{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.861003861003861,
  "eval_steps": 1000,
  "global_step": 5000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.019305019305019305,
      "grad_norm": 7.541379928588867,
      "learning_rate": 5.000000000000001e-07,
      "loss": 1.5011,
      "step": 25
    },
    {
      "epoch": 0.03861003861003861,
      "grad_norm": 5.699951648712158,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 1.1952,
      "step": 50
    },
    {
      "epoch": 0.05791505791505792,
      "grad_norm": 4.949509143829346,
      "learning_rate": 1.5e-06,
      "loss": 0.8322,
      "step": 75
    },
    {
      "epoch": 0.07722007722007722,
      "grad_norm": 4.566027641296387,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.7302,
      "step": 100
    },
    {
      "epoch": 0.09652509652509653,
      "grad_norm": 5.739590167999268,
      "learning_rate": 2.5e-06,
      "loss": 0.658,
      "step": 125
    },
    {
      "epoch": 0.11583011583011583,
      "grad_norm": 5.6211347579956055,
      "learning_rate": 3e-06,
      "loss": 0.6784,
      "step": 150
    },
    {
      "epoch": 0.13513513513513514,
      "grad_norm": 5.0315046310424805,
      "learning_rate": 3.5e-06,
      "loss": 0.6025,
      "step": 175
    },
    {
      "epoch": 0.15444015444015444,
      "grad_norm": 4.241130828857422,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.5837,
      "step": 200
    },
    {
      "epoch": 0.17374517374517376,
      "grad_norm": 4.572744369506836,
      "learning_rate": 4.5e-06,
      "loss": 0.5771,
      "step": 225
    },
    {
      "epoch": 0.19305019305019305,
      "grad_norm": 4.686110496520996,
      "learning_rate": 5e-06,
      "loss": 0.5647,
      "step": 250
    },
    {
      "epoch": 0.21235521235521235,
      "grad_norm": 4.762725353240967,
      "learning_rate": 5.500000000000001e-06,
      "loss": 0.5817,
      "step": 275
    },
    {
      "epoch": 0.23166023166023167,
      "grad_norm": 4.6283278465271,
      "learning_rate": 6e-06,
      "loss": 0.5409,
      "step": 300
    },
    {
      "epoch": 0.25096525096525096,
      "grad_norm": 3.748809576034546,
      "learning_rate": 6.5000000000000004e-06,
      "loss": 0.4916,
      "step": 325
    },
    {
      "epoch": 0.2702702702702703,
      "grad_norm": 4.640991687774658,
      "learning_rate": 7e-06,
      "loss": 0.5268,
      "step": 350
    },
    {
      "epoch": 0.28957528957528955,
      "grad_norm": 4.591019630432129,
      "learning_rate": 7.500000000000001e-06,
      "loss": 0.5422,
      "step": 375
    },
    {
      "epoch": 0.3088803088803089,
      "grad_norm": 5.206230640411377,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.5214,
      "step": 400
    },
    {
      "epoch": 0.3281853281853282,
      "grad_norm": 4.378481388092041,
      "learning_rate": 8.5e-06,
      "loss": 0.5085,
      "step": 425
    },
    {
      "epoch": 0.3474903474903475,
      "grad_norm": 4.319910526275635,
      "learning_rate": 9e-06,
      "loss": 0.4955,
      "step": 450
    },
    {
      "epoch": 0.3667953667953668,
      "grad_norm": 4.216291427612305,
      "learning_rate": 9.5e-06,
      "loss": 0.5112,
      "step": 475
    },
    {
      "epoch": 0.3861003861003861,
      "grad_norm": 4.130216121673584,
      "learning_rate": 1e-05,
      "loss": 0.4917,
      "step": 500
    },
    {
      "epoch": 0.40540540540540543,
      "grad_norm": 3.5388808250427246,
      "learning_rate": 9.944444444444445e-06,
      "loss": 0.4995,
      "step": 525
    },
    {
      "epoch": 0.4247104247104247,
      "grad_norm": 3.57562255859375,
      "learning_rate": 9.88888888888889e-06,
      "loss": 0.4747,
      "step": 550
    },
    {
      "epoch": 0.444015444015444,
      "grad_norm": 3.1933300495147705,
      "learning_rate": 9.833333333333333e-06,
      "loss": 0.4789,
      "step": 575
    },
    {
      "epoch": 0.46332046332046334,
      "grad_norm": 3.9434165954589844,
      "learning_rate": 9.777777777777779e-06,
      "loss": 0.4642,
      "step": 600
    },
    {
      "epoch": 0.4826254826254826,
      "grad_norm": 3.4227051734924316,
      "learning_rate": 9.722222222222223e-06,
      "loss": 0.4493,
      "step": 625
    },
    {
      "epoch": 0.5019305019305019,
      "grad_norm": 3.3387601375579834,
      "learning_rate": 9.666666666666667e-06,
      "loss": 0.4561,
      "step": 650
    },
    {
      "epoch": 0.5212355212355212,
      "grad_norm": 3.66536021232605,
      "learning_rate": 9.611111111111112e-06,
      "loss": 0.4703,
      "step": 675
    },
    {
      "epoch": 0.5405405405405406,
      "grad_norm": 3.9533724784851074,
      "learning_rate": 9.555555555555556e-06,
      "loss": 0.4602,
      "step": 700
    },
    {
      "epoch": 0.5598455598455598,
      "grad_norm": 3.853438377380371,
      "learning_rate": 9.5e-06,
      "loss": 0.4524,
      "step": 725
    },
    {
      "epoch": 0.5791505791505791,
      "grad_norm": 3.8207361698150635,
      "learning_rate": 9.444444444444445e-06,
      "loss": 0.4422,
      "step": 750
    },
    {
      "epoch": 0.5984555984555985,
      "grad_norm": 3.5014588832855225,
      "learning_rate": 9.38888888888889e-06,
      "loss": 0.4261,
      "step": 775
    },
    {
      "epoch": 0.6177606177606177,
      "grad_norm": 3.4164435863494873,
      "learning_rate": 9.333333333333334e-06,
      "loss": 0.4244,
      "step": 800
    },
    {
      "epoch": 0.637065637065637,
      "grad_norm": 3.5216405391693115,
      "learning_rate": 9.277777777777778e-06,
      "loss": 0.4406,
      "step": 825
    },
    {
      "epoch": 0.6563706563706564,
      "grad_norm": 3.0401480197906494,
      "learning_rate": 9.222222222222224e-06,
      "loss": 0.4047,
      "step": 850
    },
    {
      "epoch": 0.6756756756756757,
      "grad_norm": 3.4229395389556885,
      "learning_rate": 9.166666666666666e-06,
      "loss": 0.4205,
      "step": 875
    },
    {
      "epoch": 0.694980694980695,
      "grad_norm": 3.6540348529815674,
      "learning_rate": 9.111111111111112e-06,
      "loss": 0.4143,
      "step": 900
    },
    {
      "epoch": 0.7142857142857143,
      "grad_norm": 3.199246883392334,
      "learning_rate": 9.055555555555556e-06,
      "loss": 0.4062,
      "step": 925
    },
    {
      "epoch": 0.7335907335907336,
      "grad_norm": 3.220662832260132,
      "learning_rate": 9e-06,
      "loss": 0.405,
      "step": 950
    },
    {
      "epoch": 0.752895752895753,
      "grad_norm": 3.5012640953063965,
      "learning_rate": 8.944444444444446e-06,
      "loss": 0.3942,
      "step": 975
    },
    {
      "epoch": 0.7722007722007722,
      "grad_norm": 3.518545150756836,
      "learning_rate": 8.888888888888888e-06,
      "loss": 0.3936,
      "step": 1000
    },
    {
      "epoch": 0.7722007722007722,
      "eval_loss": 0.45976704359054565,
      "eval_runtime": 2976.7472,
      "eval_samples_per_second": 2.359,
      "eval_steps_per_second": 0.147,
      "eval_wer": 0.34102477535968423,
      "step": 1000
    },
    {
      "epoch": 0.7915057915057915,
      "grad_norm": 3.2509634494781494,
      "learning_rate": 8.833333333333334e-06,
      "loss": 0.4169,
      "step": 1025
    },
    {
      "epoch": 0.8108108108108109,
      "grad_norm": 2.5278899669647217,
      "learning_rate": 8.777777777777778e-06,
      "loss": 0.3895,
      "step": 1050
    },
    {
      "epoch": 0.8301158301158301,
      "grad_norm": 3.232598066329956,
      "learning_rate": 8.722222222222224e-06,
      "loss": 0.3782,
      "step": 1075
    },
    {
      "epoch": 0.8494208494208494,
      "grad_norm": 3.384092092514038,
      "learning_rate": 8.666666666666668e-06,
      "loss": 0.406,
      "step": 1100
    },
    {
      "epoch": 0.8687258687258688,
      "grad_norm": 3.261749267578125,
      "learning_rate": 8.611111111111112e-06,
      "loss": 0.4397,
      "step": 1125
    },
    {
      "epoch": 0.888030888030888,
      "grad_norm": 3.817667007446289,
      "learning_rate": 8.555555555555556e-06,
      "loss": 0.3846,
      "step": 1150
    },
    {
      "epoch": 0.9073359073359073,
      "grad_norm": 3.3195998668670654,
      "learning_rate": 8.5e-06,
      "loss": 0.3748,
      "step": 1175
    },
    {
      "epoch": 0.9266409266409267,
      "grad_norm": 3.510660171508789,
      "learning_rate": 8.444444444444446e-06,
      "loss": 0.3977,
      "step": 1200
    },
    {
      "epoch": 0.9459459459459459,
      "grad_norm": 2.8854782581329346,
      "learning_rate": 8.38888888888889e-06,
      "loss": 0.3768,
      "step": 1225
    },
    {
      "epoch": 0.9652509652509652,
      "grad_norm": 3.2596817016601562,
      "learning_rate": 8.333333333333334e-06,
      "loss": 0.3791,
      "step": 1250
    },
    {
      "epoch": 0.9845559845559846,
      "grad_norm": 2.9499335289001465,
      "learning_rate": 8.277777777777778e-06,
      "loss": 0.3839,
      "step": 1275
    },
    {
      "epoch": 1.0038610038610039,
      "grad_norm": 2.457566738128662,
      "learning_rate": 8.222222222222222e-06,
      "loss": 0.3371,
      "step": 1300
    },
    {
      "epoch": 1.0231660231660231,
      "grad_norm": 2.8226237297058105,
      "learning_rate": 8.166666666666668e-06,
      "loss": 0.2705,
      "step": 1325
    },
    {
      "epoch": 1.0424710424710424,
      "grad_norm": 3.664156198501587,
      "learning_rate": 8.111111111111112e-06,
      "loss": 0.2729,
      "step": 1350
    },
    {
      "epoch": 1.0617760617760619,
      "grad_norm": 2.497749090194702,
      "learning_rate": 8.055555555555557e-06,
      "loss": 0.2613,
      "step": 1375
    },
    {
      "epoch": 1.0810810810810811,
      "grad_norm": 2.437830686569214,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.2718,
      "step": 1400
    },
    {
      "epoch": 1.1003861003861004,
      "grad_norm": 2.5171914100646973,
      "learning_rate": 7.944444444444445e-06,
      "loss": 0.2649,
      "step": 1425
    },
    {
      "epoch": 1.1196911196911197,
      "grad_norm": 3.023686647415161,
      "learning_rate": 7.88888888888889e-06,
      "loss": 0.2662,
      "step": 1450
    },
    {
      "epoch": 1.138996138996139,
      "grad_norm": 2.358494520187378,
      "learning_rate": 7.833333333333333e-06,
      "loss": 0.2511,
      "step": 1475
    },
    {
      "epoch": 1.1583011583011582,
      "grad_norm": 2.428818941116333,
      "learning_rate": 7.77777777777778e-06,
      "loss": 0.2626,
      "step": 1500
    },
    {
      "epoch": 1.1776061776061777,
      "grad_norm": 3.066359281539917,
      "learning_rate": 7.722222222222223e-06,
      "loss": 0.2504,
      "step": 1525
    },
    {
      "epoch": 1.196911196911197,
      "grad_norm": 2.8853089809417725,
      "learning_rate": 7.666666666666667e-06,
      "loss": 0.2565,
      "step": 1550
    },
    {
      "epoch": 1.2162162162162162,
      "grad_norm": 2.476994514465332,
      "learning_rate": 7.611111111111111e-06,
      "loss": 0.274,
      "step": 1575
    },
    {
      "epoch": 1.2355212355212355,
      "grad_norm": 3.0519564151763916,
      "learning_rate": 7.555555555555556e-06,
      "loss": 0.2604,
      "step": 1600
    },
    {
      "epoch": 1.2548262548262548,
      "grad_norm": 2.985853910446167,
      "learning_rate": 7.500000000000001e-06,
      "loss": 0.2832,
      "step": 1625
    },
    {
      "epoch": 1.2741312741312742,
      "grad_norm": 2.780881643295288,
      "learning_rate": 7.444444444444445e-06,
      "loss": 0.2743,
      "step": 1650
    },
    {
      "epoch": 1.2934362934362935,
      "grad_norm": 2.6131482124328613,
      "learning_rate": 7.38888888888889e-06,
      "loss": 0.2483,
      "step": 1675
    },
    {
      "epoch": 1.3127413127413128,
      "grad_norm": 3.0259549617767334,
      "learning_rate": 7.333333333333333e-06,
      "loss": 0.2686,
      "step": 1700
    },
    {
      "epoch": 1.332046332046332,
      "grad_norm": 2.420754909515381,
      "learning_rate": 7.277777777777778e-06,
      "loss": 0.265,
      "step": 1725
    },
    {
      "epoch": 1.3513513513513513,
      "grad_norm": 3.001450300216675,
      "learning_rate": 7.222222222222223e-06,
      "loss": 0.2772,
      "step": 1750
    },
    {
      "epoch": 1.3706563706563706,
      "grad_norm": 3.064401626586914,
      "learning_rate": 7.166666666666667e-06,
      "loss": 0.2844,
      "step": 1775
    },
    {
      "epoch": 1.3899613899613898,
      "grad_norm": 2.651357412338257,
      "learning_rate": 7.111111111111112e-06,
      "loss": 0.2681,
      "step": 1800
    },
    {
      "epoch": 1.4092664092664093,
      "grad_norm": 2.363473415374756,
      "learning_rate": 7.055555555555557e-06,
      "loss": 0.2567,
      "step": 1825
    },
    {
      "epoch": 1.4285714285714286,
      "grad_norm": 3.044689178466797,
      "learning_rate": 7e-06,
      "loss": 0.2566,
      "step": 1850
    },
    {
      "epoch": 1.4478764478764479,
      "grad_norm": 2.3913726806640625,
      "learning_rate": 6.944444444444445e-06,
      "loss": 0.2506,
      "step": 1875
    },
    {
      "epoch": 1.4671814671814671,
      "grad_norm": 2.2928853034973145,
      "learning_rate": 6.88888888888889e-06,
      "loss": 0.2727,
      "step": 1900
    },
    {
      "epoch": 1.4864864864864864,
      "grad_norm": 2.741959571838379,
      "learning_rate": 6.833333333333334e-06,
      "loss": 0.2467,
      "step": 1925
    },
    {
      "epoch": 1.505791505791506,
      "grad_norm": 2.0824573040008545,
      "learning_rate": 6.777777777777779e-06,
      "loss": 0.2635,
      "step": 1950
    },
    {
      "epoch": 1.525096525096525,
      "grad_norm": 2.4026057720184326,
      "learning_rate": 6.7222222222222235e-06,
      "loss": 0.2494,
      "step": 1975
    },
    {
      "epoch": 1.5444015444015444,
      "grad_norm": 3.137629270553589,
      "learning_rate": 6.666666666666667e-06,
      "loss": 0.2493,
      "step": 2000
    },
    {
      "epoch": 1.5444015444015444,
      "eval_loss": 0.40415582060813904,
      "eval_runtime": 2884.198,
      "eval_samples_per_second": 2.435,
      "eval_steps_per_second": 0.152,
      "eval_wer": 0.29395418895756503,
      "step": 2000
    },
    {
      "epoch": 1.5637065637065637,
      "grad_norm": 2.6668949127197266,
      "learning_rate": 6.6111111111111115e-06,
      "loss": 0.2522,
      "step": 2025
    },
    {
      "epoch": 1.583011583011583,
      "grad_norm": 2.5405499935150146,
      "learning_rate": 6.555555555555556e-06,
      "loss": 0.2747,
      "step": 2050
    },
    {
      "epoch": 1.6023166023166024,
      "grad_norm": 2.4073848724365234,
      "learning_rate": 6.5000000000000004e-06,
      "loss": 0.2563,
      "step": 2075
    },
    {
      "epoch": 1.6216216216216215,
      "grad_norm": 3.3368024826049805,
      "learning_rate": 6.444444444444445e-06,
      "loss": 0.2477,
      "step": 2100
    },
    {
      "epoch": 1.640926640926641,
      "grad_norm": 2.741755723953247,
      "learning_rate": 6.3888888888888885e-06,
      "loss": 0.2444,
      "step": 2125
    },
    {
      "epoch": 1.6602316602316602,
      "grad_norm": 2.5123753547668457,
      "learning_rate": 6.333333333333333e-06,
      "loss": 0.253,
      "step": 2150
    },
    {
      "epoch": 1.6795366795366795,
      "grad_norm": 2.8450229167938232,
      "learning_rate": 6.277777777777778e-06,
      "loss": 0.2695,
      "step": 2175
    },
    {
      "epoch": 1.698841698841699,
      "grad_norm": 2.2329864501953125,
      "learning_rate": 6.222222222222223e-06,
      "loss": 0.2531,
      "step": 2200
    },
    {
      "epoch": 1.718146718146718,
      "grad_norm": 2.8518948554992676,
      "learning_rate": 6.166666666666667e-06,
      "loss": 0.2671,
      "step": 2225
    },
    {
      "epoch": 1.7374517374517375,
      "grad_norm": 2.6565253734588623,
      "learning_rate": 6.111111111111112e-06,
      "loss": 0.2477,
      "step": 2250
    },
    {
      "epoch": 1.7567567567567568,
      "grad_norm": 2.2332699298858643,
      "learning_rate": 6.055555555555555e-06,
      "loss": 0.2461,
      "step": 2275
    },
    {
      "epoch": 1.776061776061776,
      "grad_norm": 2.5508103370666504,
      "learning_rate": 6e-06,
      "loss": 0.2574,
      "step": 2300
    },
    {
      "epoch": 1.7953667953667953,
      "grad_norm": 2.47121000289917,
      "learning_rate": 5.944444444444445e-06,
      "loss": 0.2519,
      "step": 2325
    },
    {
      "epoch": 1.8146718146718146,
      "grad_norm": 2.798379898071289,
      "learning_rate": 5.88888888888889e-06,
      "loss": 0.2774,
      "step": 2350
    },
    {
      "epoch": 1.833976833976834,
      "grad_norm": 2.6081783771514893,
      "learning_rate": 5.833333333333334e-06,
      "loss": 0.2271,
      "step": 2375
    },
    {
      "epoch": 1.8532818532818531,
      "grad_norm": 2.355163335800171,
      "learning_rate": 5.777777777777778e-06,
      "loss": 0.248,
      "step": 2400
    },
    {
      "epoch": 1.8725868725868726,
      "grad_norm": 2.8276679515838623,
      "learning_rate": 5.722222222222222e-06,
      "loss": 0.2575,
      "step": 2425
    },
    {
      "epoch": 1.8918918918918919,
      "grad_norm": 2.9946837425231934,
      "learning_rate": 5.666666666666667e-06,
      "loss": 0.234,
      "step": 2450
    },
    {
      "epoch": 1.9111969111969112,
      "grad_norm": 2.660792350769043,
      "learning_rate": 5.611111111111112e-06,
      "loss": 0.2373,
      "step": 2475
    },
    {
      "epoch": 1.9305019305019306,
      "grad_norm": 2.4244725704193115,
      "learning_rate": 5.555555555555557e-06,
      "loss": 0.2443,
      "step": 2500
    },
    {
      "epoch": 1.9498069498069497,
      "grad_norm": 2.6997570991516113,
      "learning_rate": 5.500000000000001e-06,
      "loss": 0.2553,
      "step": 2525
    },
    {
      "epoch": 1.9691119691119692,
      "grad_norm": 2.6614644527435303,
      "learning_rate": 5.444444444444445e-06,
      "loss": 0.2526,
      "step": 2550
    },
    {
      "epoch": 1.9884169884169884,
      "grad_norm": 2.8852739334106445,
      "learning_rate": 5.388888888888889e-06,
      "loss": 0.2363,
      "step": 2575
    },
    {
      "epoch": 2.0077220077220077,
      "grad_norm": 1.8691250085830688,
      "learning_rate": 5.333333333333334e-06,
      "loss": 0.2197,
      "step": 2600
    },
    {
      "epoch": 2.027027027027027,
      "grad_norm": 1.8070666790008545,
      "learning_rate": 5.2777777777777785e-06,
      "loss": 0.1556,
      "step": 2625
    },
    {
      "epoch": 2.0463320463320462,
      "grad_norm": 2.2826385498046875,
      "learning_rate": 5.2222222222222226e-06,
      "loss": 0.1476,
      "step": 2650
    },
    {
      "epoch": 2.0656370656370657,
      "grad_norm": 2.1625537872314453,
      "learning_rate": 5.1666666666666675e-06,
      "loss": 0.1628,
      "step": 2675
    },
    {
      "epoch": 2.0849420849420848,
      "grad_norm": 2.158252000808716,
      "learning_rate": 5.1111111111111115e-06,
      "loss": 0.1353,
      "step": 2700
    },
    {
      "epoch": 2.1042471042471043,
      "grad_norm": 2.2190427780151367,
      "learning_rate": 5.0555555555555555e-06,
      "loss": 0.1697,
      "step": 2725
    },
    {
      "epoch": 2.1235521235521237,
      "grad_norm": 1.9213645458221436,
      "learning_rate": 5e-06,
      "loss": 0.1535,
      "step": 2750
    },
    {
      "epoch": 2.142857142857143,
      "grad_norm": 2.329991340637207,
      "learning_rate": 4.944444444444445e-06,
      "loss": 0.1582,
      "step": 2775
    },
    {
      "epoch": 2.1621621621621623,
      "grad_norm": 2.3484299182891846,
      "learning_rate": 4.888888888888889e-06,
      "loss": 0.1464,
      "step": 2800
    },
    {
      "epoch": 2.1814671814671813,
      "grad_norm": 2.418144941329956,
      "learning_rate": 4.833333333333333e-06,
      "loss": 0.1557,
      "step": 2825
    },
    {
      "epoch": 2.200772200772201,
      "grad_norm": 2.182532787322998,
      "learning_rate": 4.777777777777778e-06,
      "loss": 0.1463,
      "step": 2850
    },
    {
      "epoch": 2.2200772200772203,
      "grad_norm": 1.8456135988235474,
      "learning_rate": 4.722222222222222e-06,
      "loss": 0.1497,
      "step": 2875
    },
    {
      "epoch": 2.2393822393822393,
      "grad_norm": 2.3882248401641846,
      "learning_rate": 4.666666666666667e-06,
      "loss": 0.1487,
      "step": 2900
    },
    {
      "epoch": 2.258687258687259,
      "grad_norm": 2.5126445293426514,
      "learning_rate": 4.611111111111112e-06,
      "loss": 0.1577,
      "step": 2925
    },
    {
      "epoch": 2.277992277992278,
      "grad_norm": 2.2082295417785645,
      "learning_rate": 4.555555555555556e-06,
      "loss": 0.1491,
      "step": 2950
    },
    {
      "epoch": 2.2972972972972974,
      "grad_norm": 2.6382830142974854,
      "learning_rate": 4.5e-06,
      "loss": 0.1428,
      "step": 2975
    },
    {
      "epoch": 2.3166023166023164,
      "grad_norm": 2.447270631790161,
      "learning_rate": 4.444444444444444e-06,
      "loss": 0.1548,
      "step": 3000
    },
    {
      "epoch": 2.3166023166023164,
      "eval_loss": 0.39975878596305847,
      "eval_runtime": 2877.301,
      "eval_samples_per_second": 2.44,
      "eval_steps_per_second": 0.153,
      "eval_wer": 0.28095621461590403,
      "step": 3000
    },
    {
      "epoch": 2.335907335907336,
      "grad_norm": 1.664686679840088,
      "learning_rate": 4.388888888888889e-06,
      "loss": 0.1407,
      "step": 3025
    },
    {
      "epoch": 2.3552123552123554,
      "grad_norm": 2.261843681335449,
      "learning_rate": 4.333333333333334e-06,
      "loss": 0.1511,
      "step": 3050
    },
    {
      "epoch": 2.3745173745173744,
      "grad_norm": 2.157227039337158,
      "learning_rate": 4.277777777777778e-06,
      "loss": 0.1433,
      "step": 3075
    },
    {
      "epoch": 2.393822393822394,
      "grad_norm": 2.434943199157715,
      "learning_rate": 4.222222222222223e-06,
      "loss": 0.1476,
      "step": 3100
    },
    {
      "epoch": 2.413127413127413,
      "grad_norm": 2.106074333190918,
      "learning_rate": 4.166666666666667e-06,
      "loss": 0.1496,
      "step": 3125
    },
    {
      "epoch": 2.4324324324324325,
      "grad_norm": 2.263333320617676,
      "learning_rate": 4.111111111111111e-06,
      "loss": 0.1409,
      "step": 3150
    },
    {
      "epoch": 2.4517374517374515,
      "grad_norm": 2.9890952110290527,
      "learning_rate": 4.055555555555556e-06,
      "loss": 0.1497,
      "step": 3175
    },
    {
      "epoch": 2.471042471042471,
      "grad_norm": 2.174651622772217,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.1418,
      "step": 3200
    },
    {
      "epoch": 2.4903474903474905,
      "grad_norm": 2.380537509918213,
      "learning_rate": 3.944444444444445e-06,
      "loss": 0.1369,
      "step": 3225
    },
    {
      "epoch": 2.5096525096525095,
      "grad_norm": 2.1118268966674805,
      "learning_rate": 3.88888888888889e-06,
      "loss": 0.1465,
      "step": 3250
    },
    {
      "epoch": 2.528957528957529,
      "grad_norm": 2.221266746520996,
      "learning_rate": 3.833333333333334e-06,
      "loss": 0.1392,
      "step": 3275
    },
    {
      "epoch": 2.5482625482625485,
      "grad_norm": 2.216095209121704,
      "learning_rate": 3.777777777777778e-06,
      "loss": 0.143,
      "step": 3300
    },
    {
      "epoch": 2.5675675675675675,
      "grad_norm": 2.23201322555542,
      "learning_rate": 3.7222222222222225e-06,
      "loss": 0.1399,
      "step": 3325
    },
    {
      "epoch": 2.586872586872587,
      "grad_norm": 2.174283742904663,
      "learning_rate": 3.6666666666666666e-06,
      "loss": 0.1531,
      "step": 3350
    },
    {
      "epoch": 2.606177606177606,
      "grad_norm": 2.455362558364868,
      "learning_rate": 3.6111111111111115e-06,
      "loss": 0.1466,
      "step": 3375
    },
    {
      "epoch": 2.6254826254826256,
      "grad_norm": 2.6259799003601074,
      "learning_rate": 3.555555555555556e-06,
      "loss": 0.1458,
      "step": 3400
    },
    {
      "epoch": 2.6447876447876446,
      "grad_norm": 2.1615617275238037,
      "learning_rate": 3.5e-06,
      "loss": 0.145,
      "step": 3425
    },
    {
      "epoch": 2.664092664092664,
      "grad_norm": 2.36684513092041,
      "learning_rate": 3.444444444444445e-06,
      "loss": 0.1462,
      "step": 3450
    },
    {
      "epoch": 2.6833976833976836,
      "grad_norm": 2.027125835418701,
      "learning_rate": 3.3888888888888893e-06,
      "loss": 0.1445,
      "step": 3475
    },
    {
      "epoch": 2.7027027027027026,
      "grad_norm": 2.1827645301818848,
      "learning_rate": 3.3333333333333333e-06,
      "loss": 0.1466,
      "step": 3500
    },
    {
      "epoch": 2.722007722007722,
      "grad_norm": 2.1014013290405273,
      "learning_rate": 3.277777777777778e-06,
      "loss": 0.1453,
      "step": 3525
    },
    {
      "epoch": 2.741312741312741,
      "grad_norm": 1.92685866355896,
      "learning_rate": 3.2222222222222227e-06,
      "loss": 0.1341,
      "step": 3550
    },
    {
      "epoch": 2.7606177606177607,
      "grad_norm": 2.028932571411133,
      "learning_rate": 3.1666666666666667e-06,
      "loss": 0.1399,
      "step": 3575
    },
    {
      "epoch": 2.7799227799227797,
      "grad_norm": 1.9122258424758911,
      "learning_rate": 3.1111111111111116e-06,
      "loss": 0.1433,
      "step": 3600
    },
    {
      "epoch": 2.799227799227799,
      "grad_norm": 2.2335093021392822,
      "learning_rate": 3.055555555555556e-06,
      "loss": 0.1438,
      "step": 3625
    },
    {
      "epoch": 2.8185328185328187,
      "grad_norm": 2.2860329151153564,
      "learning_rate": 3e-06,
      "loss": 0.1392,
      "step": 3650
    },
    {
      "epoch": 2.8378378378378377,
      "grad_norm": 1.873155951499939,
      "learning_rate": 2.944444444444445e-06,
      "loss": 0.1396,
      "step": 3675
    },
    {
      "epoch": 2.857142857142857,
      "grad_norm": 2.6895735263824463,
      "learning_rate": 2.888888888888889e-06,
      "loss": 0.1419,
      "step": 3700
    },
    {
      "epoch": 2.8764478764478767,
      "grad_norm": 1.6812546253204346,
      "learning_rate": 2.8333333333333335e-06,
      "loss": 0.138,
      "step": 3725
    },
    {
      "epoch": 2.8957528957528957,
      "grad_norm": 2.893087387084961,
      "learning_rate": 2.7777777777777783e-06,
      "loss": 0.1479,
      "step": 3750
    },
    {
      "epoch": 2.915057915057915,
      "grad_norm": 2.508882999420166,
      "learning_rate": 2.7222222222222224e-06,
      "loss": 0.1427,
      "step": 3775
    },
    {
      "epoch": 2.9343629343629343,
      "grad_norm": 2.536713123321533,
      "learning_rate": 2.666666666666667e-06,
      "loss": 0.1476,
      "step": 3800
    },
    {
      "epoch": 2.9536679536679538,
      "grad_norm": 2.256779432296753,
      "learning_rate": 2.6111111111111113e-06,
      "loss": 0.1357,
      "step": 3825
    },
    {
      "epoch": 2.972972972972973,
      "grad_norm": 2.0813608169555664,
      "learning_rate": 2.5555555555555557e-06,
      "loss": 0.1362,
      "step": 3850
    },
    {
      "epoch": 2.9922779922779923,
      "grad_norm": 2.6842668056488037,
      "learning_rate": 2.5e-06,
      "loss": 0.1527,
      "step": 3875
    },
    {
      "epoch": 3.011583011583012,
      "grad_norm": 1.5669656991958618,
      "learning_rate": 2.4444444444444447e-06,
      "loss": 0.1045,
      "step": 3900
    },
    {
      "epoch": 3.030888030888031,
      "grad_norm": 1.9614019393920898,
      "learning_rate": 2.388888888888889e-06,
      "loss": 0.0861,
      "step": 3925
    },
    {
      "epoch": 3.0501930501930503,
      "grad_norm": 1.6240154504776,
      "learning_rate": 2.3333333333333336e-06,
      "loss": 0.0811,
      "step": 3950
    },
    {
      "epoch": 3.0694980694980694,
      "grad_norm": 1.675820231437683,
      "learning_rate": 2.277777777777778e-06,
      "loss": 0.0808,
      "step": 3975
    },
    {
      "epoch": 3.088803088803089,
      "grad_norm": 1.5205894708633423,
      "learning_rate": 2.222222222222222e-06,
      "loss": 0.0755,
      "step": 4000
    },
    {
      "epoch": 3.088803088803089,
      "eval_loss": 0.42321887612342834,
      "eval_runtime": 2878.9611,
      "eval_samples_per_second": 2.439,
      "eval_steps_per_second": 0.152,
      "eval_wer": 0.27565833896016206,
      "step": 4000
    },
    {
      "epoch": 3.108108108108108,
      "grad_norm": 1.8952137231826782,
      "learning_rate": 2.166666666666667e-06,
      "loss": 0.0774,
      "step": 4025
    },
    {
      "epoch": 3.1274131274131274,
      "grad_norm": 1.7920564413070679,
      "learning_rate": 2.1111111111111114e-06,
      "loss": 0.0796,
      "step": 4050
    },
    {
      "epoch": 3.146718146718147,
      "grad_norm": 1.4591903686523438,
      "learning_rate": 2.0555555555555555e-06,
      "loss": 0.0784,
      "step": 4075
    },
    {
      "epoch": 3.166023166023166,
      "grad_norm": 1.7589879035949707,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.08,
      "step": 4100
    },
    {
      "epoch": 3.1853281853281854,
      "grad_norm": 2.2422068119049072,
      "learning_rate": 1.944444444444445e-06,
      "loss": 0.0784,
      "step": 4125
    },
    {
      "epoch": 3.2046332046332044,
      "grad_norm": 2.1973836421966553,
      "learning_rate": 1.888888888888889e-06,
      "loss": 0.0747,
      "step": 4150
    },
    {
      "epoch": 3.223938223938224,
      "grad_norm": 1.9121280908584595,
      "learning_rate": 1.8333333333333333e-06,
      "loss": 0.08,
      "step": 4175
    },
    {
      "epoch": 3.2432432432432434,
      "grad_norm": 1.950451135635376,
      "learning_rate": 1.777777777777778e-06,
      "loss": 0.0787,
      "step": 4200
    },
    {
      "epoch": 3.2625482625482625,
      "grad_norm": 1.5729962587356567,
      "learning_rate": 1.7222222222222224e-06,
      "loss": 0.0708,
      "step": 4225
    },
    {
      "epoch": 3.281853281853282,
      "grad_norm": 1.7916944026947021,
      "learning_rate": 1.6666666666666667e-06,
      "loss": 0.074,
      "step": 4250
    },
    {
      "epoch": 3.301158301158301,
      "grad_norm": 2.091259717941284,
      "learning_rate": 1.6111111111111113e-06,
      "loss": 0.0734,
      "step": 4275
    },
    {
      "epoch": 3.3204633204633205,
      "grad_norm": 2.2973198890686035,
      "learning_rate": 1.5555555555555558e-06,
      "loss": 0.0763,
      "step": 4300
    },
    {
      "epoch": 3.33976833976834,
      "grad_norm": 2.0894815921783447,
      "learning_rate": 1.5e-06,
      "loss": 0.0748,
      "step": 4325
    },
    {
      "epoch": 3.359073359073359,
      "grad_norm": 1.7151942253112793,
      "learning_rate": 1.4444444444444445e-06,
      "loss": 0.0719,
      "step": 4350
    },
    {
      "epoch": 3.3783783783783785,
      "grad_norm": 2.2781429290771484,
      "learning_rate": 1.3888888888888892e-06,
      "loss": 0.0795,
      "step": 4375
    },
    {
      "epoch": 3.3976833976833976,
      "grad_norm": 1.7880635261535645,
      "learning_rate": 1.3333333333333334e-06,
      "loss": 0.0765,
      "step": 4400
    },
    {
      "epoch": 3.416988416988417,
      "grad_norm": 1.8607178926467896,
      "learning_rate": 1.2777777777777779e-06,
      "loss": 0.0806,
      "step": 4425
    },
    {
      "epoch": 3.436293436293436,
      "grad_norm": 1.5629892349243164,
      "learning_rate": 1.2222222222222223e-06,
      "loss": 0.0731,
      "step": 4450
    },
    {
      "epoch": 3.4555984555984556,
      "grad_norm": 1.976099967956543,
      "learning_rate": 1.1666666666666668e-06,
      "loss": 0.0727,
      "step": 4475
    },
    {
      "epoch": 3.474903474903475,
      "grad_norm": 1.7316856384277344,
      "learning_rate": 1.111111111111111e-06,
      "loss": 0.078,
      "step": 4500
    },
    {
      "epoch": 3.494208494208494,
      "grad_norm": 1.796291470527649,
      "learning_rate": 1.0555555555555557e-06,
      "loss": 0.0653,
      "step": 4525
    },
    {
      "epoch": 3.5135135135135136,
      "grad_norm": 1.6075197458267212,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.0788,
      "step": 4550
    },
    {
      "epoch": 3.532818532818533,
      "grad_norm": 1.6467320919036865,
      "learning_rate": 9.444444444444445e-07,
      "loss": 0.0695,
      "step": 4575
    },
    {
      "epoch": 3.552123552123552,
      "grad_norm": 1.1778756380081177,
      "learning_rate": 8.88888888888889e-07,
      "loss": 0.0683,
      "step": 4600
    },
    {
      "epoch": 3.571428571428571,
      "grad_norm": 1.402613878250122,
      "learning_rate": 8.333333333333333e-07,
      "loss": 0.0698,
      "step": 4625
    },
    {
      "epoch": 3.5907335907335907,
      "grad_norm": 2.3424105644226074,
      "learning_rate": 7.777777777777779e-07,
      "loss": 0.0788,
      "step": 4650
    },
    {
      "epoch": 3.61003861003861,
      "grad_norm": 1.551446795463562,
      "learning_rate": 7.222222222222222e-07,
      "loss": 0.0765,
      "step": 4675
    },
    {
      "epoch": 3.629343629343629,
      "grad_norm": 2.174612045288086,
      "learning_rate": 6.666666666666667e-07,
      "loss": 0.0759,
      "step": 4700
    },
    {
      "epoch": 3.6486486486486487,
      "grad_norm": 1.595508337020874,
      "learning_rate": 6.111111111111112e-07,
      "loss": 0.0783,
      "step": 4725
    },
    {
      "epoch": 3.667953667953668,
      "grad_norm": 1.6800012588500977,
      "learning_rate": 5.555555555555555e-07,
      "loss": 0.0768,
      "step": 4750
    },
    {
      "epoch": 3.687258687258687,
      "grad_norm": 2.050546884536743,
      "learning_rate": 5.000000000000001e-07,
      "loss": 0.0736,
      "step": 4775
    },
    {
      "epoch": 3.7065637065637067,
      "grad_norm": 1.9666138887405396,
      "learning_rate": 4.444444444444445e-07,
      "loss": 0.0735,
      "step": 4800
    },
    {
      "epoch": 3.7258687258687258,
      "grad_norm": 2.0180180072784424,
      "learning_rate": 3.8888888888888895e-07,
      "loss": 0.071,
      "step": 4825
    },
    {
      "epoch": 3.7451737451737452,
      "grad_norm": 2.1212894916534424,
      "learning_rate": 3.3333333333333335e-07,
      "loss": 0.0907,
      "step": 4850
    },
    {
      "epoch": 3.7644787644787643,
      "grad_norm": 1.6568444967269897,
      "learning_rate": 2.7777777777777776e-07,
      "loss": 0.0746,
      "step": 4875
    },
    {
      "epoch": 3.7837837837837838,
      "grad_norm": 2.122925281524658,
      "learning_rate": 2.2222222222222224e-07,
      "loss": 0.0734,
      "step": 4900
    },
    {
      "epoch": 3.8030888030888033,
      "grad_norm": 1.5614521503448486,
      "learning_rate": 1.6666666666666668e-07,
      "loss": 0.0737,
      "step": 4925
    },
    {
      "epoch": 3.8223938223938223,
      "grad_norm": 3.5003342628479004,
      "learning_rate": 1.1111111111111112e-07,
      "loss": 0.0743,
      "step": 4950
    },
    {
      "epoch": 3.841698841698842,
      "grad_norm": 1.8813285827636719,
      "learning_rate": 5.555555555555556e-08,
      "loss": 0.0712,
      "step": 4975
    },
    {
      "epoch": 3.861003861003861,
      "grad_norm": 1.836345911026001,
      "learning_rate": 0.0,
      "loss": 0.0767,
      "step": 5000
    },
    {
      "epoch": 3.861003861003861,
      "eval_loss": 0.4226454198360443,
      "eval_runtime": 2912.2253,
      "eval_samples_per_second": 2.411,
      "eval_steps_per_second": 0.151,
      "eval_wer": 0.2671921259024568,
      "step": 5000
    },
    {
      "epoch": 3.861003861003861,
      "step": 5000,
      "total_flos": 5.435487665750016e+20,
      "train_loss": 0.2543565913915634,
      "train_runtime": 60445.9813,
      "train_samples_per_second": 2.647,
      "train_steps_per_second": 0.083
    }
  ],
  "logging_steps": 25,
  "max_steps": 5000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "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": 5.435487665750016e+20,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}