step": 3225
    },
    {
      "epoch": 1.83,
      "grad_norm": 18.194766998291016,
      "learning_rate": 3.88888888888889e-06,
      "loss": 0.7521,
      "step": 3250
    },
    {
      "epoch": 1.85,
      "grad_norm": 14.401911735534668,
      "learning_rate": 3.833333333333334e-06,
      "loss": 0.7586,
      "step": 3275
    },
    {
      "epoch": 1.86,
      "grad_norm": 17.611783981323242,
      "learning_rate": 3.777777777777778e-06,
      "loss": 0.7286,
      "step": 3300
    },
    {
      "epoch": 1.88,
      "grad_norm": 22.452648162841797,
      "learning_rate": 3.7222222222222225e-06,
      "loss": 0.7609,
      "step": 3325
    },
    {
      "epoch": 1.89,
      "grad_norm": 19.66781997680664,
      "learning_rate": 3.6666666666666666e-06,
      "loss": 0.7042,
      "step": 3350
    },
    {
      "epoch": 1.9,
      "grad_norm": 17.88923454284668,
      "learning_rate": 3.6111111111111115e-06,
      "loss": 0.7208,
      "step": 3375
    },
    {
      "epoch": 1.92,
      "grad_norm": 14.06191635131836,
      "learning_rate": 3.555555555555556e-06,
      "loss": 0.8189,
      "step": 3400
    },
    {
      "epoch": 1.93,
      "grad_norm": 17.72393798828125,
      "learning_rate": 3.5e-06,
      "loss": 0.7533,
      "step": 3425
    },
    {
      "epoch": 1.95,
      "grad_norm": 15.48258113861084,
      "learning_rate": 3.444444444444445e-06,
      "loss": 0.8195,
      "step": 3450
    },
    {
      "epoch": 1.96,
      "grad_norm": 13.074950218200684,
      "learning_rate": 3.3888888888888893e-06,
      "loss": 0.7019,
      "step": 3475
    },
    {
      "epoch": 1.97,
      "grad_norm": 13.651080131530762,
      "learning_rate": 3.3333333333333333e-06,
      "loss": 0.8497,
      "step": 3500
    },
    {
      "epoch": 1.99,
      "grad_norm": 15.199976921081543,
      "learning_rate": 3.277777777777778e-06,
      "loss": 0.7122,
      "step": 3525
    },
    {
      "epoch": 2.0,
      "grad_norm": 15.230713844299316,
      "learning_rate": 3.2222222222222227e-06,
      "loss": 0.6821,
      "step": 3550
    },
    {
      "epoch": 2.02,
      "grad_norm": 10.71396255493164,
      "learning_rate": 3.1666666666666667e-06,
      "loss": 0.5199,
      "step": 3575
    },
    {
      "epoch": 2.03,
      "grad_norm": 14.009171485900879,
      "learning_rate": 3.1111111111111116e-06,
      "loss": 0.5152,
      "step": 3600
    },
    {
      "epoch": 2.04,
      "grad_norm": 14.985831260681152,
      "learning_rate": 3.055555555555556e-06,
      "loss": 0.6146,
      "step": 3625
    },
    {
      "epoch": 2.06,
      "grad_norm": 11.974893569946289,
      "learning_rate": 3e-06,
      "loss": 0.5589,
      "step": 3650
    },
    {
      "epoch": 2.07,
      "grad_norm": 8.09765625,
      "learning_rate": 2.944444444444445e-06,
      "loss": 0.473,
      "step": 3675
    },
    {
      "epoch": 2.09,
      "grad_norm": 13.75256061553955,
      "learning_rate": 2.888888888888889e-06,
      "loss": 0.5609,
      "step": 3700
    },
    {
      "epoch": 2.1,
      "grad_norm": 17.261474609375,
      "learning_rate": 2.8333333333333335e-06,
      "loss": 0.5907,
      "step": 3725
    },
    {
      "epoch": 2.12,
      "grad_norm": 15.0508451461792,
      "learning_rate": 2.7777777777777783e-06,
      "loss": 0.541,
      "step": 3750
    },
    {
      "epoch": 2.13,
      "grad_norm": 13.755702018737793,
      "learning_rate": 2.7222222222222224e-06,
      "loss": 0.6274,
      "step": 3775
    },
    {
      "epoch": 2.14,
      "grad_norm": 13.00074577331543,
      "learning_rate": 2.666666666666667e-06,
      "loss": 0.4982,
      "step": 3800
    },
    {
      "epoch": 2.16,
      "grad_norm": 14.188249588012695,
      "learning_rate": 2.6111111111111113e-06,
      "loss": 0.4688,
      "step": 3825
    },
    {
      "epoch": 2.17,
      "grad_norm": 17.107295989990234,
      "learning_rate": 2.5555555555555557e-06,
      "loss": 0.5059,
      "step": 3850
    },
    {
      "epoch": 2.19,
      "grad_norm": 11.36642074584961,
      "learning_rate": 2.5e-06,
      "loss": 0.563,
      "step": 3875
    },
    {
      "epoch": 2.2,
      "grad_norm": 10.15700912475586,
      "learning_rate": 2.4444444444444447e-06,
      "loss": 0.5302,
      "step": 3900
    },
    {
      "epoch": 2.21,
      "grad_norm": 13.058419227600098,
      "learning_rate": 2.388888888888889e-06,
      "loss": 0.5041,
      "step": 3925
    },
    {
      "epoch": 2.23,
      "grad_norm": 15.872893333435059,
      "learning_rate": 2.3333333333333336e-06,
      "loss": 0.5845,
      "step": 3950
    },
    {
      "epoch": 2.24,
      "grad_norm": 9.651241302490234,
      "learning_rate": 2.277777777777778e-06,
      "loss": 0.5145,
      "step": 3975
    },
    {
      "epoch": 2.26,
      "grad_norm": 11.108153343200684,
      "learning_rate": 2.222222222222222e-06,
      "loss": 0.556,
      "step": 4000
    },
    {
      "epoch": 2.26,
      "eval_cer": 43.06371086758564,
      "eval_loss": 1.0082812309265137,
      "eval_runtime": 913.0221,
      "eval_samples_per_second": 3.884,
      "eval_steps_per_second": 0.486,
      "eval_wer": 69.88922457200403,
      "step": 4000
    },
    {
      "epoch": 2.27,
      "grad_norm": 14.091168403625488,
      "learning_rate": 2.166666666666667e-06,
      "loss": 0.4955,
      "step": 4025
    },
    {
      "epoch": 2.28,
      "grad_norm": 16.199934005737305,
      "learning_rate": 2.1111111111111114e-06,
      "loss": 0.529,
      "step": 4050
    },
    {
      "epoch": 2.3,
      "grad_norm": 11.609833717346191,
      "learning_rate": 2.0555555555555555e-06,
      "loss": 0.4852,
      "step": 4075
    },
    {
      "epoch": 2.31,
      "grad_norm": 12.764281272888184,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.5283,
      "step": 4100
    },
    {
      "epoch": 2.33,
      "grad_norm": 8.958930015563965,
      "learning_rate": 1.944444444444445e-06,
      "loss": 0.4375,
      "step": 4125
    },
    {
      "epoch": 2.34,
      "grad_norm": 12.987662315368652,
      "learning_rate": 1.888888888888889e-06,
      "loss": 0.4881,
      "step": 4150
    },
    {
      "epoch": 2.35,
      "grad_norm": 10.337594032287598,
      "learning_rate": 1.8333333333333333e-06,
      "loss": 0.5212,
      "step": 4175
    },
    {
      "epoch": 2.37,
      "grad_norm": 15.952834129333496,
      "learning_rate": 1.777777777777778e-06,
      "loss": 0.5109,
      "step": 4200
    },
    {
      "epoch": 2.38,
      "grad_norm": 11.717411041259766,
      "learning_rate": 1.7222222222222224e-06,
      "loss": 0.5598,
      "step": 4225
    },
    {
      "epoch": 2.4,
      "grad_norm": 11.100090026855469,
      "learning_rate": 1.6666666666666667e-06,
      "loss": 0.5073,
      "step": 4250
    },
    {
      "epoch": 2.41,
      "grad_norm": 17.794734954833984,
      "learning_rate": 1.6111111111111113e-06,
      "loss": 0.522,
      "step": 4275
    },
    {
      "epoch": 2.43,
      "grad_norm": 16.543949127197266,
      "learning_rate": 1.5555555555555558e-06,
      "loss": 0.577,
      "step": 4300
    },
    {
      "epoch": 2.44,
      "grad_norm": 14.382917404174805,
      "learning_rate": 1.5e-06,
      "loss": 0.4889,
      "step": 4325
    },
    {
      "epoch": 2.45,
      "grad_norm": 6.578629970550537,
      "learning_rate": 1.4444444444444445e-06,
      "loss": 0.4715,
      "step": 4350
    },
    {
      "epoch": 2.47,
      "grad_norm": 14.872430801391602,
      "learning_rate": 1.3888888888888892e-06,
      "loss": 0.6008,
      "step": 4375
    },
    {
      "epoch": 2.48,
      "grad_norm": 16.687318801879883,
      "learning_rate": 1.3333333333333334e-06,
      "loss": 0.489,
      "step": 4400
    },
    {
      "epoch": 2.5,
      "grad_norm": 16.002120971679688,
      "learning_rate": 1.2777777777777779e-06,
      "loss": 0.5201,
      "step": 4425
    },
    {
      "epoch": 2.51,
      "grad_norm": 8.543059349060059,
      "learning_rate": 1.2222222222222223e-06,
      "loss": 0.5222,
      "step": 4450
    },
    {
      "epoch": 2.52,
      "grad_norm": 14.853689193725586,
      "learning_rate": 1.1666666666666668e-06,
      "loss": 0.5044,
      "step": 4475
    },
    {
      "epoch": 2.54,
      "grad_norm": 19.702091217041016,
      "learning_rate": 1.111111111111111e-06,
      "loss": 0.5847,
      "step": 4500
    },
    {
      "epoch": 2.55,
      "grad_norm": 18.463783264160156,
      "learning_rate": 1.0555555555555557e-06,
      "loss": 0.6202,
      "step": 4525
    },
    {
      "epoch": 2.57,
      "grad_norm": 10.704026222229004,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.5371,
      "step": 4550
    },
    {
      "epoch": 2.58,
      "grad_norm": 13.03596305847168,
      "learning_rate": 9.444444444444445e-07,
      "loss": 0.5034,
      "step": 4575
    },
    {
      "epoch": 2.59,
      "grad_norm": 8.77291488647461,
      "learning_rate": 8.88888888888889e-07,
      "loss": 0.5054,
      "step": 4600
    },
    {
      "epoch": 2.61,
      "grad_norm": 12.519658088684082,
      "learning_rate": 8.333333333333333e-07,
      "loss": 0.4917,
      "step": 4625
    },
    {
      "epoch": 2.62,
      "grad_norm": 13.293140411376953,
      "learning_rate": 7.777777777777779e-07,
      "loss": 0.5367,
      "step": 4650
    },
    {
      "epoch": 2.64,
      "grad_norm": 10.27344799041748,
      "learning_rate": 7.222222222222222e-07,
      "loss": 0.5368,
      "step": 4675
    },
    {
      "epoch": 2.65,
      "grad_norm": 17.37248992919922,
      "learning_rate": 6.666666666666667e-07,
      "loss": 0.5907,
      "step": 4700
    },
    {
      "epoch": 2.66,
      "grad_norm": 11.495616912841797,
      "learning_rate": 6.111111111111112e-07,
      "loss": 0.5692,
      "step": 4725
    },
    {
      "epoch": 2.68,
      "grad_norm": 13.74099349975586,
      "learning_rate": 5.555555555555555e-07,
      "loss": 0.4925,
      "step": 4750
    },
    {
      "epoch": 2.69,
      "grad_norm": 14.522932052612305,
      "learning_rate": 5.000000000000001e-07,
      "loss": 0.5133,
      "step": 4775
    },
    {
      "epoch": 2.71,
      "grad_norm": 14.491698265075684,
      "learning_rate": 4.444444444444445e-07,
      "loss": 0.5781,
      "step": 4800
    },
    {
      "epoch": 2.72,
      "grad_norm": 7.774211406707764,
      "learning_rate": 3.8888888888888895e-07,
      "loss": 0.4805,
      "step": 4825
    },
    {
      "epoch": 2.74,
      "grad_norm": 15.983479499816895,
      "learning_rate": 3.3333333333333335e-07,
      "loss": 0.582,
      "step": 4850
    },
    {
      "epoch": 2.75,
      "grad_norm": 14.96191692352295,
      "learning_rate": 2.7777777777777776e-07,
      "loss": 0.5464,
      "step": 4875
    },
    {
      "epoch": 2.76,
      "grad_norm": 11.486920356750488,
      "learning_rate": 2.2222222222222224e-07,
      "loss": 0.5766,
      "step": 4900
    },
    {
      "epoch": 2.78,
      "grad_norm": 9.038966178894043,
      "learning_rate": 1.6666666666666668e-07,
      "loss": 0.4745,
      "step": 4925
    },
    {
      "epoch": 2.79,
      "grad_norm": 13.27407455444336,
      "learning_rate": 1.1111111111111112e-07,
      "loss": 0.55,
      "step": 4950
    },
    {
      "epoch": 2.81,
      "grad_norm": 19.504352569580078,
      "learning_rate": 5.555555555555556e-08,
      "loss": 0.5034,
      "step": 4975
    },
    {
      "epoch": 2.82,
      "grad_norm": 12.238847732543945,
      "learning_rate": 0.0,
      "loss": 0.4653,
      "step": 5000
    },
    {
      "epoch": 2.82,
      "eval_cer": 41.698590134078565,
      "eval_loss": 1.0031510591506958,
      "eval_runtime": 912.0081,
      "eval_samples_per_second": 3.888,
      "eval_steps_per_second": 0.487,
      "eval_wer": 68.70090634441087,
      "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
}