{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.7056277056277054,
  "eval_steps": 1000,
  "global_step": 5000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.013528138528138528,
      "grad_norm": 18.23776626586914,
      "learning_rate": 4.800000000000001e-07,
      "loss": 1.7881,
      "step": 25
    },
    {
      "epoch": 0.027056277056277056,
      "grad_norm": 11.490796089172363,
      "learning_rate": 9.800000000000001e-07,
      "loss": 1.4477,
      "step": 50
    },
    {
      "epoch": 0.040584415584415584,
      "grad_norm": 10.601150512695312,
      "learning_rate": 1.48e-06,
      "loss": 1.0989,
      "step": 75
    },
    {
      "epoch": 0.05411255411255411,
      "grad_norm": 12.294251441955566,
      "learning_rate": 1.98e-06,
      "loss": 0.9367,
      "step": 100
    },
    {
      "epoch": 0.06764069264069264,
      "grad_norm": 11.920494079589844,
      "learning_rate": 2.4800000000000004e-06,
      "loss": 0.8698,
      "step": 125
    },
    {
      "epoch": 0.08116883116883117,
      "grad_norm": 11.758705139160156,
      "learning_rate": 2.9800000000000003e-06,
      "loss": 0.8531,
      "step": 150
    },
    {
      "epoch": 0.0946969696969697,
      "grad_norm": 11.037556648254395,
      "learning_rate": 3.48e-06,
      "loss": 0.812,
      "step": 175
    },
    {
      "epoch": 0.10822510822510822,
      "grad_norm": 10.065262794494629,
      "learning_rate": 3.980000000000001e-06,
      "loss": 0.7987,
      "step": 200
    },
    {
      "epoch": 0.12175324675324675,
      "grad_norm": 9.124336242675781,
      "learning_rate": 4.48e-06,
      "loss": 0.7455,
      "step": 225
    },
    {
      "epoch": 0.13528138528138528,
      "grad_norm": 10.971399307250977,
      "learning_rate": 4.980000000000001e-06,
      "loss": 0.7564,
      "step": 250
    },
    {
      "epoch": 0.1488095238095238,
      "grad_norm": 9.226423263549805,
      "learning_rate": 5.480000000000001e-06,
      "loss": 0.7163,
      "step": 275
    },
    {
      "epoch": 0.16233766233766234,
      "grad_norm": 9.523015022277832,
      "learning_rate": 5.98e-06,
      "loss": 0.7341,
      "step": 300
    },
    {
      "epoch": 0.17586580086580086,
      "grad_norm": 8.390256881713867,
      "learning_rate": 6.480000000000001e-06,
      "loss": 0.7301,
      "step": 325
    },
    {
      "epoch": 0.1893939393939394,
      "grad_norm": 9.996743202209473,
      "learning_rate": 6.98e-06,
      "loss": 0.6897,
      "step": 350
    },
    {
      "epoch": 0.20292207792207792,
      "grad_norm": 9.470787048339844,
      "learning_rate": 7.48e-06,
      "loss": 0.6728,
      "step": 375
    },
    {
      "epoch": 0.21645021645021645,
      "grad_norm": 8.221435546875,
      "learning_rate": 7.980000000000002e-06,
      "loss": 0.6853,
      "step": 400
    },
    {
      "epoch": 0.22997835497835498,
      "grad_norm": 9.243407249450684,
      "learning_rate": 8.48e-06,
      "loss": 0.6899,
      "step": 425
    },
    {
      "epoch": 0.2435064935064935,
      "grad_norm": 8.308032989501953,
      "learning_rate": 8.96e-06,
      "loss": 0.634,
      "step": 450
    },
    {
      "epoch": 0.25703463203463206,
      "grad_norm": 8.970362663269043,
      "learning_rate": 9.460000000000001e-06,
      "loss": 0.6191,
      "step": 475
    },
    {
      "epoch": 0.27056277056277056,
      "grad_norm": 9.167222023010254,
      "learning_rate": 9.960000000000001e-06,
      "loss": 0.6348,
      "step": 500
    },
    {
      "epoch": 0.2840909090909091,
      "grad_norm": 8.824662208557129,
      "learning_rate": 9.94888888888889e-06,
      "loss": 0.6228,
      "step": 525
    },
    {
      "epoch": 0.2976190476190476,
      "grad_norm": 10.600458145141602,
      "learning_rate": 9.893333333333334e-06,
      "loss": 0.622,
      "step": 550
    },
    {
      "epoch": 0.31114718614718617,
      "grad_norm": 10.680913925170898,
      "learning_rate": 9.837777777777778e-06,
      "loss": 0.6115,
      "step": 575
    },
    {
      "epoch": 0.3246753246753247,
      "grad_norm": 8.21044635772705,
      "learning_rate": 9.782222222222222e-06,
      "loss": 0.6219,
      "step": 600
    },
    {
      "epoch": 0.33820346320346323,
      "grad_norm": 8.558358192443848,
      "learning_rate": 9.726666666666668e-06,
      "loss": 0.5936,
      "step": 625
    },
    {
      "epoch": 0.35173160173160173,
      "grad_norm": 6.6742095947265625,
      "learning_rate": 9.671111111111112e-06,
      "loss": 0.5871,
      "step": 650
    },
    {
      "epoch": 0.3652597402597403,
      "grad_norm": 8.321733474731445,
      "learning_rate": 9.615555555555558e-06,
      "loss": 0.5767,
      "step": 675
    },
    {
      "epoch": 0.3787878787878788,
      "grad_norm": 8.88350772857666,
      "learning_rate": 9.56e-06,
      "loss": 0.5387,
      "step": 700
    },
    {
      "epoch": 0.39231601731601734,
      "grad_norm": 8.155245780944824,
      "learning_rate": 9.504444444444446e-06,
      "loss": 0.5809,
      "step": 725
    },
    {
      "epoch": 0.40584415584415584,
      "grad_norm": 8.907155990600586,
      "learning_rate": 9.44888888888889e-06,
      "loss": 0.5778,
      "step": 750
    },
    {
      "epoch": 0.4193722943722944,
      "grad_norm": 9.427032470703125,
      "learning_rate": 9.393333333333334e-06,
      "loss": 0.5776,
      "step": 775
    },
    {
      "epoch": 0.4329004329004329,
      "grad_norm": 6.904598236083984,
      "learning_rate": 9.33777777777778e-06,
      "loss": 0.5287,
      "step": 800
    },
    {
      "epoch": 0.44642857142857145,
      "grad_norm": 7.866734504699707,
      "learning_rate": 9.282222222222222e-06,
      "loss": 0.5555,
      "step": 825
    },
    {
      "epoch": 0.45995670995670995,
      "grad_norm": 7.301151752471924,
      "learning_rate": 9.226666666666668e-06,
      "loss": 0.5296,
      "step": 850
    },
    {
      "epoch": 0.4734848484848485,
      "grad_norm": 9.370705604553223,
      "learning_rate": 9.171111111111112e-06,
      "loss": 0.5516,
      "step": 875
    },
    {
      "epoch": 0.487012987012987,
      "grad_norm": 7.46251916885376,
      "learning_rate": 9.115555555555556e-06,
      "loss": 0.534,
      "step": 900
    },
    {
      "epoch": 0.5005411255411255,
      "grad_norm": 7.885534286499023,
      "learning_rate": 9.060000000000001e-06,
      "loss": 0.5113,
      "step": 925
    },
    {
      "epoch": 0.5140692640692641,
      "grad_norm": 6.12823486328125,
      "learning_rate": 9.004444444444445e-06,
      "loss": 0.5274,
      "step": 950
    },
    {
      "epoch": 0.5275974025974026,
      "grad_norm": 7.1373515129089355,
      "learning_rate": 8.94888888888889e-06,
      "loss": 0.5241,
      "step": 975
    },
    {
      "epoch": 0.5411255411255411,
      "grad_norm": 7.099331378936768,
      "learning_rate": 8.893333333333333e-06,
      "loss": 0.5152,
      "step": 1000
    },
    {
      "epoch": 0.5411255411255411,
      "eval_loss": 0.4954243302345276,
      "eval_runtime": 1779.4215,
      "eval_samples_per_second": 2.192,
      "eval_steps_per_second": 0.137,
      "eval_wer": 0.3535207186322805,
      "step": 1000
    },
    {
      "epoch": 0.5546536796536796,
      "grad_norm": 8.30470085144043,
      "learning_rate": 8.83777777777778e-06,
      "loss": 0.531,
      "step": 1025
    },
    {
      "epoch": 0.5681818181818182,
      "grad_norm": 6.959774971008301,
      "learning_rate": 8.782222222222223e-06,
      "loss": 0.5038,
      "step": 1050
    },
    {
      "epoch": 0.5817099567099567,
      "grad_norm": 7.844577789306641,
      "learning_rate": 8.726666666666667e-06,
      "loss": 0.5196,
      "step": 1075
    },
    {
      "epoch": 0.5952380952380952,
      "grad_norm": 6.599257946014404,
      "learning_rate": 8.671111111111113e-06,
      "loss": 0.4982,
      "step": 1100
    },
    {
      "epoch": 0.6087662337662337,
      "grad_norm": 5.671600818634033,
      "learning_rate": 8.615555555555555e-06,
      "loss": 0.497,
      "step": 1125
    },
    {
      "epoch": 0.6222943722943723,
      "grad_norm": 6.545307636260986,
      "learning_rate": 8.560000000000001e-06,
      "loss": 0.4875,
      "step": 1150
    },
    {
      "epoch": 0.6358225108225108,
      "grad_norm": 6.877360820770264,
      "learning_rate": 8.504444444444445e-06,
      "loss": 0.5162,
      "step": 1175
    },
    {
      "epoch": 0.6493506493506493,
      "grad_norm": 7.325205326080322,
      "learning_rate": 8.448888888888889e-06,
      "loss": 0.5151,
      "step": 1200
    },
    {
      "epoch": 0.6628787878787878,
      "grad_norm": 6.775233745574951,
      "learning_rate": 8.393333333333335e-06,
      "loss": 0.498,
      "step": 1225
    },
    {
      "epoch": 0.6764069264069265,
      "grad_norm": 7.457151412963867,
      "learning_rate": 8.337777777777777e-06,
      "loss": 0.5012,
      "step": 1250
    },
    {
      "epoch": 0.689935064935065,
      "grad_norm": 7.285881042480469,
      "learning_rate": 8.282222222222223e-06,
      "loss": 0.4684,
      "step": 1275
    },
    {
      "epoch": 0.7034632034632035,
      "grad_norm": 9.163443565368652,
      "learning_rate": 8.226666666666667e-06,
      "loss": 0.5079,
      "step": 1300
    },
    {
      "epoch": 0.716991341991342,
      "grad_norm": 7.168745994567871,
      "learning_rate": 8.171111111111113e-06,
      "loss": 0.475,
      "step": 1325
    },
    {
      "epoch": 0.7305194805194806,
      "grad_norm": 7.457499027252197,
      "learning_rate": 8.115555555555557e-06,
      "loss": 0.488,
      "step": 1350
    },
    {
      "epoch": 0.7440476190476191,
      "grad_norm": 6.2372846603393555,
      "learning_rate": 8.06e-06,
      "loss": 0.4822,
      "step": 1375
    },
    {
      "epoch": 0.7575757575757576,
      "grad_norm": 5.880990505218506,
      "learning_rate": 8.004444444444445e-06,
      "loss": 0.531,
      "step": 1400
    },
    {
      "epoch": 0.7711038961038961,
      "grad_norm": 7.057967185974121,
      "learning_rate": 7.948888888888889e-06,
      "loss": 0.4872,
      "step": 1425
    },
    {
      "epoch": 0.7846320346320347,
      "grad_norm": 7.299345970153809,
      "learning_rate": 7.893333333333335e-06,
      "loss": 0.4749,
      "step": 1450
    },
    {
      "epoch": 0.7981601731601732,
      "grad_norm": 6.807291030883789,
      "learning_rate": 7.837777777777779e-06,
      "loss": 0.4676,
      "step": 1475
    },
    {
      "epoch": 0.8116883116883117,
      "grad_norm": 5.556617736816406,
      "learning_rate": 7.782222222222223e-06,
      "loss": 0.4614,
      "step": 1500
    },
    {
      "epoch": 0.8252164502164502,
      "grad_norm": 6.165937900543213,
      "learning_rate": 7.726666666666667e-06,
      "loss": 0.4459,
      "step": 1525
    },
    {
      "epoch": 0.8387445887445888,
      "grad_norm": 6.99851655960083,
      "learning_rate": 7.67111111111111e-06,
      "loss": 0.4734,
      "step": 1550
    },
    {
      "epoch": 0.8522727272727273,
      "grad_norm": 7.385776519775391,
      "learning_rate": 7.6155555555555564e-06,
      "loss": 0.4547,
      "step": 1575
    },
    {
      "epoch": 0.8658008658008658,
      "grad_norm": 6.626092910766602,
      "learning_rate": 7.5600000000000005e-06,
      "loss": 0.4462,
      "step": 1600
    },
    {
      "epoch": 0.8793290043290043,
      "grad_norm": 6.563342094421387,
      "learning_rate": 7.504444444444445e-06,
      "loss": 0.4511,
      "step": 1625
    },
    {
      "epoch": 0.8928571428571429,
      "grad_norm": 9.904861450195312,
      "learning_rate": 7.44888888888889e-06,
      "loss": 0.4728,
      "step": 1650
    },
    {
      "epoch": 0.9063852813852814,
      "grad_norm": 7.5107622146606445,
      "learning_rate": 7.393333333333333e-06,
      "loss": 0.4867,
      "step": 1675
    },
    {
      "epoch": 0.9199134199134199,
      "grad_norm": 6.618627548217773,
      "learning_rate": 7.337777777777778e-06,
      "loss": 0.4512,
      "step": 1700
    },
    {
      "epoch": 0.9334415584415584,
      "grad_norm": 7.19182014465332,
      "learning_rate": 7.282222222222222e-06,
      "loss": 0.4657,
      "step": 1725
    },
    {
      "epoch": 0.946969696969697,
      "grad_norm": 6.207240104675293,
      "learning_rate": 7.226666666666667e-06,
      "loss": 0.4455,
      "step": 1750
    },
    {
      "epoch": 0.9604978354978355,
      "grad_norm": 8.109068870544434,
      "learning_rate": 7.171111111111112e-06,
      "loss": 0.4744,
      "step": 1775
    },
    {
      "epoch": 0.974025974025974,
      "grad_norm": 7.550827503204346,
      "learning_rate": 7.115555555555557e-06,
      "loss": 0.4565,
      "step": 1800
    },
    {
      "epoch": 0.9875541125541125,
      "grad_norm": 5.667859077453613,
      "learning_rate": 7.06e-06,
      "loss": 0.4368,
      "step": 1825
    },
    {
      "epoch": 1.001082251082251,
      "grad_norm": 5.609886646270752,
      "learning_rate": 7.004444444444445e-06,
      "loss": 0.4406,
      "step": 1850
    },
    {
      "epoch": 1.0146103896103895,
      "grad_norm": 4.862238883972168,
      "learning_rate": 6.948888888888889e-06,
      "loss": 0.3379,
      "step": 1875
    },
    {
      "epoch": 1.0281385281385282,
      "grad_norm": 6.2563066482543945,
      "learning_rate": 6.893333333333334e-06,
      "loss": 0.3386,
      "step": 1900
    },
    {
      "epoch": 1.0416666666666667,
      "grad_norm": 6.764842987060547,
      "learning_rate": 6.837777777777779e-06,
      "loss": 0.3534,
      "step": 1925
    },
    {
      "epoch": 1.0551948051948052,
      "grad_norm": 5.962332248687744,
      "learning_rate": 6.782222222222222e-06,
      "loss": 0.3212,
      "step": 1950
    },
    {
      "epoch": 1.0687229437229437,
      "grad_norm": 5.471970081329346,
      "learning_rate": 6.726666666666667e-06,
      "loss": 0.3572,
      "step": 1975
    },
    {
      "epoch": 1.0822510822510822,
      "grad_norm": 6.054861545562744,
      "learning_rate": 6.671111111111112e-06,
      "loss": 0.3339,
      "step": 2000
    },
    {
      "epoch": 1.0822510822510822,
      "eval_loss": 0.42054763436317444,
      "eval_runtime": 1782.8878,
      "eval_samples_per_second": 2.188,
      "eval_steps_per_second": 0.137,
      "eval_wer": 0.3197865353037767,
      "step": 2000
    },
    {
      "epoch": 1.0957792207792207,
      "grad_norm": 6.194203853607178,
      "learning_rate": 6.615555555555556e-06,
      "loss": 0.339,
      "step": 2025
    },
    {
      "epoch": 1.1093073593073592,
      "grad_norm": 5.470515727996826,
      "learning_rate": 6.560000000000001e-06,
      "loss": 0.3375,
      "step": 2050
    },
    {
      "epoch": 1.1228354978354977,
      "grad_norm": 5.1414618492126465,
      "learning_rate": 6.504444444444446e-06,
      "loss": 0.3348,
      "step": 2075
    },
    {
      "epoch": 1.1363636363636362,
      "grad_norm": 5.000445365905762,
      "learning_rate": 6.448888888888889e-06,
      "loss": 0.3255,
      "step": 2100
    },
    {
      "epoch": 1.149891774891775,
      "grad_norm": 5.545360088348389,
      "learning_rate": 6.393333333333334e-06,
      "loss": 0.3296,
      "step": 2125
    },
    {
      "epoch": 1.1634199134199135,
      "grad_norm": 5.920198440551758,
      "learning_rate": 6.3377777777777786e-06,
      "loss": 0.3436,
      "step": 2150
    },
    {
      "epoch": 1.176948051948052,
      "grad_norm": 5.722521781921387,
      "learning_rate": 6.282222222222223e-06,
      "loss": 0.3366,
      "step": 2175
    },
    {
      "epoch": 1.1904761904761905,
      "grad_norm": 6.066483020782471,
      "learning_rate": 6.2266666666666675e-06,
      "loss": 0.332,
      "step": 2200
    },
    {
      "epoch": 1.204004329004329,
      "grad_norm": 6.301929473876953,
      "learning_rate": 6.171111111111112e-06,
      "loss": 0.3207,
      "step": 2225
    },
    {
      "epoch": 1.2175324675324675,
      "grad_norm": 5.607754230499268,
      "learning_rate": 6.1155555555555555e-06,
      "loss": 0.3338,
      "step": 2250
    },
    {
      "epoch": 1.231060606060606,
      "grad_norm": 5.145053863525391,
      "learning_rate": 6.0600000000000004e-06,
      "loss": 0.3268,
      "step": 2275
    },
    {
      "epoch": 1.2445887445887447,
      "grad_norm": 5.448360443115234,
      "learning_rate": 6.004444444444445e-06,
      "loss": 0.3365,
      "step": 2300
    },
    {
      "epoch": 1.2581168831168832,
      "grad_norm": 5.5156474113464355,
      "learning_rate": 5.948888888888889e-06,
      "loss": 0.3268,
      "step": 2325
    },
    {
      "epoch": 1.2716450216450217,
      "grad_norm": 5.252381324768066,
      "learning_rate": 5.893333333333334e-06,
      "loss": 0.3228,
      "step": 2350
    },
    {
      "epoch": 1.2851731601731602,
      "grad_norm": 5.7689313888549805,
      "learning_rate": 5.837777777777777e-06,
      "loss": 0.3304,
      "step": 2375
    },
    {
      "epoch": 1.2987012987012987,
      "grad_norm": 4.822956085205078,
      "learning_rate": 5.782222222222222e-06,
      "loss": 0.3066,
      "step": 2400
    },
    {
      "epoch": 1.3122294372294372,
      "grad_norm": 5.012087345123291,
      "learning_rate": 5.726666666666667e-06,
      "loss": 0.3323,
      "step": 2425
    },
    {
      "epoch": 1.3257575757575757,
      "grad_norm": 5.262439250946045,
      "learning_rate": 5.671111111111112e-06,
      "loss": 0.3256,
      "step": 2450
    },
    {
      "epoch": 1.3392857142857144,
      "grad_norm": 5.300339221954346,
      "learning_rate": 5.615555555555556e-06,
      "loss": 0.3287,
      "step": 2475
    },
    {
      "epoch": 1.3528138528138527,
      "grad_norm": 6.058621883392334,
      "learning_rate": 5.560000000000001e-06,
      "loss": 0.3283,
      "step": 2500
    },
    {
      "epoch": 1.3663419913419914,
      "grad_norm": 6.223220348358154,
      "learning_rate": 5.504444444444444e-06,
      "loss": 0.3288,
      "step": 2525
    },
    {
      "epoch": 1.37987012987013,
      "grad_norm": 5.865265369415283,
      "learning_rate": 5.448888888888889e-06,
      "loss": 0.3303,
      "step": 2550
    },
    {
      "epoch": 1.3933982683982684,
      "grad_norm": 4.715255260467529,
      "learning_rate": 5.393333333333334e-06,
      "loss": 0.3432,
      "step": 2575
    },
    {
      "epoch": 1.406926406926407,
      "grad_norm": 5.57729434967041,
      "learning_rate": 5.337777777777779e-06,
      "loss": 0.3174,
      "step": 2600
    },
    {
      "epoch": 1.4204545454545454,
      "grad_norm": 6.372653484344482,
      "learning_rate": 5.282222222222223e-06,
      "loss": 0.3251,
      "step": 2625
    },
    {
      "epoch": 1.433982683982684,
      "grad_norm": 6.7026848793029785,
      "learning_rate": 5.226666666666667e-06,
      "loss": 0.3258,
      "step": 2650
    },
    {
      "epoch": 1.4475108225108224,
      "grad_norm": 5.12203311920166,
      "learning_rate": 5.171111111111111e-06,
      "loss": 0.3217,
      "step": 2675
    },
    {
      "epoch": 1.4610389610389611,
      "grad_norm": 7.778601169586182,
      "learning_rate": 5.115555555555556e-06,
      "loss": 0.3182,
      "step": 2700
    },
    {
      "epoch": 1.4745670995670996,
      "grad_norm": 4.994805335998535,
      "learning_rate": 5.060000000000001e-06,
      "loss": 0.3142,
      "step": 2725
    },
    {
      "epoch": 1.4880952380952381,
      "grad_norm": 6.392801761627197,
      "learning_rate": 5.004444444444445e-06,
      "loss": 0.3289,
      "step": 2750
    },
    {
      "epoch": 1.5016233766233766,
      "grad_norm": 5.523842811584473,
      "learning_rate": 4.94888888888889e-06,
      "loss": 0.3248,
      "step": 2775
    },
    {
      "epoch": 1.5151515151515151,
      "grad_norm": 5.348546981811523,
      "learning_rate": 4.893333333333334e-06,
      "loss": 0.303,
      "step": 2800
    },
    {
      "epoch": 1.5286796536796536,
      "grad_norm": 5.714568614959717,
      "learning_rate": 4.837777777777778e-06,
      "loss": 0.3144,
      "step": 2825
    },
    {
      "epoch": 1.5422077922077921,
      "grad_norm": 5.544715404510498,
      "learning_rate": 4.7822222222222226e-06,
      "loss": 0.3183,
      "step": 2850
    },
    {
      "epoch": 1.5557359307359309,
      "grad_norm": 6.49782133102417,
      "learning_rate": 4.7266666666666674e-06,
      "loss": 0.2981,
      "step": 2875
    },
    {
      "epoch": 1.5692640692640691,
      "grad_norm": 5.356492042541504,
      "learning_rate": 4.6711111111111115e-06,
      "loss": 0.3159,
      "step": 2900
    },
    {
      "epoch": 1.5827922077922079,
      "grad_norm": 5.4491119384765625,
      "learning_rate": 4.6155555555555555e-06,
      "loss": 0.3333,
      "step": 2925
    },
    {
      "epoch": 1.5963203463203464,
      "grad_norm": 5.832214832305908,
      "learning_rate": 4.56e-06,
      "loss": 0.3141,
      "step": 2950
    },
    {
      "epoch": 1.6098484848484849,
      "grad_norm": 5.139626979827881,
      "learning_rate": 4.504444444444444e-06,
      "loss": 0.308,
      "step": 2975
    },
    {
      "epoch": 1.6233766233766234,
      "grad_norm": 5.6519999504089355,
      "learning_rate": 4.448888888888889e-06,
      "loss": 0.3189,
      "step": 3000
    },
    {
      "epoch": 1.6233766233766234,
      "eval_loss": 0.3910607397556305,
      "eval_runtime": 1770.6955,
      "eval_samples_per_second": 2.203,
      "eval_steps_per_second": 0.138,
      "eval_wer": 0.29134067420071474,
      "step": 3000
    },
    {
      "epoch": 1.6369047619047619,
      "grad_norm": 5.791579246520996,
      "learning_rate": 4.393333333333334e-06,
      "loss": 0.3122,
      "step": 3025
    },
    {
      "epoch": 1.6504329004329006,
      "grad_norm": 5.640200614929199,
      "learning_rate": 4.337777777777778e-06,
      "loss": 0.3062,
      "step": 3050
    },
    {
      "epoch": 1.6639610389610389,
      "grad_norm": 5.585713863372803,
      "learning_rate": 4.282222222222222e-06,
      "loss": 0.3208,
      "step": 3075
    },
    {
      "epoch": 1.6774891774891776,
      "grad_norm": 5.871087074279785,
      "learning_rate": 4.226666666666667e-06,
      "loss": 0.3071,
      "step": 3100
    },
    {
      "epoch": 1.6910173160173159,
      "grad_norm": 5.412327766418457,
      "learning_rate": 4.171111111111111e-06,
      "loss": 0.3167,
      "step": 3125
    },
    {
      "epoch": 1.7045454545454546,
      "grad_norm": 5.0698561668396,
      "learning_rate": 4.115555555555556e-06,
      "loss": 0.3231,
      "step": 3150
    },
    {
      "epoch": 1.718073593073593,
      "grad_norm": 5.63693904876709,
      "learning_rate": 4.060000000000001e-06,
      "loss": 0.3128,
      "step": 3175
    },
    {
      "epoch": 1.7316017316017316,
      "grad_norm": 5.766589164733887,
      "learning_rate": 4.004444444444445e-06,
      "loss": 0.3229,
      "step": 3200
    },
    {
      "epoch": 1.74512987012987,
      "grad_norm": 5.414788246154785,
      "learning_rate": 3.948888888888889e-06,
      "loss": 0.2917,
      "step": 3225
    },
    {
      "epoch": 1.7586580086580086,
      "grad_norm": 5.106072902679443,
      "learning_rate": 3.893333333333333e-06,
      "loss": 0.311,
      "step": 3250
    },
    {
      "epoch": 1.7721861471861473,
      "grad_norm": 4.694611549377441,
      "learning_rate": 3.837777777777778e-06,
      "loss": 0.3228,
      "step": 3275
    },
    {
      "epoch": 1.7857142857142856,
      "grad_norm": 6.422979354858398,
      "learning_rate": 3.782222222222223e-06,
      "loss": 0.314,
      "step": 3300
    },
    {
      "epoch": 1.7992424242424243,
      "grad_norm": 5.5537567138671875,
      "learning_rate": 3.726666666666667e-06,
      "loss": 0.303,
      "step": 3325
    },
    {
      "epoch": 1.8127705627705628,
      "grad_norm": 6.503033638000488,
      "learning_rate": 3.6711111111111113e-06,
      "loss": 0.336,
      "step": 3350
    },
    {
      "epoch": 1.8262987012987013,
      "grad_norm": 5.406898021697998,
      "learning_rate": 3.615555555555556e-06,
      "loss": 0.3031,
      "step": 3375
    },
    {
      "epoch": 1.8398268398268398,
      "grad_norm": 6.486941337585449,
      "learning_rate": 3.5600000000000002e-06,
      "loss": 0.3203,
      "step": 3400
    },
    {
      "epoch": 1.8533549783549783,
      "grad_norm": 7.027703285217285,
      "learning_rate": 3.5044444444444447e-06,
      "loss": 0.3159,
      "step": 3425
    },
    {
      "epoch": 1.866883116883117,
      "grad_norm": 5.475865364074707,
      "learning_rate": 3.4488888888888896e-06,
      "loss": 0.3239,
      "step": 3450
    },
    {
      "epoch": 1.8804112554112553,
      "grad_norm": 6.124994277954102,
      "learning_rate": 3.3933333333333336e-06,
      "loss": 0.2928,
      "step": 3475
    },
    {
      "epoch": 1.893939393939394,
      "grad_norm": 4.759301662445068,
      "learning_rate": 3.337777777777778e-06,
      "loss": 0.2862,
      "step": 3500
    },
    {
      "epoch": 1.9074675324675323,
      "grad_norm": 5.548280239105225,
      "learning_rate": 3.282222222222223e-06,
      "loss": 0.312,
      "step": 3525
    },
    {
      "epoch": 1.920995670995671,
      "grad_norm": 5.691162109375,
      "learning_rate": 3.226666666666667e-06,
      "loss": 0.3161,
      "step": 3550
    },
    {
      "epoch": 1.9345238095238095,
      "grad_norm": 6.089394569396973,
      "learning_rate": 3.1711111111111114e-06,
      "loss": 0.3028,
      "step": 3575
    },
    {
      "epoch": 1.948051948051948,
      "grad_norm": 5.725650310516357,
      "learning_rate": 3.1155555555555555e-06,
      "loss": 0.3058,
      "step": 3600
    },
    {
      "epoch": 1.9615800865800865,
      "grad_norm": 5.124326705932617,
      "learning_rate": 3.0600000000000003e-06,
      "loss": 0.2947,
      "step": 3625
    },
    {
      "epoch": 1.975108225108225,
      "grad_norm": 6.62967586517334,
      "learning_rate": 3.004444444444445e-06,
      "loss": 0.318,
      "step": 3650
    },
    {
      "epoch": 1.9886363636363638,
      "grad_norm": 6.150094985961914,
      "learning_rate": 2.948888888888889e-06,
      "loss": 0.3257,
      "step": 3675
    },
    {
      "epoch": 2.002164502164502,
      "grad_norm": 3.962730884552002,
      "learning_rate": 2.8933333333333337e-06,
      "loss": 0.29,
      "step": 3700
    },
    {
      "epoch": 2.0156926406926408,
      "grad_norm": 3.999758005142212,
      "learning_rate": 2.837777777777778e-06,
      "loss": 0.2127,
      "step": 3725
    },
    {
      "epoch": 2.029220779220779,
      "grad_norm": 4.3916015625,
      "learning_rate": 2.7822222222222222e-06,
      "loss": 0.2073,
      "step": 3750
    },
    {
      "epoch": 2.0427489177489178,
      "grad_norm": 4.647676944732666,
      "learning_rate": 2.726666666666667e-06,
      "loss": 0.2065,
      "step": 3775
    },
    {
      "epoch": 2.0562770562770565,
      "grad_norm": 5.331233501434326,
      "learning_rate": 2.6711111111111116e-06,
      "loss": 0.208,
      "step": 3800
    },
    {
      "epoch": 2.0698051948051948,
      "grad_norm": 4.8974995613098145,
      "learning_rate": 2.6155555555555556e-06,
      "loss": 0.2007,
      "step": 3825
    },
    {
      "epoch": 2.0833333333333335,
      "grad_norm": 4.972061634063721,
      "learning_rate": 2.56e-06,
      "loss": 0.2017,
      "step": 3850
    },
    {
      "epoch": 2.0968614718614718,
      "grad_norm": 5.035933494567871,
      "learning_rate": 2.504444444444445e-06,
      "loss": 0.2075,
      "step": 3875
    },
    {
      "epoch": 2.1103896103896105,
      "grad_norm": 6.620712757110596,
      "learning_rate": 2.448888888888889e-06,
      "loss": 0.2074,
      "step": 3900
    },
    {
      "epoch": 2.1239177489177488,
      "grad_norm": 7.161535739898682,
      "learning_rate": 2.3933333333333334e-06,
      "loss": 0.2075,
      "step": 3925
    },
    {
      "epoch": 2.1374458874458875,
      "grad_norm": 5.531479358673096,
      "learning_rate": 2.337777777777778e-06,
      "loss": 0.2088,
      "step": 3950
    },
    {
      "epoch": 2.150974025974026,
      "grad_norm": 4.983880043029785,
      "learning_rate": 2.2822222222222223e-06,
      "loss": 0.2103,
      "step": 3975
    },
    {
      "epoch": 2.1645021645021645,
      "grad_norm": 4.373054504394531,
      "learning_rate": 2.226666666666667e-06,
      "loss": 0.2051,
      "step": 4000
    },
    {
      "epoch": 2.1645021645021645,
      "eval_loss": 0.38633546233177185,
      "eval_runtime": 1768.38,
      "eval_samples_per_second": 2.206,
      "eval_steps_per_second": 0.138,
      "eval_wer": 0.27895296049454266,
      "step": 4000
    },
    {
      "epoch": 2.178030303030303,
      "grad_norm": 5.22249174118042,
      "learning_rate": 2.1711111111111113e-06,
      "loss": 0.1988,
      "step": 4025
    },
    {
      "epoch": 2.1915584415584415,
      "grad_norm": 3.8173792362213135,
      "learning_rate": 2.1155555555555557e-06,
      "loss": 0.2241,
      "step": 4050
    },
    {
      "epoch": 2.20508658008658,
      "grad_norm": 5.271940231323242,
      "learning_rate": 2.06e-06,
      "loss": 0.201,
      "step": 4075
    },
    {
      "epoch": 2.2186147186147185,
      "grad_norm": 4.359199523925781,
      "learning_rate": 2.0044444444444446e-06,
      "loss": 0.1946,
      "step": 4100
    },
    {
      "epoch": 2.232142857142857,
      "grad_norm": 4.675993919372559,
      "learning_rate": 1.948888888888889e-06,
      "loss": 0.2123,
      "step": 4125
    },
    {
      "epoch": 2.2456709956709955,
      "grad_norm": 4.090628147125244,
      "learning_rate": 1.8933333333333333e-06,
      "loss": 0.2045,
      "step": 4150
    },
    {
      "epoch": 2.259199134199134,
      "grad_norm": 3.5872035026550293,
      "learning_rate": 1.837777777777778e-06,
      "loss": 0.2042,
      "step": 4175
    },
    {
      "epoch": 2.2727272727272725,
      "grad_norm": 4.375498294830322,
      "learning_rate": 1.7822222222222225e-06,
      "loss": 0.2056,
      "step": 4200
    },
    {
      "epoch": 2.286255411255411,
      "grad_norm": 4.972550868988037,
      "learning_rate": 1.7266666666666667e-06,
      "loss": 0.2024,
      "step": 4225
    },
    {
      "epoch": 2.29978354978355,
      "grad_norm": 4.7940168380737305,
      "learning_rate": 1.6711111111111112e-06,
      "loss": 0.1996,
      "step": 4250
    },
    {
      "epoch": 2.313311688311688,
      "grad_norm": 4.399414539337158,
      "learning_rate": 1.6155555555555559e-06,
      "loss": 0.2101,
      "step": 4275
    },
    {
      "epoch": 2.326839826839827,
      "grad_norm": 5.292896747589111,
      "learning_rate": 1.56e-06,
      "loss": 0.1929,
      "step": 4300
    },
    {
      "epoch": 2.340367965367965,
      "grad_norm": 4.333370685577393,
      "learning_rate": 1.5044444444444446e-06,
      "loss": 0.208,
      "step": 4325
    },
    {
      "epoch": 2.353896103896104,
      "grad_norm": 4.69057035446167,
      "learning_rate": 1.4488888888888892e-06,
      "loss": 0.1905,
      "step": 4350
    },
    {
      "epoch": 2.367424242424242,
      "grad_norm": 4.56622838973999,
      "learning_rate": 1.3933333333333335e-06,
      "loss": 0.2051,
      "step": 4375
    },
    {
      "epoch": 2.380952380952381,
      "grad_norm": 4.605253219604492,
      "learning_rate": 1.337777777777778e-06,
      "loss": 0.1978,
      "step": 4400
    },
    {
      "epoch": 2.3944805194805197,
      "grad_norm": 5.727032661437988,
      "learning_rate": 1.2822222222222222e-06,
      "loss": 0.2002,
      "step": 4425
    },
    {
      "epoch": 2.408008658008658,
      "grad_norm": 5.882457256317139,
      "learning_rate": 1.2266666666666666e-06,
      "loss": 0.202,
      "step": 4450
    },
    {
      "epoch": 2.4215367965367967,
      "grad_norm": 4.464743614196777,
      "learning_rate": 1.171111111111111e-06,
      "loss": 0.2145,
      "step": 4475
    },
    {
      "epoch": 2.435064935064935,
      "grad_norm": 4.503987789154053,
      "learning_rate": 1.1155555555555558e-06,
      "loss": 0.2101,
      "step": 4500
    },
    {
      "epoch": 2.4485930735930737,
      "grad_norm": 5.735741138458252,
      "learning_rate": 1.06e-06,
      "loss": 0.1913,
      "step": 4525
    },
    {
      "epoch": 2.462121212121212,
      "grad_norm": 4.6319098472595215,
      "learning_rate": 1.0044444444444445e-06,
      "loss": 0.2001,
      "step": 4550
    },
    {
      "epoch": 2.4756493506493507,
      "grad_norm": 5.589540958404541,
      "learning_rate": 9.488888888888889e-07,
      "loss": 0.1981,
      "step": 4575
    },
    {
      "epoch": 2.4891774891774894,
      "grad_norm": 4.481135845184326,
      "learning_rate": 8.933333333333334e-07,
      "loss": 0.198,
      "step": 4600
    },
    {
      "epoch": 2.5027056277056277,
      "grad_norm": 6.087165355682373,
      "learning_rate": 8.37777777777778e-07,
      "loss": 0.2131,
      "step": 4625
    },
    {
      "epoch": 2.5162337662337664,
      "grad_norm": 4.635289669036865,
      "learning_rate": 7.822222222222223e-07,
      "loss": 0.2088,
      "step": 4650
    },
    {
      "epoch": 2.5297619047619047,
      "grad_norm": 4.698585510253906,
      "learning_rate": 7.266666666666668e-07,
      "loss": 0.2057,
      "step": 4675
    },
    {
      "epoch": 2.5432900432900434,
      "grad_norm": 4.562716960906982,
      "learning_rate": 6.711111111111111e-07,
      "loss": 0.2117,
      "step": 4700
    },
    {
      "epoch": 2.5568181818181817,
      "grad_norm": 5.381985187530518,
      "learning_rate": 6.155555555555556e-07,
      "loss": 0.1975,
      "step": 4725
    },
    {
      "epoch": 2.5703463203463204,
      "grad_norm": 5.667773723602295,
      "learning_rate": 5.6e-07,
      "loss": 0.2409,
      "step": 4750
    },
    {
      "epoch": 2.583874458874459,
      "grad_norm": 4.565330982208252,
      "learning_rate": 5.044444444444445e-07,
      "loss": 0.1915,
      "step": 4775
    },
    {
      "epoch": 2.5974025974025974,
      "grad_norm": 5.17742395401001,
      "learning_rate": 4.488888888888889e-07,
      "loss": 0.1973,
      "step": 4800
    },
    {
      "epoch": 2.6109307359307357,
      "grad_norm": 4.878474712371826,
      "learning_rate": 3.9333333333333336e-07,
      "loss": 0.2209,
      "step": 4825
    },
    {
      "epoch": 2.6244588744588744,
      "grad_norm": 5.2556328773498535,
      "learning_rate": 3.3777777777777777e-07,
      "loss": 0.204,
      "step": 4850
    },
    {
      "epoch": 2.637987012987013,
      "grad_norm": 3.8071792125701904,
      "learning_rate": 2.822222222222222e-07,
      "loss": 0.2052,
      "step": 4875
    },
    {
      "epoch": 2.6515151515151514,
      "grad_norm": 4.218277454376221,
      "learning_rate": 2.266666666666667e-07,
      "loss": 0.1989,
      "step": 4900
    },
    {
      "epoch": 2.66504329004329,
      "grad_norm": 5.260907173156738,
      "learning_rate": 1.7111111111111114e-07,
      "loss": 0.1915,
      "step": 4925
    },
    {
      "epoch": 2.678571428571429,
      "grad_norm": 4.497314453125,
      "learning_rate": 1.1555555555555556e-07,
      "loss": 0.1997,
      "step": 4950
    },
    {
      "epoch": 2.692099567099567,
      "grad_norm": 4.543353080749512,
      "learning_rate": 6.000000000000001e-08,
      "loss": 0.1866,
      "step": 4975
    },
    {
      "epoch": 2.7056277056277054,
      "grad_norm": 6.265724182128906,
      "learning_rate": 4.444444444444445e-09,
      "loss": 0.202,
      "step": 5000
    },
    {
      "epoch": 2.7056277056277054,
      "eval_loss": 0.38099274039268494,
      "eval_runtime": 1768.0451,
      "eval_samples_per_second": 2.206,
      "eval_steps_per_second": 0.138,
      "eval_wer": 0.27501690331304934,
      "step": 5000
    },
    {
      "epoch": 2.7056277056277054,
      "step": 5000,
      "total_flos": 5.434978041004032e+20,
      "train_loss": 0.39443905401229856,
      "train_runtime": 57517.0908,
      "train_samples_per_second": 2.782,
      "train_steps_per_second": 0.087
    }
  ],
  "logging_steps": 25,
  "max_steps": 5000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 5.434978041004032e+20,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}