{
  "best_metric": 74.92953873650924,
  "best_model_checkpoint": "./whisper-small-finetune_iraqi/checkpoint-3000",
  "epoch": 3.977724741447892,
  "eval_steps": 1000,
  "global_step": 5000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "grad_norm": 31.307842254638672,
      "learning_rate": 5.000000000000001e-07,
      "loss": 1.8461,
      "step": 25
    },
    {
      "epoch": 0.04,
      "grad_norm": 24.101598739624023,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 1.8684,
      "step": 50
    },
    {
      "epoch": 0.06,
      "grad_norm": 32.56498336791992,
      "learning_rate": 1.5e-06,
      "loss": 1.6799,
      "step": 75
    },
    {
      "epoch": 0.08,
      "grad_norm": 26.423107147216797,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 1.8388,
      "step": 100
    },
    {
      "epoch": 0.1,
      "grad_norm": 34.08100128173828,
      "learning_rate": 2.5e-06,
      "loss": 1.5596,
      "step": 125
    },
    {
      "epoch": 0.12,
      "grad_norm": 29.480880737304688,
      "learning_rate": 3e-06,
      "loss": 1.741,
      "step": 150
    },
    {
      "epoch": 0.14,
      "grad_norm": 20.32571792602539,
      "learning_rate": 3.5e-06,
      "loss": 1.5891,
      "step": 175
    },
    {
      "epoch": 0.16,
      "grad_norm": 22.268383026123047,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.5566,
      "step": 200
    },
    {
      "epoch": 0.18,
      "grad_norm": 25.940326690673828,
      "learning_rate": 4.5e-06,
      "loss": 1.5812,
      "step": 225
    },
    {
      "epoch": 0.2,
      "grad_norm": 20.023916244506836,
      "learning_rate": 5e-06,
      "loss": 1.4958,
      "step": 250
    },
    {
      "epoch": 0.22,
      "grad_norm": 24.432737350463867,
      "learning_rate": 5.500000000000001e-06,
      "loss": 1.5452,
      "step": 275
    },
    {
      "epoch": 0.24,
      "grad_norm": 24.809738159179688,
      "learning_rate": 6e-06,
      "loss": 1.4721,
      "step": 300
    },
    {
      "epoch": 0.26,
      "grad_norm": 26.316701889038086,
      "learning_rate": 6.5000000000000004e-06,
      "loss": 1.4539,
      "step": 325
    },
    {
      "epoch": 0.28,
      "grad_norm": 20.42281723022461,
      "learning_rate": 7e-06,
      "loss": 1.4232,
      "step": 350
    },
    {
      "epoch": 0.3,
      "grad_norm": 23.9652156829834,
      "learning_rate": 7.500000000000001e-06,
      "loss": 1.3441,
      "step": 375
    },
    {
      "epoch": 0.32,
      "grad_norm": 22.536012649536133,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.419,
      "step": 400
    },
    {
      "epoch": 0.34,
      "grad_norm": 24.226743698120117,
      "learning_rate": 8.5e-06,
      "loss": 1.5099,
      "step": 425
    },
    {
      "epoch": 0.36,
      "grad_norm": 23.327442169189453,
      "learning_rate": 9e-06,
      "loss": 1.451,
      "step": 450
    },
    {
      "epoch": 0.38,
      "grad_norm": 21.532514572143555,
      "learning_rate": 9.5e-06,
      "loss": 1.3318,
      "step": 475
    },
    {
      "epoch": 0.4,
      "grad_norm": 22.450647354125977,
      "learning_rate": 1e-05,
      "loss": 1.3831,
      "step": 500
    },
    {
      "epoch": 0.42,
      "grad_norm": 23.443180084228516,
      "learning_rate": 9.944444444444445e-06,
      "loss": 1.3584,
      "step": 525
    },
    {
      "epoch": 0.44,
      "grad_norm": 15.108329772949219,
      "learning_rate": 9.88888888888889e-06,
      "loss": 1.3257,
      "step": 550
    },
    {
      "epoch": 0.46,
      "grad_norm": 21.810331344604492,
      "learning_rate": 9.833333333333333e-06,
      "loss": 1.3358,
      "step": 575
    },
    {
      "epoch": 0.48,
      "grad_norm": 20.90282440185547,
      "learning_rate": 9.777777777777779e-06,
      "loss": 1.3223,
      "step": 600
    },
    {
      "epoch": 0.5,
      "grad_norm": 24.409061431884766,
      "learning_rate": 9.722222222222223e-06,
      "loss": 1.3732,
      "step": 625
    },
    {
      "epoch": 0.52,
      "grad_norm": 25.396240234375,
      "learning_rate": 9.666666666666667e-06,
      "loss": 1.4752,
      "step": 650
    },
    {
      "epoch": 0.54,
      "grad_norm": 22.128602981567383,
      "learning_rate": 9.611111111111112e-06,
      "loss": 1.316,
      "step": 675
    },
    {
      "epoch": 0.56,
      "grad_norm": 23.54676055908203,
      "learning_rate": 9.555555555555556e-06,
      "loss": 1.3471,
      "step": 700
    },
    {
      "epoch": 0.58,
      "grad_norm": 19.39595603942871,
      "learning_rate": 9.5e-06,
      "loss": 1.3318,
      "step": 725
    },
    {
      "epoch": 0.6,
      "grad_norm": 16.275251388549805,
      "learning_rate": 9.444444444444445e-06,
      "loss": 1.179,
      "step": 750
    },
    {
      "epoch": 0.62,
      "grad_norm": 20.7484073638916,
      "learning_rate": 9.38888888888889e-06,
      "loss": 1.2591,
      "step": 775
    },
    {
      "epoch": 0.64,
      "grad_norm": 25.535470962524414,
      "learning_rate": 9.333333333333334e-06,
      "loss": 1.2721,
      "step": 800
    },
    {
      "epoch": 0.66,
      "grad_norm": 21.40460777282715,
      "learning_rate": 9.277777777777778e-06,
      "loss": 1.2564,
      "step": 825
    },
    {
      "epoch": 0.68,
      "grad_norm": 23.19025993347168,
      "learning_rate": 9.222222222222224e-06,
      "loss": 1.2542,
      "step": 850
    },
    {
      "epoch": 0.7,
      "grad_norm": 18.77362632751465,
      "learning_rate": 9.166666666666666e-06,
      "loss": 1.1861,
      "step": 875
    },
    {
      "epoch": 0.72,
      "grad_norm": 27.533153533935547,
      "learning_rate": 9.111111111111112e-06,
      "loss": 1.3505,
      "step": 900
    },
    {
      "epoch": 0.74,
      "grad_norm": 21.657012939453125,
      "learning_rate": 9.055555555555556e-06,
      "loss": 1.1985,
      "step": 925
    },
    {
      "epoch": 0.76,
      "grad_norm": 27.935184478759766,
      "learning_rate": 9e-06,
      "loss": 1.2384,
      "step": 950
    },
    {
      "epoch": 0.78,
      "grad_norm": 22.997709274291992,
      "learning_rate": 8.944444444444446e-06,
      "loss": 1.2963,
      "step": 975
    },
    {
      "epoch": 0.8,
      "grad_norm": 22.01042938232422,
      "learning_rate": 8.888888888888888e-06,
      "loss": 1.2175,
      "step": 1000
    },
    {
      "epoch": 0.8,
      "eval_cer": 53.03030303030303,
      "eval_loss": 1.2452795505523682,
      "eval_runtime": 621.4587,
      "eval_samples_per_second": 4.044,
      "eval_steps_per_second": 0.507,
      "eval_wer": 81.205746889393,
      "step": 1000
    },
    {
      "epoch": 0.82,
      "grad_norm": 22.637540817260742,
      "learning_rate": 8.833333333333334e-06,
      "loss": 1.3225,
      "step": 1025
    },
    {
      "epoch": 0.84,
      "grad_norm": 22.166149139404297,
      "learning_rate": 8.777777777777778e-06,
      "loss": 1.2662,
      "step": 1050
    },
    {
      "epoch": 0.86,
      "grad_norm": 21.192859649658203,
      "learning_rate": 8.722222222222224e-06,
      "loss": 1.2507,
      "step": 1075
    },
    {
      "epoch": 0.88,
      "grad_norm": 25.296127319335938,
      "learning_rate": 8.666666666666668e-06,
      "loss": 1.3007,
      "step": 1100
    },
    {
      "epoch": 0.89,
      "grad_norm": 19.297321319580078,
      "learning_rate": 8.611111111111112e-06,
      "loss": 1.2212,
      "step": 1125
    },
    {
      "epoch": 0.91,
      "grad_norm": 24.768157958984375,
      "learning_rate": 8.555555555555556e-06,
      "loss": 1.2646,
      "step": 1150
    },
    {
      "epoch": 0.93,
      "grad_norm": 13.836922645568848,
      "learning_rate": 8.5e-06,
      "loss": 1.1391,
      "step": 1175
    },
    {
      "epoch": 0.95,
      "grad_norm": 18.015613555908203,
      "learning_rate": 8.444444444444446e-06,
      "loss": 1.3267,
      "step": 1200
    },
    {
      "epoch": 0.97,
      "grad_norm": 18.187437057495117,
      "learning_rate": 8.38888888888889e-06,
      "loss": 1.1999,
      "step": 1225
    },
    {
      "epoch": 0.99,
      "grad_norm": 20.086061477661133,
      "learning_rate": 8.333333333333334e-06,
      "loss": 1.2112,
      "step": 1250
    },
    {
      "epoch": 1.01,
      "grad_norm": 14.848949432373047,
      "learning_rate": 8.277777777777778e-06,
      "loss": 1.0858,
      "step": 1275
    },
    {
      "epoch": 1.03,
      "grad_norm": 18.931108474731445,
      "learning_rate": 8.222222222222222e-06,
      "loss": 0.904,
      "step": 1300
    },
    {
      "epoch": 1.05,
      "grad_norm": 17.61419105529785,
      "learning_rate": 8.166666666666668e-06,
      "loss": 0.9141,
      "step": 1325
    },
    {
      "epoch": 1.07,
      "grad_norm": 14.918651580810547,
      "learning_rate": 8.111111111111112e-06,
      "loss": 0.8404,
      "step": 1350
    },
    {
      "epoch": 1.09,
      "grad_norm": 13.606977462768555,
      "learning_rate": 8.055555555555557e-06,
      "loss": 0.9239,
      "step": 1375
    },
    {
      "epoch": 1.11,
      "grad_norm": 10.916210174560547,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.9142,
      "step": 1400
    },
    {
      "epoch": 1.13,
      "grad_norm": 15.314620971679688,
      "learning_rate": 7.944444444444445e-06,
      "loss": 0.8994,
      "step": 1425
    },
    {
      "epoch": 1.15,
      "grad_norm": 17.887939453125,
      "learning_rate": 7.88888888888889e-06,
      "loss": 0.9692,
      "step": 1450
    },
    {
      "epoch": 1.17,
      "grad_norm": 18.108110427856445,
      "learning_rate": 7.833333333333333e-06,
      "loss": 1.0226,
      "step": 1475
    },
    {
      "epoch": 1.19,
      "grad_norm": 19.22450828552246,
      "learning_rate": 7.77777777777778e-06,
      "loss": 0.9862,
      "step": 1500
    },
    {
      "epoch": 1.21,
      "grad_norm": 16.892333984375,
      "learning_rate": 7.722222222222223e-06,
      "loss": 0.8676,
      "step": 1525
    },
    {
      "epoch": 1.23,
      "grad_norm": 19.769046783447266,
      "learning_rate": 7.666666666666667e-06,
      "loss": 1.009,
      "step": 1550
    },
    {
      "epoch": 1.25,
      "grad_norm": 17.185529708862305,
      "learning_rate": 7.611111111111111e-06,
      "loss": 0.9733,
      "step": 1575
    },
    {
      "epoch": 1.27,
      "grad_norm": 16.545021057128906,
      "learning_rate": 7.555555555555556e-06,
      "loss": 0.8958,
      "step": 1600
    },
    {
      "epoch": 1.29,
      "grad_norm": 12.434943199157715,
      "learning_rate": 7.500000000000001e-06,
      "loss": 0.8863,
      "step": 1625
    },
    {
      "epoch": 1.31,
      "grad_norm": 17.79275894165039,
      "learning_rate": 7.444444444444445e-06,
      "loss": 0.9409,
      "step": 1650
    },
    {
      "epoch": 1.33,
      "grad_norm": 18.283716201782227,
      "learning_rate": 7.38888888888889e-06,
      "loss": 0.9022,
      "step": 1675
    },
    {
      "epoch": 1.35,
      "grad_norm": 12.344863891601562,
      "learning_rate": 7.333333333333333e-06,
      "loss": 0.9641,
      "step": 1700
    },
    {
      "epoch": 1.37,
      "grad_norm": 19.406063079833984,
      "learning_rate": 7.277777777777778e-06,
      "loss": 0.9584,
      "step": 1725
    },
    {
      "epoch": 1.39,
      "grad_norm": 14.66451644897461,
      "learning_rate": 7.222222222222223e-06,
      "loss": 0.9161,
      "step": 1750
    },
    {
      "epoch": 1.41,
      "grad_norm": 22.2624454498291,
      "learning_rate": 7.166666666666667e-06,
      "loss": 0.8841,
      "step": 1775
    },
    {
      "epoch": 1.43,
      "grad_norm": 21.720008850097656,
      "learning_rate": 7.111111111111112e-06,
      "loss": 0.8905,
      "step": 1800
    },
    {
      "epoch": 1.45,
      "grad_norm": 18.579879760742188,
      "learning_rate": 7.055555555555557e-06,
      "loss": 0.9017,
      "step": 1825
    },
    {
      "epoch": 1.47,
      "grad_norm": 13.965752601623535,
      "learning_rate": 7e-06,
      "loss": 0.9258,
      "step": 1850
    },
    {
      "epoch": 1.49,
      "grad_norm": 12.050531387329102,
      "learning_rate": 6.944444444444445e-06,
      "loss": 0.8406,
      "step": 1875
    },
    {
      "epoch": 1.51,
      "grad_norm": 20.574005126953125,
      "learning_rate": 6.88888888888889e-06,
      "loss": 0.9911,
      "step": 1900
    },
    {
      "epoch": 1.53,
      "grad_norm": 13.421600341796875,
      "learning_rate": 6.833333333333334e-06,
      "loss": 0.9802,
      "step": 1925
    },
    {
      "epoch": 1.55,
      "grad_norm": 15.10280704498291,
      "learning_rate": 6.777777777777779e-06,
      "loss": 0.9281,
      "step": 1950
    },
    {
      "epoch": 1.57,
      "grad_norm": 16.386531829833984,
      "learning_rate": 6.7222222222222235e-06,
      "loss": 0.9103,
      "step": 1975
    },
    {
      "epoch": 1.59,
      "grad_norm": 21.13428497314453,
      "learning_rate": 6.666666666666667e-06,
      "loss": 0.8821,
      "step": 2000
    },
    {
      "epoch": 1.59,
      "eval_cer": 58.570669096984886,
      "eval_loss": 1.163949728012085,
      "eval_runtime": 688.3941,
      "eval_samples_per_second": 3.651,
      "eval_steps_per_second": 0.458,
      "eval_wer": 80.6833023991201,
      "step": 2000
    },
    {
      "epoch": 1.61,
      "grad_norm": 12.398762702941895,
      "learning_rate": 6.6111111111111115e-06,
      "loss": 0.9263,
      "step": 2025
    },
    {
      "epoch": 1.63,
      "grad_norm": 19.7735538482666,
      "learning_rate": 6.555555555555556e-06,
      "loss": 0.9111,
      "step": 2050
    },
    {
      "epoch": 1.65,
      "grad_norm": 17.024389266967773,
      "learning_rate": 6.5000000000000004e-06,
      "loss": 0.8293,
      "step": 2075
    },
    {
      "epoch": 1.67,
      "grad_norm": 25.842208862304688,
      "learning_rate": 6.444444444444445e-06,
      "loss": 0.9776,
      "step": 2100
    },
    {
      "epoch": 1.69,
      "grad_norm": 13.255263328552246,
      "learning_rate": 6.3888888888888885e-06,
      "loss": 0.8695,
      "step": 2125
    },
    {
      "epoch": 1.71,
      "grad_norm": 23.875268936157227,
      "learning_rate": 6.333333333333333e-06,
      "loss": 0.965,
      "step": 2150
    },
    {
      "epoch": 1.73,
      "grad_norm": 12.608431816101074,
      "learning_rate": 6.277777777777778e-06,
      "loss": 0.9176,
      "step": 2175
    },
    {
      "epoch": 1.75,
      "grad_norm": 18.595909118652344,
      "learning_rate": 6.222222222222223e-06,
      "loss": 0.9389,
      "step": 2200
    },
    {
      "epoch": 1.77,
      "grad_norm": 17.5798397064209,
      "learning_rate": 6.166666666666667e-06,
      "loss": 0.8804,
      "step": 2225
    },
    {
      "epoch": 1.79,
      "grad_norm": 20.442508697509766,
      "learning_rate": 6.111111111111112e-06,
      "loss": 0.9779,
      "step": 2250
    },
    {
      "epoch": 1.81,
      "grad_norm": 17.262054443359375,
      "learning_rate": 6.055555555555555e-06,
      "loss": 0.9289,
      "step": 2275
    },
    {
      "epoch": 1.83,
      "grad_norm": 11.93671989440918,
      "learning_rate": 6e-06,
      "loss": 0.8674,
      "step": 2300
    },
    {
      "epoch": 1.85,
      "grad_norm": 19.9697322845459,
      "learning_rate": 5.944444444444445e-06,
      "loss": 0.8906,
      "step": 2325
    },
    {
      "epoch": 1.87,
      "grad_norm": 19.09278678894043,
      "learning_rate": 5.88888888888889e-06,
      "loss": 0.949,
      "step": 2350
    },
    {
      "epoch": 1.89,
      "grad_norm": 13.083151817321777,
      "learning_rate": 5.833333333333334e-06,
      "loss": 0.8925,
      "step": 2375
    },
    {
      "epoch": 1.91,
      "grad_norm": 16.96628761291504,
      "learning_rate": 5.777777777777778e-06,
      "loss": 0.8806,
      "step": 2400
    },
    {
      "epoch": 1.93,
      "grad_norm": 17.384708404541016,
      "learning_rate": 5.722222222222222e-06,
      "loss": 0.909,
      "step": 2425
    },
    {
      "epoch": 1.95,
      "grad_norm": 16.435260772705078,
      "learning_rate": 5.666666666666667e-06,
      "loss": 0.9044,
      "step": 2450
    },
    {
      "epoch": 1.97,
      "grad_norm": 19.849964141845703,
      "learning_rate": 5.611111111111112e-06,
      "loss": 0.8914,
      "step": 2475
    },
    {
      "epoch": 1.99,
      "grad_norm": 18.659517288208008,
      "learning_rate": 5.555555555555557e-06,
      "loss": 0.8724,
      "step": 2500
    },
    {
      "epoch": 2.01,
      "grad_norm": 11.454329490661621,
      "learning_rate": 5.500000000000001e-06,
      "loss": 0.6817,
      "step": 2525
    },
    {
      "epoch": 2.03,
      "grad_norm": 15.80736255645752,
      "learning_rate": 5.444444444444445e-06,
      "loss": 0.5733,
      "step": 2550
    },
    {
      "epoch": 2.05,
      "grad_norm": 22.253652572631836,
      "learning_rate": 5.388888888888889e-06,
      "loss": 0.6498,
      "step": 2575
    },
    {
      "epoch": 2.07,
      "grad_norm": 16.574365615844727,
      "learning_rate": 5.333333333333334e-06,
      "loss": 0.586,
      "step": 2600
    },
    {
      "epoch": 2.09,
      "grad_norm": 15.736115455627441,
      "learning_rate": 5.2777777777777785e-06,
      "loss": 0.5926,
      "step": 2625
    },
    {
      "epoch": 2.11,
      "grad_norm": 14.3034086227417,
      "learning_rate": 5.2222222222222226e-06,
      "loss": 0.6365,
      "step": 2650
    },
    {
      "epoch": 2.13,
      "grad_norm": 14.139527320861816,
      "learning_rate": 5.1666666666666675e-06,
      "loss": 0.6302,
      "step": 2675
    },
    {
      "epoch": 2.15,
      "grad_norm": 11.31550121307373,
      "learning_rate": 5.1111111111111115e-06,
      "loss": 0.5432,
      "step": 2700
    },
    {
      "epoch": 2.17,
      "grad_norm": 20.395217895507812,
      "learning_rate": 5.0555555555555555e-06,
      "loss": 0.6269,
      "step": 2725
    },
    {
      "epoch": 2.19,
      "grad_norm": 17.261280059814453,
      "learning_rate": 5e-06,
      "loss": 0.6443,
      "step": 2750
    },
    {
      "epoch": 2.21,
      "grad_norm": 12.30817699432373,
      "learning_rate": 4.944444444444445e-06,
      "loss": 0.5756,
      "step": 2775
    },
    {
      "epoch": 2.23,
      "grad_norm": 9.878785133361816,
      "learning_rate": 4.888888888888889e-06,
      "loss": 0.639,
      "step": 2800
    },
    {
      "epoch": 2.25,
      "grad_norm": 14.701591491699219,
      "learning_rate": 4.833333333333333e-06,
      "loss": 0.6066,
      "step": 2825
    },
    {
      "epoch": 2.27,
      "grad_norm": 16.308835983276367,
      "learning_rate": 4.777777777777778e-06,
      "loss": 0.6264,
      "step": 2850
    },
    {
      "epoch": 2.29,
      "grad_norm": 19.56263542175293,
      "learning_rate": 4.722222222222222e-06,
      "loss": 0.6369,
      "step": 2875
    },
    {
      "epoch": 2.31,
      "grad_norm": 6.289492607116699,
      "learning_rate": 4.666666666666667e-06,
      "loss": 0.6256,
      "step": 2900
    },
    {
      "epoch": 2.33,
      "grad_norm": 13.804625511169434,
      "learning_rate": 4.611111111111112e-06,
      "loss": 0.6128,
      "step": 2925
    },
    {
      "epoch": 2.35,
      "grad_norm": 15.281023979187012,
      "learning_rate": 4.555555555555556e-06,
      "loss": 0.645,
      "step": 2950
    },
    {
      "epoch": 2.37,
      "grad_norm": 14.549323081970215,
      "learning_rate": 4.5e-06,
      "loss": 0.7194,
      "step": 2975
    },
    {
      "epoch": 2.39,
      "grad_norm": 15.436699867248535,
      "learning_rate": 4.444444444444444e-06,
      "loss": 0.6527,
      "step": 3000
    },
    {
      "epoch": 2.39,
      "eval_cer": 48.39877471456418,
      "eval_loss": 1.1647427082061768,
      "eval_runtime": 640.6707,
      "eval_samples_per_second": 3.922,
      "eval_steps_per_second": 0.492,
      "eval_wer": 74.92953873650924,
      "step": 3000
    },
    {
      "epoch": 2.41,
      "grad_norm": 10.962376594543457,
      "learning_rate": 4.388888888888889e-06,
      "loss": 0.6551,
      "step": 3025
    },
    {
      "epoch": 2.43,
      "grad_norm": 14.927529335021973,
      "learning_rate": 4.333333333333334e-06,
      "loss": 0.5794,
      "step": 3050
    },
    {
      "epoch": 2.45,
      "grad_norm": 18.27710723876953,
      "learning_rate": 4.277777777777778e-06,
      "loss": 0.6381,
      "step": 3075
    },
    {
      "epoch": 2.47,
      "grad_norm": 12.02866268157959,
      "learning_rate": 4.222222222222223e-06,
      "loss": 0.559,
      "step": 3100
    },
    {
      "epoch": 2.49,
      "grad_norm": 18.277559280395508,
      "learning_rate": 4.166666666666667e-06,
      "loss": 0.6653,
      "step": 3125
    },
    {
      "epoch": 2.51,
      "grad_norm": 17.41031265258789,
      "learning_rate": 4.111111111111111e-06,
      "loss": 0.648,
      "step": 3150
    },
    {
      "epoch": 2.53,
      "grad_norm": 17.889631271362305,
      "learning_rate": 4.055555555555556e-06,
      "loss": 0.5711,
      "step": 3175
    },
    {
      "epoch": 2.55,
      "grad_norm": 12.564661026000977,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.6177,
      "step": 3200
    },
    {
      "epoch": 2.57,
      "grad_norm": 12.346695899963379,
      "learning_rate": 3.944444444444445e-06,
      "loss": 0.607,
      "step": 3225
    },
    {
      "epoch": 2.59,
      "grad_norm": 15.423745155334473,
      "learning_rate": 3.88888888888889e-06,
      "loss": 0.6286,
      "step": 3250
    },
    {
      "epoch": 2.61,
      "grad_norm": 9.96914005279541,
      "learning_rate": 3.833333333333334e-06,
      "loss": 0.609,
      "step": 3275
    },
    {
      "epoch": 2.63,
      "grad_norm": 17.189481735229492,
      "learning_rate": 3.777777777777778e-06,
      "loss": 0.5851,
      "step": 3300
    },
    {
      "epoch": 2.65,
      "grad_norm": 12.618528366088867,
      "learning_rate": 3.7222222222222225e-06,
      "loss": 0.6227,
      "step": 3325
    },
    {
      "epoch": 2.67,
      "grad_norm": 14.24203109741211,
      "learning_rate": 3.6666666666666666e-06,
      "loss": 0.637,
      "step": 3350
    },
    {
      "epoch": 2.68,
      "grad_norm": 19.71699333190918,
      "learning_rate": 3.6111111111111115e-06,
      "loss": 0.6088,
      "step": 3375
    },
    {
      "epoch": 2.7,
      "grad_norm": 18.02073860168457,
      "learning_rate": 3.555555555555556e-06,
      "loss": 0.6252,
      "step": 3400
    },
    {
      "epoch": 2.72,
      "grad_norm": 14.6773099899292,
      "learning_rate": 3.5e-06,
      "loss": 0.5767,
      "step": 3425
    },
    {
      "epoch": 2.74,
      "grad_norm": 13.4647855758667,
      "learning_rate": 3.444444444444445e-06,
      "loss": 0.6186,
      "step": 3450
    },
    {
      "epoch": 2.76,
      "grad_norm": 15.786669731140137,
      "learning_rate": 3.3888888888888893e-06,
      "loss": 0.5699,
      "step": 3475
    },
    {
      "epoch": 2.78,
      "grad_norm": 10.151808738708496,
      "learning_rate": 3.3333333333333333e-06,
      "loss": 0.571,
      "step": 3500
    },
    {
      "epoch": 2.8,
      "grad_norm": 11.594339370727539,
      "learning_rate": 3.277777777777778e-06,
      "loss": 0.6104,
      "step": 3525
    },
    {
      "epoch": 2.82,
      "grad_norm": 19.481727600097656,
      "learning_rate": 3.2222222222222227e-06,
      "loss": 0.584,
      "step": 3550
    },
    {
      "epoch": 2.84,
      "grad_norm": 15.422442436218262,
      "learning_rate": 3.1666666666666667e-06,
      "loss": 0.5463,
      "step": 3575
    },
    {
      "epoch": 2.86,
      "grad_norm": 11.132479667663574,
      "learning_rate": 3.1111111111111116e-06,
      "loss": 0.5999,
      "step": 3600
    },
    {
      "epoch": 2.88,
      "grad_norm": 7.921601295471191,
      "learning_rate": 3.055555555555556e-06,
      "loss": 0.5775,
      "step": 3625
    },
    {
      "epoch": 2.9,
      "grad_norm": 18.08513832092285,
      "learning_rate": 3e-06,
      "loss": 0.6061,
      "step": 3650
    },
    {
      "epoch": 2.92,
      "grad_norm": 12.494170188903809,
      "learning_rate": 2.944444444444445e-06,
      "loss": 0.5436,
      "step": 3675
    },
    {
      "epoch": 2.94,
      "grad_norm": 12.588512420654297,
      "learning_rate": 2.888888888888889e-06,
      "loss": 0.5468,
      "step": 3700
    },
    {
      "epoch": 2.96,
      "grad_norm": 13.475690841674805,
      "learning_rate": 2.8333333333333335e-06,
      "loss": 0.5645,
      "step": 3725
    },
    {
      "epoch": 2.98,
      "grad_norm": 10.302828788757324,
      "learning_rate": 2.7777777777777783e-06,
      "loss": 0.5733,
      "step": 3750
    },
    {
      "epoch": 3.0,
      "grad_norm": 12.000411033630371,
      "learning_rate": 2.7222222222222224e-06,
      "loss": 0.5959,
      "step": 3775
    },
    {
      "epoch": 3.02,
      "grad_norm": 9.559981346130371,
      "learning_rate": 2.666666666666667e-06,
      "loss": 0.3823,
      "step": 3800
    },
    {
      "epoch": 3.04,
      "grad_norm": 6.879577159881592,
      "learning_rate": 2.6111111111111113e-06,
      "loss": 0.3876,
      "step": 3825
    },
    {
      "epoch": 3.06,
      "grad_norm": 11.854860305786133,
      "learning_rate": 2.5555555555555557e-06,
      "loss": 0.3963,
      "step": 3850
    },
    {
      "epoch": 3.08,
      "grad_norm": 8.722013473510742,
      "learning_rate": 2.5e-06,
      "loss": 0.4091,
      "step": 3875
    },
    {
      "epoch": 3.1,
      "grad_norm": 8.34683609008789,
      "learning_rate": 2.4444444444444447e-06,
      "loss": 0.4084,
      "step": 3900
    },
    {
      "epoch": 3.12,
      "grad_norm": 14.6788969039917,
      "learning_rate": 2.388888888888889e-06,
      "loss": 0.4166,
      "step": 3925
    },
    {
      "epoch": 3.14,
      "grad_norm": 12.618093490600586,
      "learning_rate": 2.3333333333333336e-06,
      "loss": 0.3751,
      "step": 3950
    },
    {
      "epoch": 3.16,
      "grad_norm": 7.677009582519531,
      "learning_rate": 2.277777777777778e-06,
      "loss": 0.3864,
      "step": 3975
    },
    {
      "epoch": 3.18,
      "grad_norm": 6.518270492553711,
      "learning_rate": 2.222222222222222e-06,
      "loss": 0.3078,
      "step": 4000
    },
    {
      "epoch": 3.18,
      "eval_cer": 53.61383256120098,
      "eval_loss": 1.188154935836792,
      "eval_runtime": 654.7314,
      "eval_samples_per_second": 3.838,
      "eval_steps_per_second": 0.481,
      "eval_wer": 85.45404550766482,
      "step": 4000
    },
    {
      "epoch": 3.2,
      "grad_norm": 12.325676918029785,
      "learning_rate": 2.166666666666667e-06,
      "loss": 0.4185,
      "step": 4025
    },
    {
      "epoch": 3.22,
      "grad_norm": 9.633131980895996,
      "learning_rate": 2.1111111111111114e-06,
      "loss": 0.5039,
      "step": 4050
    },
    {
      "epoch": 3.24,
      "grad_norm": 12.042168617248535,
      "learning_rate": 2.0555555555555555e-06,
      "loss": 0.3689,
      "step": 4075
    },
    {
      "epoch": 3.26,
      "grad_norm": 11.479193687438965,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.3856,
      "step": 4100
    },
    {
      "epoch": 3.28,
      "grad_norm": 15.161530494689941,
      "learning_rate": 1.944444444444445e-06,
      "loss": 0.4055,
      "step": 4125
    },
    {
      "epoch": 3.3,
      "grad_norm": 13.629692077636719,
      "learning_rate": 1.888888888888889e-06,
      "loss": 0.4257,
      "step": 4150
    },
    {
      "epoch": 3.32,
      "grad_norm": 7.724813461303711,
      "learning_rate": 1.8333333333333333e-06,
      "loss": 0.4206,
      "step": 4175
    },
    {
      "epoch": 3.34,
      "grad_norm": 11.88393783569336,
      "learning_rate": 1.777777777777778e-06,
      "loss": 0.3937,
      "step": 4200
    },
    {
      "epoch": 3.36,
      "grad_norm": 11.900699615478516,
      "learning_rate": 1.7222222222222224e-06,
      "loss": 0.4389,
      "step": 4225
    },
    {
      "epoch": 3.38,
      "grad_norm": 15.184126853942871,
      "learning_rate": 1.6666666666666667e-06,
      "loss": 0.4195,
      "step": 4250
    },
    {
      "epoch": 3.4,
      "grad_norm": 17.46352767944336,
      "learning_rate": 1.6111111111111113e-06,
      "loss": 0.4278,
      "step": 4275
    },
    {
      "epoch": 3.42,
      "grad_norm": 6.922684669494629,
      "learning_rate": 1.5555555555555558e-06,
      "loss": 0.3707,
      "step": 4300
    },
    {
      "epoch": 3.44,
      "grad_norm": 8.876372337341309,
      "learning_rate": 1.5e-06,
      "loss": 0.4042,
      "step": 4325
    },
    {
      "epoch": 3.46,
      "grad_norm": 11.79426097869873,
      "learning_rate": 1.4444444444444445e-06,
      "loss": 0.4306,
      "step": 4350
    },
    {
      "epoch": 3.48,
      "grad_norm": 13.896903038024902,
      "learning_rate": 1.3888888888888892e-06,
      "loss": 0.4667,
      "step": 4375
    },
    {
      "epoch": 3.5,
      "grad_norm": 15.71197509765625,
      "learning_rate": 1.3333333333333334e-06,
      "loss": 0.3795,
      "step": 4400
    },
    {
      "epoch": 3.52,
      "grad_norm": 24.959102630615234,
      "learning_rate": 1.2777777777777779e-06,
      "loss": 0.3962,
      "step": 4425
    },
    {
      "epoch": 3.54,
      "grad_norm": 14.953639030456543,
      "learning_rate": 1.2222222222222223e-06,
      "loss": 0.405,
      "step": 4450
    },
    {
      "epoch": 3.56,
      "grad_norm": 9.408991813659668,
      "learning_rate": 1.1666666666666668e-06,
      "loss": 0.3741,
      "step": 4475
    },
    {
      "epoch": 3.58,
      "grad_norm": 13.057042121887207,
      "learning_rate": 1.111111111111111e-06,
      "loss": 0.443,
      "step": 4500
    },
    {
      "epoch": 3.6,
      "grad_norm": 13.196622848510742,
      "learning_rate": 1.0555555555555557e-06,
      "loss": 0.3873,
      "step": 4525
    },
    {
      "epoch": 3.62,
      "grad_norm": 13.450845718383789,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.4196,
      "step": 4550
    },
    {
      "epoch": 3.64,
      "grad_norm": 11.10022258758545,
      "learning_rate": 9.444444444444445e-07,
      "loss": 0.4027,
      "step": 4575
    },
    {
      "epoch": 3.66,
      "grad_norm": 11.709542274475098,
      "learning_rate": 8.88888888888889e-07,
      "loss": 0.3875,
      "step": 4600
    },
    {
      "epoch": 3.68,
      "grad_norm": 18.268943786621094,
      "learning_rate": 8.333333333333333e-07,
      "loss": 0.4404,
      "step": 4625
    },
    {
      "epoch": 3.7,
      "grad_norm": 9.290139198303223,
      "learning_rate": 7.777777777777779e-07,
      "loss": 0.3566,
      "step": 4650
    },
    {
      "epoch": 3.72,
      "grad_norm": 9.714057922363281,
      "learning_rate": 7.222222222222222e-07,
      "loss": 0.3539,
      "step": 4675
    },
    {
      "epoch": 3.74,
      "grad_norm": 12.28147029876709,
      "learning_rate": 6.666666666666667e-07,
      "loss": 0.4019,
      "step": 4700
    },
    {
      "epoch": 3.76,
      "grad_norm": 14.317485809326172,
      "learning_rate": 6.111111111111112e-07,
      "loss": 0.4342,
      "step": 4725
    },
    {
      "epoch": 3.78,
      "grad_norm": 14.031659126281738,
      "learning_rate": 5.555555555555555e-07,
      "loss": 0.3736,
      "step": 4750
    },
    {
      "epoch": 3.8,
      "grad_norm": 11.357044219970703,
      "learning_rate": 5.000000000000001e-07,
      "loss": 0.4058,
      "step": 4775
    },
    {
      "epoch": 3.82,
      "grad_norm": 15.167471885681152,
      "learning_rate": 4.444444444444445e-07,
      "loss": 0.3479,
      "step": 4800
    },
    {
      "epoch": 3.84,
      "grad_norm": 10.228837966918945,
      "learning_rate": 3.8888888888888895e-07,
      "loss": 0.3747,
      "step": 4825
    },
    {
      "epoch": 3.86,
      "grad_norm": 15.795928955078125,
      "learning_rate": 3.3333333333333335e-07,
      "loss": 0.4141,
      "step": 4850
    },
    {
      "epoch": 3.88,
      "grad_norm": 14.220040321350098,
      "learning_rate": 2.7777777777777776e-07,
      "loss": 0.3817,
      "step": 4875
    },
    {
      "epoch": 3.9,
      "grad_norm": 15.395153999328613,
      "learning_rate": 2.2222222222222224e-07,
      "loss": 0.3983,
      "step": 4900
    },
    {
      "epoch": 3.92,
      "grad_norm": 16.81828498840332,
      "learning_rate": 1.6666666666666668e-07,
      "loss": 0.4367,
      "step": 4925
    },
    {
      "epoch": 3.94,
      "grad_norm": 12.835176467895508,
      "learning_rate": 1.1111111111111112e-07,
      "loss": 0.3874,
      "step": 4950
    },
    {
      "epoch": 3.96,
      "grad_norm": 16.80717658996582,
      "learning_rate": 5.555555555555556e-08,
      "loss": 0.3914,
      "step": 4975
    },
    {
      "epoch": 3.98,
      "grad_norm": 17.128372192382812,
      "learning_rate": 0.0,
      "loss": 0.403,
      "step": 5000
    },
    {
      "epoch": 3.98,
      "eval_cer": 51.96957039062302,
      "eval_loss": 1.188420295715332,
      "eval_runtime": 653.7956,
      "eval_samples_per_second": 3.844,
      "eval_steps_per_second": 0.482,
      "eval_wer": 79.83776723723103,
      "step": 5000
    }
  ],
  "logging_steps": 25,
  "max_steps": 5000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 1000,
  "total_flos": 1.15390872317952e+19,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}