{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 14.79466271312083,
  "eval_steps": 1000,
  "global_step": 5000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07412898443291327,
      "grad_norm": 11.929125785827637,
      "learning_rate": 4.4e-07,
      "loss": 1.6381,
      "step": 25
    },
    {
      "epoch": 0.14825796886582654,
      "grad_norm": 8.018507957458496,
      "learning_rate": 9.400000000000001e-07,
      "loss": 1.3896,
      "step": 50
    },
    {
      "epoch": 0.2223869532987398,
      "grad_norm": 6.990742206573486,
      "learning_rate": 1.44e-06,
      "loss": 1.2335,
      "step": 75
    },
    {
      "epoch": 0.2965159377316531,
      "grad_norm": 7.199395656585693,
      "learning_rate": 1.94e-06,
      "loss": 1.0625,
      "step": 100
    },
    {
      "epoch": 0.37064492216456635,
      "grad_norm": 5.8580546379089355,
      "learning_rate": 2.4400000000000004e-06,
      "loss": 0.9354,
      "step": 125
    },
    {
      "epoch": 0.4447739065974796,
      "grad_norm": 6.414823055267334,
      "learning_rate": 2.9400000000000002e-06,
      "loss": 0.8125,
      "step": 150
    },
    {
      "epoch": 0.5189028910303929,
      "grad_norm": 6.512091159820557,
      "learning_rate": 3.44e-06,
      "loss": 0.7764,
      "step": 175
    },
    {
      "epoch": 0.5930318754633062,
      "grad_norm": 6.021186351776123,
      "learning_rate": 3.94e-06,
      "loss": 0.7381,
      "step": 200
    },
    {
      "epoch": 0.6671608598962194,
      "grad_norm": 5.814141273498535,
      "learning_rate": 4.440000000000001e-06,
      "loss": 0.6923,
      "step": 225
    },
    {
      "epoch": 0.7412898443291327,
      "grad_norm": 5.2472968101501465,
      "learning_rate": 4.94e-06,
      "loss": 0.6586,
      "step": 250
    },
    {
      "epoch": 0.815418828762046,
      "grad_norm": 5.4379401206970215,
      "learning_rate": 5.4400000000000004e-06,
      "loss": 0.6041,
      "step": 275
    },
    {
      "epoch": 0.8895478131949592,
      "grad_norm": 5.554481029510498,
      "learning_rate": 5.94e-06,
      "loss": 0.5964,
      "step": 300
    },
    {
      "epoch": 0.9636767976278725,
      "grad_norm": 6.5946502685546875,
      "learning_rate": 6.440000000000001e-06,
      "loss": 0.5586,
      "step": 325
    },
    {
      "epoch": 1.0355819125277983,
      "grad_norm": 4.164698600769043,
      "learning_rate": 6.9400000000000005e-06,
      "loss": 0.4838,
      "step": 350
    },
    {
      "epoch": 1.1097108969607117,
      "grad_norm": 6.609652519226074,
      "learning_rate": 7.440000000000001e-06,
      "loss": 0.4744,
      "step": 375
    },
    {
      "epoch": 1.1838398813936248,
      "grad_norm": 4.401813507080078,
      "learning_rate": 7.94e-06,
      "loss": 0.4447,
      "step": 400
    },
    {
      "epoch": 1.2579688658265382,
      "grad_norm": 4.387920379638672,
      "learning_rate": 8.44e-06,
      "loss": 0.4425,
      "step": 425
    },
    {
      "epoch": 1.3320978502594514,
      "grad_norm": 4.439324855804443,
      "learning_rate": 8.94e-06,
      "loss": 0.3978,
      "step": 450
    },
    {
      "epoch": 1.4062268346923648,
      "grad_norm": 4.0201826095581055,
      "learning_rate": 9.440000000000001e-06,
      "loss": 0.4059,
      "step": 475
    },
    {
      "epoch": 1.480355819125278,
      "grad_norm": 4.707728385925293,
      "learning_rate": 9.940000000000001e-06,
      "loss": 0.3959,
      "step": 500
    },
    {
      "epoch": 1.5544848035581913,
      "grad_norm": 4.224379062652588,
      "learning_rate": 9.951111111111111e-06,
      "loss": 0.3747,
      "step": 525
    },
    {
      "epoch": 1.6286137879911045,
      "grad_norm": 4.27752161026001,
      "learning_rate": 9.895555555555557e-06,
      "loss": 0.3749,
      "step": 550
    },
    {
      "epoch": 1.7027427724240178,
      "grad_norm": 4.656911373138428,
      "learning_rate": 9.84e-06,
      "loss": 0.3667,
      "step": 575
    },
    {
      "epoch": 1.7768717568569312,
      "grad_norm": 4.724191188812256,
      "learning_rate": 9.784444444444445e-06,
      "loss": 0.3474,
      "step": 600
    },
    {
      "epoch": 1.8510007412898444,
      "grad_norm": 4.367356300354004,
      "learning_rate": 9.72888888888889e-06,
      "loss": 0.3445,
      "step": 625
    },
    {
      "epoch": 1.9251297257227575,
      "grad_norm": 4.204794406890869,
      "learning_rate": 9.673333333333334e-06,
      "loss": 0.3353,
      "step": 650
    },
    {
      "epoch": 1.999258710155671,
      "grad_norm": 3.796111583709717,
      "learning_rate": 9.617777777777778e-06,
      "loss": 0.3136,
      "step": 675
    },
    {
      "epoch": 2.0711638250555966,
      "grad_norm": 3.014369249343872,
      "learning_rate": 9.562222222222223e-06,
      "loss": 0.2109,
      "step": 700
    },
    {
      "epoch": 2.14529280948851,
      "grad_norm": 3.4962964057922363,
      "learning_rate": 9.506666666666667e-06,
      "loss": 0.2112,
      "step": 725
    },
    {
      "epoch": 2.2194217939214234,
      "grad_norm": 3.688762903213501,
      "learning_rate": 9.451111111111112e-06,
      "loss": 0.2122,
      "step": 750
    },
    {
      "epoch": 2.2935507783543367,
      "grad_norm": 4.557786464691162,
      "learning_rate": 9.395555555555556e-06,
      "loss": 0.2107,
      "step": 775
    },
    {
      "epoch": 2.3676797627872497,
      "grad_norm": 3.3230926990509033,
      "learning_rate": 9.340000000000002e-06,
      "loss": 0.2115,
      "step": 800
    },
    {
      "epoch": 2.441808747220163,
      "grad_norm": 3.251685380935669,
      "learning_rate": 9.284444444444444e-06,
      "loss": 0.2021,
      "step": 825
    },
    {
      "epoch": 2.5159377316530764,
      "grad_norm": 3.852593421936035,
      "learning_rate": 9.22888888888889e-06,
      "loss": 0.1999,
      "step": 850
    },
    {
      "epoch": 2.59006671608599,
      "grad_norm": 3.0748980045318604,
      "learning_rate": 9.173333333333334e-06,
      "loss": 0.1936,
      "step": 875
    },
    {
      "epoch": 2.6641957005189028,
      "grad_norm": 2.834657907485962,
      "learning_rate": 9.117777777777778e-06,
      "loss": 0.2031,
      "step": 900
    },
    {
      "epoch": 2.738324684951816,
      "grad_norm": 2.9382002353668213,
      "learning_rate": 9.062222222222224e-06,
      "loss": 0.199,
      "step": 925
    },
    {
      "epoch": 2.8124536693847295,
      "grad_norm": 2.9481730461120605,
      "learning_rate": 9.006666666666666e-06,
      "loss": 0.1867,
      "step": 950
    },
    {
      "epoch": 2.8865826538176425,
      "grad_norm": 3.9300479888916016,
      "learning_rate": 8.951111111111112e-06,
      "loss": 0.203,
      "step": 975
    },
    {
      "epoch": 2.960711638250556,
      "grad_norm": 3.3928661346435547,
      "learning_rate": 8.895555555555556e-06,
      "loss": 0.2023,
      "step": 1000
    },
    {
      "epoch": 2.960711638250556,
      "eval_loss": 0.3293725550174713,
      "eval_runtime": 936.3299,
      "eval_samples_per_second": 3.081,
      "eval_steps_per_second": 0.386,
      "eval_wer": 0.18741347189225221,
      "step": 1000
    },
    {
      "epoch": 3.032616753150482,
      "grad_norm": 2.452481269836426,
      "learning_rate": 8.84e-06,
      "loss": 0.152,
      "step": 1025
    },
    {
      "epoch": 3.106745737583395,
      "grad_norm": 2.802464485168457,
      "learning_rate": 8.784444444444446e-06,
      "loss": 0.1076,
      "step": 1050
    },
    {
      "epoch": 3.1808747220163083,
      "grad_norm": 3.293215751647949,
      "learning_rate": 8.72888888888889e-06,
      "loss": 0.1114,
      "step": 1075
    },
    {
      "epoch": 3.2550037064492217,
      "grad_norm": 2.4673008918762207,
      "learning_rate": 8.673333333333334e-06,
      "loss": 0.1075,
      "step": 1100
    },
    {
      "epoch": 3.329132690882135,
      "grad_norm": 2.8392369747161865,
      "learning_rate": 8.617777777777778e-06,
      "loss": 0.1112,
      "step": 1125
    },
    {
      "epoch": 3.403261675315048,
      "grad_norm": 2.4692773818969727,
      "learning_rate": 8.562222222222224e-06,
      "loss": 0.1109,
      "step": 1150
    },
    {
      "epoch": 3.4773906597479614,
      "grad_norm": 3.809983968734741,
      "learning_rate": 8.506666666666668e-06,
      "loss": 0.1173,
      "step": 1175
    },
    {
      "epoch": 3.5515196441808747,
      "grad_norm": 2.7140321731567383,
      "learning_rate": 8.451111111111112e-06,
      "loss": 0.1131,
      "step": 1200
    },
    {
      "epoch": 3.625648628613788,
      "grad_norm": 2.5393877029418945,
      "learning_rate": 8.395555555555557e-06,
      "loss": 0.1077,
      "step": 1225
    },
    {
      "epoch": 3.6997776130467015,
      "grad_norm": 2.905407667160034,
      "learning_rate": 8.34e-06,
      "loss": 0.1054,
      "step": 1250
    },
    {
      "epoch": 3.7739065974796144,
      "grad_norm": 2.8358991146087646,
      "learning_rate": 8.284444444444446e-06,
      "loss": 0.1073,
      "step": 1275
    },
    {
      "epoch": 3.848035581912528,
      "grad_norm": 2.4956672191619873,
      "learning_rate": 8.22888888888889e-06,
      "loss": 0.1052,
      "step": 1300
    },
    {
      "epoch": 3.922164566345441,
      "grad_norm": 3.1253950595855713,
      "learning_rate": 8.173333333333334e-06,
      "loss": 0.1064,
      "step": 1325
    },
    {
      "epoch": 3.996293550778354,
      "grad_norm": 3.244260549545288,
      "learning_rate": 8.11777777777778e-06,
      "loss": 0.0973,
      "step": 1350
    },
    {
      "epoch": 4.06819866567828,
      "grad_norm": 1.2010555267333984,
      "learning_rate": 8.062222222222222e-06,
      "loss": 0.0602,
      "step": 1375
    },
    {
      "epoch": 4.142327650111193,
      "grad_norm": 1.965174913406372,
      "learning_rate": 8.006666666666667e-06,
      "loss": 0.0607,
      "step": 1400
    },
    {
      "epoch": 4.216456634544107,
      "grad_norm": 1.7773408889770508,
      "learning_rate": 7.951111111111111e-06,
      "loss": 0.0539,
      "step": 1425
    },
    {
      "epoch": 4.29058561897702,
      "grad_norm": 2.6301095485687256,
      "learning_rate": 7.895555555555557e-06,
      "loss": 0.0606,
      "step": 1450
    },
    {
      "epoch": 4.364714603409933,
      "grad_norm": 1.90291428565979,
      "learning_rate": 7.840000000000001e-06,
      "loss": 0.0584,
      "step": 1475
    },
    {
      "epoch": 4.438843587842847,
      "grad_norm": 2.0360007286071777,
      "learning_rate": 7.784444444444445e-06,
      "loss": 0.0553,
      "step": 1500
    },
    {
      "epoch": 4.51297257227576,
      "grad_norm": 2.2208218574523926,
      "learning_rate": 7.72888888888889e-06,
      "loss": 0.0656,
      "step": 1525
    },
    {
      "epoch": 4.5871015567086735,
      "grad_norm": 1.9523277282714844,
      "learning_rate": 7.673333333333333e-06,
      "loss": 0.0537,
      "step": 1550
    },
    {
      "epoch": 4.661230541141586,
      "grad_norm": 1.7108356952667236,
      "learning_rate": 7.617777777777778e-06,
      "loss": 0.0575,
      "step": 1575
    },
    {
      "epoch": 4.735359525574499,
      "grad_norm": 1.8699008226394653,
      "learning_rate": 7.562222222222223e-06,
      "loss": 0.0548,
      "step": 1600
    },
    {
      "epoch": 4.809488510007413,
      "grad_norm": 2.0190742015838623,
      "learning_rate": 7.506666666666668e-06,
      "loss": 0.0548,
      "step": 1625
    },
    {
      "epoch": 4.883617494440326,
      "grad_norm": 1.592809796333313,
      "learning_rate": 7.451111111111111e-06,
      "loss": 0.0536,
      "step": 1650
    },
    {
      "epoch": 4.957746478873239,
      "grad_norm": 1.7714742422103882,
      "learning_rate": 7.395555555555556e-06,
      "loss": 0.0522,
      "step": 1675
    },
    {
      "epoch": 5.029651593773165,
      "grad_norm": 0.9238072037696838,
      "learning_rate": 7.340000000000001e-06,
      "loss": 0.0427,
      "step": 1700
    },
    {
      "epoch": 5.103780578206079,
      "grad_norm": 1.1125034093856812,
      "learning_rate": 7.284444444444445e-06,
      "loss": 0.0269,
      "step": 1725
    },
    {
      "epoch": 5.177909562638992,
      "grad_norm": 1.247423768043518,
      "learning_rate": 7.22888888888889e-06,
      "loss": 0.0257,
      "step": 1750
    },
    {
      "epoch": 5.252038547071905,
      "grad_norm": 1.1966558694839478,
      "learning_rate": 7.173333333333335e-06,
      "loss": 0.029,
      "step": 1775
    },
    {
      "epoch": 5.326167531504819,
      "grad_norm": 0.9568462371826172,
      "learning_rate": 7.117777777777778e-06,
      "loss": 0.0294,
      "step": 1800
    },
    {
      "epoch": 5.400296515937732,
      "grad_norm": 1.0882899761199951,
      "learning_rate": 7.062222222222223e-06,
      "loss": 0.028,
      "step": 1825
    },
    {
      "epoch": 5.474425500370645,
      "grad_norm": 1.1803560256958008,
      "learning_rate": 7.006666666666667e-06,
      "loss": 0.0295,
      "step": 1850
    },
    {
      "epoch": 5.548554484803558,
      "grad_norm": 1.2876580953598022,
      "learning_rate": 6.951111111111112e-06,
      "loss": 0.0334,
      "step": 1875
    },
    {
      "epoch": 5.622683469236471,
      "grad_norm": 1.7007198333740234,
      "learning_rate": 6.8955555555555565e-06,
      "loss": 0.0294,
      "step": 1900
    },
    {
      "epoch": 5.696812453669384,
      "grad_norm": 0.9945253729820251,
      "learning_rate": 6.8400000000000014e-06,
      "loss": 0.0287,
      "step": 1925
    },
    {
      "epoch": 5.770941438102298,
      "grad_norm": 1.3555738925933838,
      "learning_rate": 6.784444444444445e-06,
      "loss": 0.0278,
      "step": 1950
    },
    {
      "epoch": 5.845070422535211,
      "grad_norm": 1.6404248476028442,
      "learning_rate": 6.7288888888888895e-06,
      "loss": 0.027,
      "step": 1975
    },
    {
      "epoch": 5.919199406968125,
      "grad_norm": 1.1645821332931519,
      "learning_rate": 6.6733333333333335e-06,
      "loss": 0.0309,
      "step": 2000
    },
    {
      "epoch": 5.919199406968125,
      "eval_loss": 0.3413762152194977,
      "eval_runtime": 911.1637,
      "eval_samples_per_second": 3.166,
      "eval_steps_per_second": 0.396,
      "eval_wer": 0.17174013365922025,
      "step": 2000
    },
    {
      "epoch": 5.993328391401038,
      "grad_norm": 1.239682674407959,
      "learning_rate": 6.617777777777778e-06,
      "loss": 0.0246,
      "step": 2025
    },
    {
      "epoch": 6.065233506300964,
      "grad_norm": 0.5780806541442871,
      "learning_rate": 6.562222222222223e-06,
      "loss": 0.0158,
      "step": 2050
    },
    {
      "epoch": 6.139362490733877,
      "grad_norm": 0.7804180383682251,
      "learning_rate": 6.5066666666666665e-06,
      "loss": 0.0149,
      "step": 2075
    },
    {
      "epoch": 6.21349147516679,
      "grad_norm": 1.2513881921768188,
      "learning_rate": 6.451111111111111e-06,
      "loss": 0.0186,
      "step": 2100
    },
    {
      "epoch": 6.287620459599704,
      "grad_norm": 1.0758005380630493,
      "learning_rate": 6.395555555555556e-06,
      "loss": 0.0176,
      "step": 2125
    },
    {
      "epoch": 6.3617494440326166,
      "grad_norm": 1.6794716119766235,
      "learning_rate": 6.34e-06,
      "loss": 0.0166,
      "step": 2150
    },
    {
      "epoch": 6.43587842846553,
      "grad_norm": 0.4483010768890381,
      "learning_rate": 6.284444444444445e-06,
      "loss": 0.0116,
      "step": 2175
    },
    {
      "epoch": 6.510007412898443,
      "grad_norm": 0.6224784255027771,
      "learning_rate": 6.22888888888889e-06,
      "loss": 0.0122,
      "step": 2200
    },
    {
      "epoch": 6.584136397331356,
      "grad_norm": 0.5704225897789001,
      "learning_rate": 6.173333333333333e-06,
      "loss": 0.0138,
      "step": 2225
    },
    {
      "epoch": 6.65826538176427,
      "grad_norm": 0.976425051689148,
      "learning_rate": 6.117777777777778e-06,
      "loss": 0.0142,
      "step": 2250
    },
    {
      "epoch": 6.732394366197183,
      "grad_norm": 1.829641580581665,
      "learning_rate": 6.062222222222223e-06,
      "loss": 0.0139,
      "step": 2275
    },
    {
      "epoch": 6.806523350630096,
      "grad_norm": 0.645282506942749,
      "learning_rate": 6.006666666666667e-06,
      "loss": 0.0154,
      "step": 2300
    },
    {
      "epoch": 6.88065233506301,
      "grad_norm": 0.9580066204071045,
      "learning_rate": 5.951111111111112e-06,
      "loss": 0.0119,
      "step": 2325
    },
    {
      "epoch": 6.954781319495923,
      "grad_norm": 1.3833215236663818,
      "learning_rate": 5.895555555555557e-06,
      "loss": 0.0163,
      "step": 2350
    },
    {
      "epoch": 7.026686434395849,
      "grad_norm": 0.3075716495513916,
      "learning_rate": 5.84e-06,
      "loss": 0.0115,
      "step": 2375
    },
    {
      "epoch": 7.100815418828762,
      "grad_norm": 1.460079312324524,
      "learning_rate": 5.784444444444445e-06,
      "loss": 0.0098,
      "step": 2400
    },
    {
      "epoch": 7.174944403261676,
      "grad_norm": 0.24449419975280762,
      "learning_rate": 5.72888888888889e-06,
      "loss": 0.0082,
      "step": 2425
    },
    {
      "epoch": 7.2490733876945885,
      "grad_norm": 0.4475979208946228,
      "learning_rate": 5.673333333333334e-06,
      "loss": 0.008,
      "step": 2450
    },
    {
      "epoch": 7.3232023721275015,
      "grad_norm": 0.9485690593719482,
      "learning_rate": 5.617777777777779e-06,
      "loss": 0.0086,
      "step": 2475
    },
    {
      "epoch": 7.397331356560415,
      "grad_norm": 0.4818005859851837,
      "learning_rate": 5.562222222222222e-06,
      "loss": 0.0079,
      "step": 2500
    },
    {
      "epoch": 7.471460340993328,
      "grad_norm": 0.2728360891342163,
      "learning_rate": 5.506666666666667e-06,
      "loss": 0.0071,
      "step": 2525
    },
    {
      "epoch": 7.545589325426242,
      "grad_norm": 0.36738815903663635,
      "learning_rate": 5.451111111111112e-06,
      "loss": 0.0071,
      "step": 2550
    },
    {
      "epoch": 7.619718309859155,
      "grad_norm": 0.5995059609413147,
      "learning_rate": 5.3955555555555565e-06,
      "loss": 0.0071,
      "step": 2575
    },
    {
      "epoch": 7.693847294292068,
      "grad_norm": 0.6667615175247192,
      "learning_rate": 5.3400000000000005e-06,
      "loss": 0.0071,
      "step": 2600
    },
    {
      "epoch": 7.767976278724982,
      "grad_norm": 0.9709383845329285,
      "learning_rate": 5.2844444444444454e-06,
      "loss": 0.0077,
      "step": 2625
    },
    {
      "epoch": 7.842105263157895,
      "grad_norm": 0.9174025058746338,
      "learning_rate": 5.228888888888889e-06,
      "loss": 0.0064,
      "step": 2650
    },
    {
      "epoch": 7.916234247590808,
      "grad_norm": 0.4343569576740265,
      "learning_rate": 5.1733333333333335e-06,
      "loss": 0.006,
      "step": 2675
    },
    {
      "epoch": 7.9903632320237215,
      "grad_norm": 1.9002019166946411,
      "learning_rate": 5.117777777777778e-06,
      "loss": 0.0071,
      "step": 2700
    },
    {
      "epoch": 8.062268346923647,
      "grad_norm": 0.3372269570827484,
      "learning_rate": 5.062222222222222e-06,
      "loss": 0.005,
      "step": 2725
    },
    {
      "epoch": 8.13639733135656,
      "grad_norm": 0.8460608720779419,
      "learning_rate": 5.006666666666667e-06,
      "loss": 0.005,
      "step": 2750
    },
    {
      "epoch": 8.210526315789474,
      "grad_norm": 0.24673226475715637,
      "learning_rate": 4.951111111111111e-06,
      "loss": 0.0066,
      "step": 2775
    },
    {
      "epoch": 8.284655300222386,
      "grad_norm": 0.20106859505176544,
      "learning_rate": 4.895555555555556e-06,
      "loss": 0.0037,
      "step": 2800
    },
    {
      "epoch": 8.3587842846553,
      "grad_norm": 0.1510283499956131,
      "learning_rate": 4.84e-06,
      "loss": 0.0044,
      "step": 2825
    },
    {
      "epoch": 8.432913269088214,
      "grad_norm": 0.7809015512466431,
      "learning_rate": 4.784444444444445e-06,
      "loss": 0.0041,
      "step": 2850
    },
    {
      "epoch": 8.507042253521126,
      "grad_norm": 0.3642624318599701,
      "learning_rate": 4.728888888888889e-06,
      "loss": 0.0048,
      "step": 2875
    },
    {
      "epoch": 8.58117123795404,
      "grad_norm": 0.20495444536209106,
      "learning_rate": 4.673333333333333e-06,
      "loss": 0.0036,
      "step": 2900
    },
    {
      "epoch": 8.655300222386954,
      "grad_norm": 0.14625044167041779,
      "learning_rate": 4.617777777777778e-06,
      "loss": 0.0041,
      "step": 2925
    },
    {
      "epoch": 8.729429206819866,
      "grad_norm": 0.1600050926208496,
      "learning_rate": 4.562222222222222e-06,
      "loss": 0.0044,
      "step": 2950
    },
    {
      "epoch": 8.80355819125278,
      "grad_norm": 1.0638575553894043,
      "learning_rate": 4.506666666666667e-06,
      "loss": 0.0036,
      "step": 2975
    },
    {
      "epoch": 8.877687175685693,
      "grad_norm": 0.16608619689941406,
      "learning_rate": 4.451111111111112e-06,
      "loss": 0.0037,
      "step": 3000
    },
    {
      "epoch": 8.877687175685693,
      "eval_loss": 0.37816545367240906,
      "eval_runtime": 954.3156,
      "eval_samples_per_second": 3.023,
      "eval_steps_per_second": 0.378,
      "eval_wer": 0.17080007520467635,
      "step": 3000
    },
    {
      "epoch": 8.951816160118607,
      "grad_norm": 0.13698986172676086,
      "learning_rate": 4.395555555555556e-06,
      "loss": 0.0043,
      "step": 3025
    },
    {
      "epoch": 9.023721275018532,
      "grad_norm": 0.15848596394062042,
      "learning_rate": 4.34e-06,
      "loss": 0.0044,
      "step": 3050
    },
    {
      "epoch": 9.097850259451446,
      "grad_norm": 0.12134752422571182,
      "learning_rate": 4.284444444444445e-06,
      "loss": 0.0031,
      "step": 3075
    },
    {
      "epoch": 9.171979243884358,
      "grad_norm": 0.13019242882728577,
      "learning_rate": 4.228888888888889e-06,
      "loss": 0.0027,
      "step": 3100
    },
    {
      "epoch": 9.246108228317272,
      "grad_norm": 0.19179771840572357,
      "learning_rate": 4.173333333333334e-06,
      "loss": 0.0028,
      "step": 3125
    },
    {
      "epoch": 9.320237212750186,
      "grad_norm": 0.20711077749729156,
      "learning_rate": 4.117777777777779e-06,
      "loss": 0.0027,
      "step": 3150
    },
    {
      "epoch": 9.394366197183098,
      "grad_norm": 0.09309308230876923,
      "learning_rate": 4.062222222222223e-06,
      "loss": 0.0026,
      "step": 3175
    },
    {
      "epoch": 9.468495181616012,
      "grad_norm": 0.09770819544792175,
      "learning_rate": 4.006666666666667e-06,
      "loss": 0.0035,
      "step": 3200
    },
    {
      "epoch": 9.542624166048926,
      "grad_norm": 0.1151503324508667,
      "learning_rate": 3.951111111111112e-06,
      "loss": 0.0027,
      "step": 3225
    },
    {
      "epoch": 9.616753150481838,
      "grad_norm": 0.1251019835472107,
      "learning_rate": 3.895555555555556e-06,
      "loss": 0.0029,
      "step": 3250
    },
    {
      "epoch": 9.690882134914752,
      "grad_norm": 0.10979008674621582,
      "learning_rate": 3.8400000000000005e-06,
      "loss": 0.0027,
      "step": 3275
    },
    {
      "epoch": 9.765011119347665,
      "grad_norm": 0.12528590857982635,
      "learning_rate": 3.784444444444445e-06,
      "loss": 0.0037,
      "step": 3300
    },
    {
      "epoch": 9.839140103780577,
      "grad_norm": 0.12703803181648254,
      "learning_rate": 3.728888888888889e-06,
      "loss": 0.0026,
      "step": 3325
    },
    {
      "epoch": 9.913269088213491,
      "grad_norm": 0.10044371336698532,
      "learning_rate": 3.673333333333334e-06,
      "loss": 0.003,
      "step": 3350
    },
    {
      "epoch": 9.987398072646405,
      "grad_norm": 0.3393096625804901,
      "learning_rate": 3.617777777777778e-06,
      "loss": 0.0032,
      "step": 3375
    },
    {
      "epoch": 10.05930318754633,
      "grad_norm": 0.08865969628095627,
      "learning_rate": 3.5622222222222224e-06,
      "loss": 0.0034,
      "step": 3400
    },
    {
      "epoch": 10.133432171979244,
      "grad_norm": 0.07572653144598007,
      "learning_rate": 3.5066666666666673e-06,
      "loss": 0.0023,
      "step": 3425
    },
    {
      "epoch": 10.207561156412158,
      "grad_norm": 0.08611512184143066,
      "learning_rate": 3.4511111111111113e-06,
      "loss": 0.0021,
      "step": 3450
    },
    {
      "epoch": 10.28169014084507,
      "grad_norm": 0.09255647659301758,
      "learning_rate": 3.3955555555555558e-06,
      "loss": 0.0021,
      "step": 3475
    },
    {
      "epoch": 10.355819125277984,
      "grad_norm": 0.09140720963478088,
      "learning_rate": 3.3400000000000006e-06,
      "loss": 0.0021,
      "step": 3500
    },
    {
      "epoch": 10.429948109710898,
      "grad_norm": 0.08117499947547913,
      "learning_rate": 3.2844444444444447e-06,
      "loss": 0.002,
      "step": 3525
    },
    {
      "epoch": 10.50407709414381,
      "grad_norm": 0.07924593240022659,
      "learning_rate": 3.228888888888889e-06,
      "loss": 0.0021,
      "step": 3550
    },
    {
      "epoch": 10.578206078576724,
      "grad_norm": 0.09375474601984024,
      "learning_rate": 3.173333333333334e-06,
      "loss": 0.0023,
      "step": 3575
    },
    {
      "epoch": 10.652335063009637,
      "grad_norm": 0.08366698771715164,
      "learning_rate": 3.117777777777778e-06,
      "loss": 0.0022,
      "step": 3600
    },
    {
      "epoch": 10.72646404744255,
      "grad_norm": 0.09042704850435257,
      "learning_rate": 3.0622222222222225e-06,
      "loss": 0.0023,
      "step": 3625
    },
    {
      "epoch": 10.800593031875463,
      "grad_norm": 0.07148691266775131,
      "learning_rate": 3.0066666666666674e-06,
      "loss": 0.002,
      "step": 3650
    },
    {
      "epoch": 10.874722016308377,
      "grad_norm": 0.3031787574291229,
      "learning_rate": 2.9511111111111114e-06,
      "loss": 0.0022,
      "step": 3675
    },
    {
      "epoch": 10.94885100074129,
      "grad_norm": 0.09258675575256348,
      "learning_rate": 2.895555555555556e-06,
      "loss": 0.002,
      "step": 3700
    },
    {
      "epoch": 11.020756115641216,
      "grad_norm": 0.06253425031900406,
      "learning_rate": 2.84e-06,
      "loss": 0.002,
      "step": 3725
    },
    {
      "epoch": 11.094885100074128,
      "grad_norm": 0.057923316955566406,
      "learning_rate": 2.784444444444445e-06,
      "loss": 0.0017,
      "step": 3750
    },
    {
      "epoch": 11.169014084507042,
      "grad_norm": 0.06114868074655533,
      "learning_rate": 2.7288888888888893e-06,
      "loss": 0.0018,
      "step": 3775
    },
    {
      "epoch": 11.243143068939956,
      "grad_norm": 0.061181288212537766,
      "learning_rate": 2.6733333333333333e-06,
      "loss": 0.0017,
      "step": 3800
    },
    {
      "epoch": 11.317272053372868,
      "grad_norm": 0.06324835866689682,
      "learning_rate": 2.617777777777778e-06,
      "loss": 0.0018,
      "step": 3825
    },
    {
      "epoch": 11.391401037805782,
      "grad_norm": 0.07077573984861374,
      "learning_rate": 2.5622222222222226e-06,
      "loss": 0.0019,
      "step": 3850
    },
    {
      "epoch": 11.465530022238696,
      "grad_norm": 0.065274178981781,
      "learning_rate": 2.5066666666666667e-06,
      "loss": 0.0019,
      "step": 3875
    },
    {
      "epoch": 11.53965900667161,
      "grad_norm": 0.08122989535331726,
      "learning_rate": 2.451111111111111e-06,
      "loss": 0.0019,
      "step": 3900
    },
    {
      "epoch": 11.613787991104521,
      "grad_norm": 0.06203175708651543,
      "learning_rate": 2.3955555555555556e-06,
      "loss": 0.0018,
      "step": 3925
    },
    {
      "epoch": 11.687916975537435,
      "grad_norm": 0.08688808977603912,
      "learning_rate": 2.3400000000000005e-06,
      "loss": 0.0018,
      "step": 3950
    },
    {
      "epoch": 11.762045959970349,
      "grad_norm": 0.06490565836429596,
      "learning_rate": 2.2844444444444445e-06,
      "loss": 0.0017,
      "step": 3975
    },
    {
      "epoch": 11.836174944403261,
      "grad_norm": 0.06775514036417007,
      "learning_rate": 2.228888888888889e-06,
      "loss": 0.0018,
      "step": 4000
    },
    {
      "epoch": 11.836174944403261,
      "eval_loss": 0.39641496539115906,
      "eval_runtime": 944.1469,
      "eval_samples_per_second": 3.056,
      "eval_steps_per_second": 0.382,
      "eval_wer": 0.16803117575674706,
      "step": 4000
    },
    {
      "epoch": 11.910303928836175,
      "grad_norm": 0.06309942156076431,
      "learning_rate": 2.1733333333333334e-06,
      "loss": 0.0017,
      "step": 4025
    },
    {
      "epoch": 11.984432913269089,
      "grad_norm": 0.061430562287569046,
      "learning_rate": 2.117777777777778e-06,
      "loss": 0.0017,
      "step": 4050
    },
    {
      "epoch": 12.056338028169014,
      "grad_norm": 0.05334235355257988,
      "learning_rate": 2.0622222222222223e-06,
      "loss": 0.0016,
      "step": 4075
    },
    {
      "epoch": 12.130467012601928,
      "grad_norm": 0.057459134608507156,
      "learning_rate": 2.006666666666667e-06,
      "loss": 0.0016,
      "step": 4100
    },
    {
      "epoch": 12.20459599703484,
      "grad_norm": 0.047365590929985046,
      "learning_rate": 1.9511111111111113e-06,
      "loss": 0.0017,
      "step": 4125
    },
    {
      "epoch": 12.278724981467754,
      "grad_norm": 0.056059811264276505,
      "learning_rate": 1.8955555555555557e-06,
      "loss": 0.0015,
      "step": 4150
    },
    {
      "epoch": 12.352853965900668,
      "grad_norm": 0.062216296792030334,
      "learning_rate": 1.8400000000000002e-06,
      "loss": 0.0016,
      "step": 4175
    },
    {
      "epoch": 12.42698295033358,
      "grad_norm": 0.055066876113414764,
      "learning_rate": 1.7844444444444444e-06,
      "loss": 0.0015,
      "step": 4200
    },
    {
      "epoch": 12.501111934766493,
      "grad_norm": 0.06531412899494171,
      "learning_rate": 1.728888888888889e-06,
      "loss": 0.0015,
      "step": 4225
    },
    {
      "epoch": 12.575240919199407,
      "grad_norm": 0.04836783558130264,
      "learning_rate": 1.6733333333333335e-06,
      "loss": 0.0016,
      "step": 4250
    },
    {
      "epoch": 12.649369903632321,
      "grad_norm": 0.050022318959236145,
      "learning_rate": 1.6177777777777778e-06,
      "loss": 0.0016,
      "step": 4275
    },
    {
      "epoch": 12.723498888065233,
      "grad_norm": 0.05791894719004631,
      "learning_rate": 1.5622222222222225e-06,
      "loss": 0.0015,
      "step": 4300
    },
    {
      "epoch": 12.797627872498147,
      "grad_norm": 0.050857629626989365,
      "learning_rate": 1.506666666666667e-06,
      "loss": 0.0016,
      "step": 4325
    },
    {
      "epoch": 12.87175685693106,
      "grad_norm": 0.05395572632551193,
      "learning_rate": 1.4511111111111112e-06,
      "loss": 0.0016,
      "step": 4350
    },
    {
      "epoch": 12.945885841363973,
      "grad_norm": 0.06613746285438538,
      "learning_rate": 1.3955555555555556e-06,
      "loss": 0.0015,
      "step": 4375
    },
    {
      "epoch": 13.0177909562639,
      "grad_norm": 0.048592809587717056,
      "learning_rate": 1.34e-06,
      "loss": 0.0014,
      "step": 4400
    },
    {
      "epoch": 13.091919940696812,
      "grad_norm": 0.055479902774095535,
      "learning_rate": 1.2844444444444445e-06,
      "loss": 0.0014,
      "step": 4425
    },
    {
      "epoch": 13.166048925129726,
      "grad_norm": 0.045032110065221786,
      "learning_rate": 1.228888888888889e-06,
      "loss": 0.0014,
      "step": 4450
    },
    {
      "epoch": 13.24017790956264,
      "grad_norm": 0.06555484235286713,
      "learning_rate": 1.1733333333333335e-06,
      "loss": 0.0014,
      "step": 4475
    },
    {
      "epoch": 13.314306893995552,
      "grad_norm": 0.048962488770484924,
      "learning_rate": 1.117777777777778e-06,
      "loss": 0.0014,
      "step": 4500
    },
    {
      "epoch": 13.388435878428465,
      "grad_norm": 0.047078557312488556,
      "learning_rate": 1.0622222222222222e-06,
      "loss": 0.0014,
      "step": 4525
    },
    {
      "epoch": 13.46256486286138,
      "grad_norm": 0.050719983875751495,
      "learning_rate": 1.0066666666666668e-06,
      "loss": 0.0014,
      "step": 4550
    },
    {
      "epoch": 13.536693847294291,
      "grad_norm": 0.060897134244441986,
      "learning_rate": 9.511111111111111e-07,
      "loss": 0.0014,
      "step": 4575
    },
    {
      "epoch": 13.610822831727205,
      "grad_norm": 0.0608171783387661,
      "learning_rate": 8.955555555555557e-07,
      "loss": 0.0016,
      "step": 4600
    },
    {
      "epoch": 13.684951816160119,
      "grad_norm": 0.050377678126096725,
      "learning_rate": 8.400000000000001e-07,
      "loss": 0.0014,
      "step": 4625
    },
    {
      "epoch": 13.759080800593033,
      "grad_norm": 0.04982956126332283,
      "learning_rate": 7.844444444444445e-07,
      "loss": 0.0013,
      "step": 4650
    },
    {
      "epoch": 13.833209785025945,
      "grad_norm": 0.049559298902750015,
      "learning_rate": 7.28888888888889e-07,
      "loss": 0.0014,
      "step": 4675
    },
    {
      "epoch": 13.907338769458859,
      "grad_norm": 0.045101020485162735,
      "learning_rate": 6.733333333333334e-07,
      "loss": 0.0014,
      "step": 4700
    },
    {
      "epoch": 13.981467753891772,
      "grad_norm": 0.04930800572037697,
      "learning_rate": 6.177777777777778e-07,
      "loss": 0.0014,
      "step": 4725
    },
    {
      "epoch": 14.053372868791698,
      "grad_norm": 0.046362608671188354,
      "learning_rate": 5.622222222222223e-07,
      "loss": 0.0013,
      "step": 4750
    },
    {
      "epoch": 14.127501853224611,
      "grad_norm": 0.05173536762595177,
      "learning_rate": 5.066666666666667e-07,
      "loss": 0.0015,
      "step": 4775
    },
    {
      "epoch": 14.201630837657524,
      "grad_norm": 0.045450981706380844,
      "learning_rate": 4.511111111111111e-07,
      "loss": 0.0013,
      "step": 4800
    },
    {
      "epoch": 14.275759822090437,
      "grad_norm": 0.04437384009361267,
      "learning_rate": 3.9555555555555557e-07,
      "loss": 0.0013,
      "step": 4825
    },
    {
      "epoch": 14.349888806523351,
      "grad_norm": 0.048076264560222626,
      "learning_rate": 3.4000000000000003e-07,
      "loss": 0.0013,
      "step": 4850
    },
    {
      "epoch": 14.424017790956263,
      "grad_norm": 0.047441426664590836,
      "learning_rate": 2.844444444444445e-07,
      "loss": 0.0014,
      "step": 4875
    },
    {
      "epoch": 14.498146775389177,
      "grad_norm": 0.04609740898013115,
      "learning_rate": 2.2888888888888892e-07,
      "loss": 0.0013,
      "step": 4900
    },
    {
      "epoch": 14.572275759822091,
      "grad_norm": 0.05346500128507614,
      "learning_rate": 1.7333333333333335e-07,
      "loss": 0.0013,
      "step": 4925
    },
    {
      "epoch": 14.646404744255003,
      "grad_norm": 0.04578370600938797,
      "learning_rate": 1.1777777777777778e-07,
      "loss": 0.0013,
      "step": 4950
    },
    {
      "epoch": 14.720533728687917,
      "grad_norm": 0.045380473136901855,
      "learning_rate": 6.222222222222223e-08,
      "loss": 0.0014,
      "step": 4975
    },
    {
      "epoch": 14.79466271312083,
      "grad_norm": 0.04897478222846985,
      "learning_rate": 6.666666666666667e-09,
      "loss": 0.0013,
      "step": 5000
    },
    {
      "epoch": 14.79466271312083,
      "eval_loss": 0.4044080376625061,
      "eval_runtime": 939.8085,
      "eval_samples_per_second": 3.07,
      "eval_steps_per_second": 0.384,
      "eval_wer": 0.1675184165997231,
      "step": 5000
    },
    {
      "epoch": 14.79466271312083,
      "step": 5000,
      "total_flos": 4.607669973179538e+19,
      "train_loss": 0.11535389684215187,
      "train_runtime": 25285.0125,
      "train_samples_per_second": 6.328,
      "train_steps_per_second": 0.198
    }
  ],
  "logging_steps": 25,
  "max_steps": 5000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 15,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.607669973179538e+19,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}