{
  "best_metric": 0.6322699785232544,
  "best_model_checkpoint": "./Hubert-common_voice-ja-demo-kana-only-cosine/checkpoint-9400",
  "epoch": 25.0,
  "eval_steps": 100,
  "global_step": 9400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.26595744680851063,
      "eval_cer": 5.820971309556485,
      "eval_loss": 43.225128173828125,
      "eval_runtime": 207.1142,
      "eval_samples_per_second": 23.953,
      "eval_steps_per_second": 2.998,
      "eval_wer": 1.5295303366256803,
      "step": 100
    },
    {
      "epoch": 0.5319148936170213,
      "eval_cer": 5.284926618983806,
      "eval_loss": 42.44519805908203,
      "eval_runtime": 204.7199,
      "eval_samples_per_second": 24.233,
      "eval_steps_per_second": 3.033,
      "eval_wer": 1.5321507760532151,
      "step": 200
    },
    {
      "epoch": 0.7978723404255319,
      "eval_cer": 1.8125998998982837,
      "eval_loss": 40.415428161621094,
      "eval_runtime": 201.5553,
      "eval_samples_per_second": 24.614,
      "eval_steps_per_second": 3.081,
      "eval_wer": 1.126184237048982,
      "step": 300
    },
    {
      "epoch": 1.0638297872340425,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 32.8021240234375,
      "eval_runtime": 198.6844,
      "eval_samples_per_second": 24.969,
      "eval_steps_per_second": 3.126,
      "eval_wer": 1.0,
      "step": 400
    },
    {
      "epoch": 1.3297872340425532,
      "grad_norm": 110.69781494140625,
      "learning_rate": 1.188e-06,
      "loss": 31.884,
      "step": 500
    },
    {
      "epoch": 1.3297872340425532,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 20.813316345214844,
      "eval_runtime": 217.9621,
      "eval_samples_per_second": 22.761,
      "eval_steps_per_second": 2.849,
      "eval_wer": 1.0,
      "step": 500
    },
    {
      "epoch": 1.5957446808510638,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 17.582351684570312,
      "eval_runtime": 200.2418,
      "eval_samples_per_second": 24.775,
      "eval_steps_per_second": 3.101,
      "eval_wer": 1.0,
      "step": 600
    },
    {
      "epoch": 1.8617021276595744,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 16.86818504333496,
      "eval_runtime": 200.8491,
      "eval_samples_per_second": 24.7,
      "eval_steps_per_second": 3.092,
      "eval_wer": 1.0,
      "step": 700
    },
    {
      "epoch": 2.127659574468085,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 16.446561813354492,
      "eval_runtime": 206.3537,
      "eval_samples_per_second": 24.041,
      "eval_steps_per_second": 3.009,
      "eval_wer": 1.0,
      "step": 800
    },
    {
      "epoch": 2.393617021276596,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 16.003677368164062,
      "eval_runtime": 203.6688,
      "eval_samples_per_second": 24.358,
      "eval_steps_per_second": 3.049,
      "eval_wer": 1.0,
      "step": 900
    },
    {
      "epoch": 2.6595744680851063,
      "grad_norm": 112.62248229980469,
      "learning_rate": 2.3880000000000003e-06,
      "loss": 14.4701,
      "step": 1000
    },
    {
      "epoch": 2.6595744680851063,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 15.540871620178223,
      "eval_runtime": 205.8626,
      "eval_samples_per_second": 24.099,
      "eval_steps_per_second": 3.017,
      "eval_wer": 1.0,
      "step": 1000
    },
    {
      "epoch": 2.925531914893617,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 15.044602394104004,
      "eval_runtime": 203.7066,
      "eval_samples_per_second": 24.354,
      "eval_steps_per_second": 3.049,
      "eval_wer": 1.0,
      "step": 1100
    },
    {
      "epoch": 3.1914893617021276,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 14.502347946166992,
      "eval_runtime": 203.5102,
      "eval_samples_per_second": 24.377,
      "eval_steps_per_second": 3.051,
      "eval_wer": 1.0,
      "step": 1200
    },
    {
      "epoch": 3.4574468085106385,
      "eval_cer": 0.9999031273713612,
      "eval_loss": 13.929794311523438,
      "eval_runtime": 205.0004,
      "eval_samples_per_second": 24.2,
      "eval_steps_per_second": 3.029,
      "eval_wer": 1.0,
      "step": 1300
    },
    {
      "epoch": 3.723404255319149,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 13.321216583251953,
      "eval_runtime": 206.9747,
      "eval_samples_per_second": 23.969,
      "eval_steps_per_second": 3.0,
      "eval_wer": 1.0,
      "step": 1400
    },
    {
      "epoch": 3.9893617021276597,
      "grad_norm": 89.55043029785156,
      "learning_rate": 3.588e-06,
      "loss": 12.1626,
      "step": 1500
    },
    {
      "epoch": 3.9893617021276597,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 12.681354522705078,
      "eval_runtime": 204.7436,
      "eval_samples_per_second": 24.23,
      "eval_steps_per_second": 3.033,
      "eval_wer": 1.0,
      "step": 1500
    },
    {
      "epoch": 4.25531914893617,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 12.00993824005127,
      "eval_runtime": 205.3753,
      "eval_samples_per_second": 24.156,
      "eval_steps_per_second": 3.024,
      "eval_wer": 1.0,
      "step": 1600
    },
    {
      "epoch": 4.5212765957446805,
      "eval_cer": 0.9999031273713612,
      "eval_loss": 11.317916870117188,
      "eval_runtime": 208.5747,
      "eval_samples_per_second": 23.785,
      "eval_steps_per_second": 2.977,
      "eval_wer": 1.0,
      "step": 1700
    },
    {
      "epoch": 4.787234042553192,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 10.601661682128906,
      "eval_runtime": 205.7117,
      "eval_samples_per_second": 24.116,
      "eval_steps_per_second": 3.019,
      "eval_wer": 1.0,
      "step": 1800
    },
    {
      "epoch": 5.053191489361702,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 9.880986213684082,
      "eval_runtime": 206.3181,
      "eval_samples_per_second": 24.045,
      "eval_steps_per_second": 3.01,
      "eval_wer": 1.0,
      "step": 1900
    },
    {
      "epoch": 5.319148936170213,
      "grad_norm": 55.87970733642578,
      "learning_rate": 4.788e-06,
      "loss": 9.5127,
      "step": 2000
    },
    {
      "epoch": 5.319148936170213,
      "eval_cer": 0.9999031273713612,
      "eval_loss": 9.15669059753418,
      "eval_runtime": 205.9181,
      "eval_samples_per_second": 24.092,
      "eval_steps_per_second": 3.016,
      "eval_wer": 1.0,
      "step": 2000
    },
    {
      "epoch": 5.585106382978723,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 8.444498062133789,
      "eval_runtime": 207.3299,
      "eval_samples_per_second": 23.928,
      "eval_steps_per_second": 2.995,
      "eval_wer": 1.0,
      "step": 2100
    },
    {
      "epoch": 5.851063829787234,
      "eval_cer": 0.9999031273713612,
      "eval_loss": 7.757309913635254,
      "eval_runtime": 207.2514,
      "eval_samples_per_second": 23.937,
      "eval_steps_per_second": 2.996,
      "eval_wer": 1.0,
      "step": 2200
    },
    {
      "epoch": 6.117021276595745,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 7.104926109313965,
      "eval_runtime": 207.5258,
      "eval_samples_per_second": 23.905,
      "eval_steps_per_second": 2.992,
      "eval_wer": 1.0,
      "step": 2300
    },
    {
      "epoch": 6.382978723404255,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 6.5016021728515625,
      "eval_runtime": 207.6536,
      "eval_samples_per_second": 23.891,
      "eval_steps_per_second": 2.991,
      "eval_wer": 1.0,
      "step": 2400
    },
    {
      "epoch": 6.648936170212766,
      "grad_norm": 40.37785339355469,
      "learning_rate": 5.988e-06,
      "loss": 6.6873,
      "step": 2500
    },
    {
      "epoch": 6.648936170212766,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 5.956511497497559,
      "eval_runtime": 208.8491,
      "eval_samples_per_second": 23.754,
      "eval_steps_per_second": 2.973,
      "eval_wer": 1.0,
      "step": 2500
    },
    {
      "epoch": 6.914893617021277,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 5.485317707061768,
      "eval_runtime": 208.251,
      "eval_samples_per_second": 23.822,
      "eval_steps_per_second": 2.982,
      "eval_wer": 1.0,
      "step": 2600
    },
    {
      "epoch": 7.180851063829787,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 5.099748134613037,
      "eval_runtime": 208.5061,
      "eval_samples_per_second": 23.793,
      "eval_steps_per_second": 2.978,
      "eval_wer": 1.0,
      "step": 2700
    },
    {
      "epoch": 7.446808510638298,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 4.788908958435059,
      "eval_runtime": 208.0405,
      "eval_samples_per_second": 23.846,
      "eval_steps_per_second": 2.985,
      "eval_wer": 1.0,
      "step": 2800
    },
    {
      "epoch": 7.712765957446808,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 4.557297706604004,
      "eval_runtime": 206.7999,
      "eval_samples_per_second": 23.989,
      "eval_steps_per_second": 3.003,
      "eval_wer": 1.0,
      "step": 2900
    },
    {
      "epoch": 7.9787234042553195,
      "grad_norm": 9.712457656860352,
      "learning_rate": 7.1880000000000005e-06,
      "loss": 4.7448,
      "step": 3000
    },
    {
      "epoch": 7.9787234042553195,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 4.38894510269165,
      "eval_runtime": 208.3141,
      "eval_samples_per_second": 23.815,
      "eval_steps_per_second": 2.981,
      "eval_wer": 1.0,
      "step": 3000
    },
    {
      "epoch": 8.24468085106383,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 4.261370658874512,
      "eval_runtime": 206.872,
      "eval_samples_per_second": 23.981,
      "eval_steps_per_second": 3.002,
      "eval_wer": 1.0,
      "step": 3100
    },
    {
      "epoch": 8.51063829787234,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 4.196043014526367,
      "eval_runtime": 206.4767,
      "eval_samples_per_second": 24.027,
      "eval_steps_per_second": 3.008,
      "eval_wer": 1.0,
      "step": 3200
    },
    {
      "epoch": 8.77659574468085,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 4.139795303344727,
      "eval_runtime": 206.4082,
      "eval_samples_per_second": 24.035,
      "eval_steps_per_second": 3.009,
      "eval_wer": 1.0,
      "step": 3300
    },
    {
      "epoch": 9.042553191489361,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 4.1092000007629395,
      "eval_runtime": 205.1387,
      "eval_samples_per_second": 24.184,
      "eval_steps_per_second": 3.027,
      "eval_wer": 1.0,
      "step": 3400
    },
    {
      "epoch": 9.308510638297872,
      "grad_norm": 1.407253384590149,
      "learning_rate": 8.388e-06,
      "loss": 4.1253,
      "step": 3500
    },
    {
      "epoch": 9.308510638297872,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 4.091124534606934,
      "eval_runtime": 206.2573,
      "eval_samples_per_second": 24.052,
      "eval_steps_per_second": 3.011,
      "eval_wer": 1.0,
      "step": 3500
    },
    {
      "epoch": 9.574468085106384,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 4.085117816925049,
      "eval_runtime": 205.8565,
      "eval_samples_per_second": 24.099,
      "eval_steps_per_second": 3.017,
      "eval_wer": 1.0,
      "step": 3600
    },
    {
      "epoch": 9.840425531914894,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 4.070712089538574,
      "eval_runtime": 206.7954,
      "eval_samples_per_second": 23.99,
      "eval_steps_per_second": 3.003,
      "eval_wer": 1.0,
      "step": 3700
    },
    {
      "epoch": 10.106382978723405,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 4.062964916229248,
      "eval_runtime": 206.5725,
      "eval_samples_per_second": 24.016,
      "eval_steps_per_second": 3.006,
      "eval_wer": 1.0,
      "step": 3800
    },
    {
      "epoch": 10.372340425531915,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 4.058863639831543,
      "eval_runtime": 206.6395,
      "eval_samples_per_second": 24.008,
      "eval_steps_per_second": 3.005,
      "eval_wer": 1.0,
      "step": 3900
    },
    {
      "epoch": 10.638297872340425,
      "grad_norm": 0.9817385077476501,
      "learning_rate": 9.588e-06,
      "loss": 4.0399,
      "step": 4000
    },
    {
      "epoch": 10.638297872340425,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 4.0573530197143555,
      "eval_runtime": 207.8093,
      "eval_samples_per_second": 23.873,
      "eval_steps_per_second": 2.988,
      "eval_wer": 1.0,
      "step": 4000
    },
    {
      "epoch": 10.904255319148936,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 4.049480438232422,
      "eval_runtime": 207.6873,
      "eval_samples_per_second": 23.887,
      "eval_steps_per_second": 2.99,
      "eval_wer": 1.0,
      "step": 4100
    },
    {
      "epoch": 11.170212765957446,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 4.03674840927124,
      "eval_runtime": 207.2045,
      "eval_samples_per_second": 23.943,
      "eval_steps_per_second": 2.997,
      "eval_wer": 1.0,
      "step": 4200
    },
    {
      "epoch": 11.436170212765958,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 4.0297136306762695,
      "eval_runtime": 207.9114,
      "eval_samples_per_second": 23.861,
      "eval_steps_per_second": 2.987,
      "eval_wer": 1.0,
      "step": 4300
    },
    {
      "epoch": 11.702127659574469,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 4.01682186126709,
      "eval_runtime": 214.2137,
      "eval_samples_per_second": 23.159,
      "eval_steps_per_second": 2.899,
      "eval_wer": 1.0,
      "step": 4400
    },
    {
      "epoch": 11.96808510638298,
      "grad_norm": 1.0635989904403687,
      "learning_rate": 1.0787999999999999e-05,
      "loss": 4.0102,
      "step": 4500
    },
    {
      "epoch": 11.96808510638298,
      "eval_cer": 0.9999031273713612,
      "eval_loss": 4.000179290771484,
      "eval_runtime": 208.8628,
      "eval_samples_per_second": 23.752,
      "eval_steps_per_second": 2.973,
      "eval_wer": 1.0,
      "step": 4500
    },
    {
      "epoch": 12.23404255319149,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 3.9823410511016846,
      "eval_runtime": 207.9568,
      "eval_samples_per_second": 23.856,
      "eval_steps_per_second": 2.986,
      "eval_wer": 1.0,
      "step": 4600
    },
    {
      "epoch": 12.5,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 3.94744873046875,
      "eval_runtime": 207.1171,
      "eval_samples_per_second": 23.953,
      "eval_steps_per_second": 2.998,
      "eval_wer": 1.0,
      "step": 4700
    },
    {
      "epoch": 12.76595744680851,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 3.88703989982605,
      "eval_runtime": 207.9782,
      "eval_samples_per_second": 23.853,
      "eval_steps_per_second": 2.986,
      "eval_wer": 1.0,
      "step": 4800
    },
    {
      "epoch": 13.03191489361702,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 3.7933156490325928,
      "eval_runtime": 207.9844,
      "eval_samples_per_second": 23.853,
      "eval_steps_per_second": 2.986,
      "eval_wer": 1.0,
      "step": 4900
    },
    {
      "epoch": 13.297872340425531,
      "grad_norm": 2.227720260620117,
      "learning_rate": 1.1988000000000001e-05,
      "loss": 3.8616,
      "step": 5000
    },
    {
      "epoch": 13.297872340425531,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 3.6575610637664795,
      "eval_runtime": 207.9134,
      "eval_samples_per_second": 23.861,
      "eval_steps_per_second": 2.987,
      "eval_wer": 1.0,
      "step": 5000
    },
    {
      "epoch": 13.563829787234042,
      "eval_cer": 0.9998627637760951,
      "eval_loss": 3.4925012588500977,
      "eval_runtime": 208.4794,
      "eval_samples_per_second": 23.796,
      "eval_steps_per_second": 2.979,
      "eval_wer": 1.0,
      "step": 5100
    },
    {
      "epoch": 13.829787234042554,
      "eval_cer": 0.9999031273713612,
      "eval_loss": 3.2550227642059326,
      "eval_runtime": 207.4858,
      "eval_samples_per_second": 23.91,
      "eval_steps_per_second": 2.993,
      "eval_wer": 1.0,
      "step": 5200
    },
    {
      "epoch": 14.095744680851064,
      "eval_cer": 0.8301338456819026,
      "eval_loss": 2.8836495876312256,
      "eval_runtime": 208.4061,
      "eval_samples_per_second": 23.804,
      "eval_steps_per_second": 2.98,
      "eval_wer": 1.0,
      "step": 5300
    },
    {
      "epoch": 14.361702127659575,
      "eval_cer": 0.6170624989909101,
      "eval_loss": 2.521071672439575,
      "eval_runtime": 209.1476,
      "eval_samples_per_second": 23.72,
      "eval_steps_per_second": 2.969,
      "eval_wer": 1.0,
      "step": 5400
    },
    {
      "epoch": 14.627659574468085,
      "grad_norm": 8.615595817565918,
      "learning_rate": 1.3188e-05,
      "loss": 3.023,
      "step": 5500
    },
    {
      "epoch": 14.627659574468085,
      "eval_cer": 0.5481053328382066,
      "eval_loss": 2.2902443408966064,
      "eval_runtime": 208.2855,
      "eval_samples_per_second": 23.818,
      "eval_steps_per_second": 2.981,
      "eval_wer": 1.0,
      "step": 5500
    },
    {
      "epoch": 14.893617021276595,
      "eval_cer": 0.5078628283578475,
      "eval_loss": 2.1006453037261963,
      "eval_runtime": 208.3604,
      "eval_samples_per_second": 23.81,
      "eval_steps_per_second": 2.98,
      "eval_wer": 1.0,
      "step": 5600
    },
    {
      "epoch": 15.159574468085106,
      "eval_cer": 0.47841354925165896,
      "eval_loss": 1.9463908672332764,
      "eval_runtime": 208.5724,
      "eval_samples_per_second": 23.786,
      "eval_steps_per_second": 2.977,
      "eval_wer": 1.0,
      "step": 5700
    },
    {
      "epoch": 15.425531914893616,
      "eval_cer": 0.45966869561005536,
      "eval_loss": 1.8196361064910889,
      "eval_runtime": 206.9409,
      "eval_samples_per_second": 23.973,
      "eval_steps_per_second": 3.001,
      "eval_wer": 1.0,
      "step": 5800
    },
    {
      "epoch": 15.691489361702128,
      "eval_cer": 0.4237531685422284,
      "eval_loss": 1.6974730491638184,
      "eval_runtime": 206.9424,
      "eval_samples_per_second": 23.973,
      "eval_steps_per_second": 3.001,
      "eval_wer": 1.0,
      "step": 5900
    },
    {
      "epoch": 15.957446808510639,
      "grad_norm": 11.626580238342285,
      "learning_rate": 1.4388000000000002e-05,
      "loss": 1.9348,
      "step": 6000
    },
    {
      "epoch": 15.957446808510639,
      "eval_cer": 0.4093110741559972,
      "eval_loss": 1.6040183305740356,
      "eval_runtime": 205.6257,
      "eval_samples_per_second": 24.126,
      "eval_steps_per_second": 3.02,
      "eval_wer": 1.0,
      "step": 6000
    },
    {
      "epoch": 16.22340425531915,
      "eval_cer": 0.40211828147956796,
      "eval_loss": 1.5034863948822021,
      "eval_runtime": 206.5176,
      "eval_samples_per_second": 24.022,
      "eval_steps_per_second": 3.007,
      "eval_wer": 1.0,
      "step": 6100
    },
    {
      "epoch": 16.48936170212766,
      "eval_cer": 0.3929638180732034,
      "eval_loss": 1.42111337184906,
      "eval_runtime": 206.0968,
      "eval_samples_per_second": 24.071,
      "eval_steps_per_second": 3.013,
      "eval_wer": 1.0,
      "step": 6200
    },
    {
      "epoch": 16.75531914893617,
      "eval_cer": 0.3802250674072041,
      "eval_loss": 1.3529201745986938,
      "eval_runtime": 207.5275,
      "eval_samples_per_second": 23.905,
      "eval_steps_per_second": 2.992,
      "eval_wer": 1.0,
      "step": 6300
    },
    {
      "epoch": 17.02127659574468,
      "eval_cer": 0.3790545231444855,
      "eval_loss": 1.2795100212097168,
      "eval_runtime": 210.6768,
      "eval_samples_per_second": 23.548,
      "eval_steps_per_second": 2.948,
      "eval_wer": 1.0,
      "step": 6400
    },
    {
      "epoch": 17.28723404255319,
      "grad_norm": 8.535591125488281,
      "learning_rate": 1.5588e-05,
      "loss": 1.4128,
      "step": 6500
    },
    {
      "epoch": 17.28723404255319,
      "eval_cer": 0.3710706040008396,
      "eval_loss": 1.219308614730835,
      "eval_runtime": 207.1346,
      "eval_samples_per_second": 23.951,
      "eval_steps_per_second": 2.998,
      "eval_wer": 1.0,
      "step": 6500
    },
    {
      "epoch": 17.5531914893617,
      "eval_cer": 0.3673571532363531,
      "eval_loss": 1.1645617485046387,
      "eval_runtime": 207.4547,
      "eval_samples_per_second": 23.914,
      "eval_steps_per_second": 2.993,
      "eval_wer": 1.0,
      "step": 6600
    },
    {
      "epoch": 17.819148936170212,
      "eval_cer": 0.37062660445291185,
      "eval_loss": 1.1192774772644043,
      "eval_runtime": 209.6296,
      "eval_samples_per_second": 23.666,
      "eval_steps_per_second": 2.962,
      "eval_wer": 1.0,
      "step": 6700
    },
    {
      "epoch": 18.085106382978722,
      "eval_cer": 0.3605841419506918,
      "eval_loss": 1.066541075706482,
      "eval_runtime": 212.2728,
      "eval_samples_per_second": 23.371,
      "eval_steps_per_second": 2.925,
      "eval_wer": 1.0,
      "step": 6800
    },
    {
      "epoch": 18.351063829787233,
      "eval_cer": 0.3590422526115246,
      "eval_loss": 1.0244266986846924,
      "eval_runtime": 211.5467,
      "eval_samples_per_second": 23.451,
      "eval_steps_per_second": 2.936,
      "eval_wer": 0.9997984277363435,
      "step": 6900
    },
    {
      "epoch": 18.617021276595743,
      "grad_norm": 6.8628058433532715,
      "learning_rate": 1.6788e-05,
      "loss": 1.1012,
      "step": 7000
    },
    {
      "epoch": 18.617021276595743,
      "eval_cer": 0.35400487592230817,
      "eval_loss": 0.9863778352737427,
      "eval_runtime": 210.3223,
      "eval_samples_per_second": 23.588,
      "eval_steps_per_second": 2.953,
      "eval_wer": 1.0,
      "step": 7000
    },
    {
      "epoch": 18.882978723404257,
      "eval_cer": 0.3553691654423043,
      "eval_loss": 0.9577982425689697,
      "eval_runtime": 208.8403,
      "eval_samples_per_second": 23.755,
      "eval_steps_per_second": 2.974,
      "eval_wer": 1.0,
      "step": 7100
    },
    {
      "epoch": 19.148936170212767,
      "eval_cer": 0.350896879086814,
      "eval_loss": 0.9308760762214661,
      "eval_runtime": 211.4474,
      "eval_samples_per_second": 23.462,
      "eval_steps_per_second": 2.937,
      "eval_wer": 0.9997984277363435,
      "step": 7200
    },
    {
      "epoch": 19.414893617021278,
      "eval_cer": 0.3495083714096582,
      "eval_loss": 0.9070402383804321,
      "eval_runtime": 213.7352,
      "eval_samples_per_second": 23.211,
      "eval_steps_per_second": 2.905,
      "eval_wer": 1.0,
      "step": 7300
    },
    {
      "epoch": 19.680851063829788,
      "eval_cer": 0.3470461920984226,
      "eval_loss": 0.8692798018455505,
      "eval_runtime": 203.6485,
      "eval_samples_per_second": 24.361,
      "eval_steps_per_second": 3.049,
      "eval_wer": 0.9997984277363435,
      "step": 7400
    },
    {
      "epoch": 19.9468085106383,
      "grad_norm": 5.451691150665283,
      "learning_rate": 1.7988e-05,
      "loss": 0.9083,
      "step": 7500
    },
    {
      "epoch": 19.9468085106383,
      "eval_cer": 0.34493921242552916,
      "eval_loss": 0.8492410778999329,
      "eval_runtime": 202.6575,
      "eval_samples_per_second": 24.48,
      "eval_steps_per_second": 3.064,
      "eval_wer": 1.0,
      "step": 7500
    },
    {
      "epoch": 20.21276595744681,
      "eval_cer": 0.34490692154931624,
      "eval_loss": 0.8214272856712341,
      "eval_runtime": 204.0849,
      "eval_samples_per_second": 24.309,
      "eval_steps_per_second": 3.043,
      "eval_wer": 1.0,
      "step": 7600
    },
    {
      "epoch": 20.47872340425532,
      "eval_cer": 0.35004117086717146,
      "eval_loss": 0.821138322353363,
      "eval_runtime": 205.8454,
      "eval_samples_per_second": 24.101,
      "eval_steps_per_second": 3.017,
      "eval_wer": 1.0,
      "step": 7700
    },
    {
      "epoch": 20.74468085106383,
      "eval_cer": 0.34516524855901964,
      "eval_loss": 0.7964152693748474,
      "eval_runtime": 205.6355,
      "eval_samples_per_second": 24.125,
      "eval_steps_per_second": 3.02,
      "eval_wer": 1.0,
      "step": 7800
    },
    {
      "epoch": 21.01063829787234,
      "eval_cer": 0.34293717810032776,
      "eval_loss": 0.77970951795578,
      "eval_runtime": 207.2008,
      "eval_samples_per_second": 23.943,
      "eval_steps_per_second": 2.997,
      "eval_wer": 1.0,
      "step": 7900
    },
    {
      "epoch": 21.27659574468085,
      "grad_norm": 5.198563098907471,
      "learning_rate": 1.9188e-05,
      "loss": 0.7546,
      "step": 8000
    },
    {
      "epoch": 21.27659574468085,
      "eval_cer": 0.34003907196021765,
      "eval_loss": 0.7633516788482666,
      "eval_runtime": 205.0619,
      "eval_samples_per_second": 24.193,
      "eval_steps_per_second": 3.028,
      "eval_wer": 1.0,
      "step": 8000
    },
    {
      "epoch": 21.54255319148936,
      "eval_cer": 0.33842452814957136,
      "eval_loss": 0.7471381425857544,
      "eval_runtime": 211.6646,
      "eval_samples_per_second": 23.438,
      "eval_steps_per_second": 2.934,
      "eval_wer": 1.0,
      "step": 8100
    },
    {
      "epoch": 21.80851063829787,
      "eval_cer": 0.33775449246815314,
      "eval_loss": 0.7400262355804443,
      "eval_runtime": 213.7944,
      "eval_samples_per_second": 23.205,
      "eval_steps_per_second": 2.905,
      "eval_wer": 1.0,
      "step": 8200
    },
    {
      "epoch": 22.074468085106382,
      "eval_cer": 0.33898961848329756,
      "eval_loss": 0.7214083075523376,
      "eval_runtime": 205.5478,
      "eval_samples_per_second": 24.135,
      "eval_steps_per_second": 3.021,
      "eval_wer": 1.0,
      "step": 8300
    },
    {
      "epoch": 22.340425531914892,
      "eval_cer": 0.33750423817750297,
      "eval_loss": 0.7061555981636047,
      "eval_runtime": 211.7804,
      "eval_samples_per_second": 23.425,
      "eval_steps_per_second": 2.932,
      "eval_wer": 0.9997984277363435,
      "step": 8400
    },
    {
      "epoch": 22.606382978723403,
      "grad_norm": 6.07712459564209,
      "learning_rate": 2.0388e-05,
      "loss": 0.651,
      "step": 8500
    },
    {
      "epoch": 22.606382978723403,
      "eval_cer": 0.33436395046579587,
      "eval_loss": 0.6972677111625671,
      "eval_runtime": 204.7478,
      "eval_samples_per_second": 24.23,
      "eval_steps_per_second": 3.033,
      "eval_wer": 1.0,
      "step": 8500
    },
    {
      "epoch": 22.872340425531917,
      "eval_cer": 0.33436395046579587,
      "eval_loss": 0.6929803490638733,
      "eval_runtime": 205.6125,
      "eval_samples_per_second": 24.128,
      "eval_steps_per_second": 3.02,
      "eval_wer": 0.9997984277363435,
      "step": 8600
    },
    {
      "epoch": 23.138297872340427,
      "eval_cer": 0.33500169527100115,
      "eval_loss": 0.6829419136047363,
      "eval_runtime": 207.2537,
      "eval_samples_per_second": 23.937,
      "eval_steps_per_second": 2.996,
      "eval_wer": 1.0,
      "step": 8700
    },
    {
      "epoch": 23.404255319148938,
      "eval_cer": 0.3331691880459176,
      "eval_loss": 0.6683320999145508,
      "eval_runtime": 206.9839,
      "eval_samples_per_second": 23.968,
      "eval_steps_per_second": 3.0,
      "eval_wer": 1.0,
      "step": 8800
    },
    {
      "epoch": 23.670212765957448,
      "eval_cer": 0.33223275263574276,
      "eval_loss": 0.6596328616142273,
      "eval_runtime": 216.6308,
      "eval_samples_per_second": 22.901,
      "eval_steps_per_second": 2.867,
      "eval_wer": 0.9997984277363435,
      "step": 8900
    },
    {
      "epoch": 23.93617021276596,
      "grad_norm": 2.6311779022216797,
      "learning_rate": 2.1588e-05,
      "loss": 0.5868,
      "step": 9000
    },
    {
      "epoch": 23.93617021276596,
      "eval_cer": 0.3320712982546781,
      "eval_loss": 0.6764070987701416,
      "eval_runtime": 207.8654,
      "eval_samples_per_second": 23.866,
      "eval_steps_per_second": 2.988,
      "eval_wer": 1.0,
      "step": 9000
    },
    {
      "epoch": 24.20212765957447,
      "eval_cer": 0.3308280995204805,
      "eval_loss": 0.6634973883628845,
      "eval_runtime": 207.5025,
      "eval_samples_per_second": 23.908,
      "eval_steps_per_second": 2.993,
      "eval_wer": 0.9997984277363435,
      "step": 9100
    },
    {
      "epoch": 24.46808510638298,
      "eval_cer": 0.3323699888596477,
      "eval_loss": 0.6560051441192627,
      "eval_runtime": 207.3198,
      "eval_samples_per_second": 23.929,
      "eval_steps_per_second": 2.995,
      "eval_wer": 1.0,
      "step": 9200
    },
    {
      "epoch": 24.73404255319149,
      "eval_cer": 0.3290198104525566,
      "eval_loss": 0.6412243843078613,
      "eval_runtime": 207.1443,
      "eval_samples_per_second": 23.949,
      "eval_steps_per_second": 2.998,
      "eval_wer": 1.0,
      "step": 9300
    },
    {
      "epoch": 25.0,
      "eval_cer": 0.33074737232994816,
      "eval_loss": 0.6322699785232544,
      "eval_runtime": 205.6853,
      "eval_samples_per_second": 24.119,
      "eval_steps_per_second": 3.019,
      "eval_wer": 1.0,
      "step": 9400
    },
    {
      "epoch": 25.0,
      "step": 9400,
      "total_flos": 1.314337503834538e+19,
      "train_loss": 5.653992247885846,
      "train_runtime": 50640.2984,
      "train_samples_per_second": 5.94,
      "train_steps_per_second": 0.186
    }
  ],
  "logging_steps": 500,
  "max_steps": 9400,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 25,
  "save_steps": 400,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.314337503834538e+19,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}