{
  "best_metric": 10.959838224228744,
  "best_model_checkpoint": "./checkpoint-2000",
  "epoch": 6.1285,
  "global_step": 4000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 4.0000000000000003e-07,
      "loss": 2.032,
      "step": 25
    },
    {
      "epoch": 0.01,
      "learning_rate": 9.000000000000001e-07,
      "loss": 1.6426,
      "step": 50
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.4000000000000001e-06,
      "loss": 1.2077,
      "step": 75
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9000000000000002e-06,
      "loss": 1.1075,
      "step": 100
    },
    {
      "epoch": 0.03,
      "learning_rate": 2.4000000000000003e-06,
      "loss": 1.3237,
      "step": 125
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.9e-06,
      "loss": 1.3296,
      "step": 150
    },
    {
      "epoch": 0.04,
      "learning_rate": 3.4000000000000005e-06,
      "loss": 1.1841,
      "step": 175
    },
    {
      "epoch": 0.05,
      "learning_rate": 3.900000000000001e-06,
      "loss": 0.9476,
      "step": 200
    },
    {
      "epoch": 0.06,
      "learning_rate": 4.4e-06,
      "loss": 0.8653,
      "step": 225
    },
    {
      "epoch": 0.06,
      "learning_rate": 4.9000000000000005e-06,
      "loss": 0.6178,
      "step": 250
    },
    {
      "epoch": 0.07,
      "learning_rate": 5.400000000000001e-06,
      "loss": 0.3619,
      "step": 275
    },
    {
      "epoch": 0.07,
      "learning_rate": 5.9e-06,
      "loss": 0.3445,
      "step": 300
    },
    {
      "epoch": 0.08,
      "learning_rate": 6.4000000000000006e-06,
      "loss": 0.2467,
      "step": 325
    },
    {
      "epoch": 0.09,
      "learning_rate": 6.9e-06,
      "loss": 0.2312,
      "step": 350
    },
    {
      "epoch": 0.09,
      "learning_rate": 7.4e-06,
      "loss": 0.3341,
      "step": 375
    },
    {
      "epoch": 0.1,
      "learning_rate": 7.9e-06,
      "loss": 0.4471,
      "step": 400
    },
    {
      "epoch": 0.11,
      "learning_rate": 8.400000000000001e-06,
      "loss": 0.3353,
      "step": 425
    },
    {
      "epoch": 0.11,
      "learning_rate": 8.900000000000001e-06,
      "loss": 0.275,
      "step": 450
    },
    {
      "epoch": 0.12,
      "learning_rate": 9.4e-06,
      "loss": 0.2695,
      "step": 475
    },
    {
      "epoch": 0.12,
      "learning_rate": 9.9e-06,
      "loss": 0.2381,
      "step": 500
    },
    {
      "epoch": 0.12,
      "eval_loss": 0.2625342905521393,
      "eval_runtime": 1146.7498,
      "eval_samples_per_second": 3.247,
      "eval_steps_per_second": 0.102,
      "eval_wer": 11.487725733634312,
      "step": 500
    },
    {
      "epoch": 0.13,
      "learning_rate": 9.942857142857145e-06,
      "loss": 0.2463,
      "step": 525
    },
    {
      "epoch": 0.14,
      "learning_rate": 9.871428571428572e-06,
      "loss": 0.2576,
      "step": 550
    },
    {
      "epoch": 0.14,
      "learning_rate": 9.800000000000001e-06,
      "loss": 0.294,
      "step": 575
    },
    {
      "epoch": 1.0,
      "learning_rate": 9.72857142857143e-06,
      "loss": 0.1854,
      "step": 600
    },
    {
      "epoch": 1.01,
      "learning_rate": 9.657142857142859e-06,
      "loss": 0.1302,
      "step": 625
    },
    {
      "epoch": 1.02,
      "learning_rate": 9.585714285714286e-06,
      "loss": 0.1011,
      "step": 650
    },
    {
      "epoch": 1.02,
      "learning_rate": 9.514285714285715e-06,
      "loss": 0.1088,
      "step": 675
    },
    {
      "epoch": 1.03,
      "learning_rate": 9.442857142857144e-06,
      "loss": 0.2162,
      "step": 700
    },
    {
      "epoch": 1.04,
      "learning_rate": 9.371428571428572e-06,
      "loss": 0.354,
      "step": 725
    },
    {
      "epoch": 1.04,
      "learning_rate": 9.3e-06,
      "loss": 0.4993,
      "step": 750
    },
    {
      "epoch": 1.05,
      "learning_rate": 9.22857142857143e-06,
      "loss": 0.2401,
      "step": 775
    },
    {
      "epoch": 1.05,
      "learning_rate": 9.157142857142857e-06,
      "loss": 0.1652,
      "step": 800
    },
    {
      "epoch": 1.06,
      "learning_rate": 9.085714285714286e-06,
      "loss": 0.1875,
      "step": 825
    },
    {
      "epoch": 1.07,
      "learning_rate": 9.014285714285715e-06,
      "loss": 0.1658,
      "step": 850
    },
    {
      "epoch": 1.07,
      "learning_rate": 8.942857142857142e-06,
      "loss": 0.1099,
      "step": 875
    },
    {
      "epoch": 1.08,
      "learning_rate": 8.871428571428571e-06,
      "loss": 0.3382,
      "step": 900
    },
    {
      "epoch": 1.09,
      "learning_rate": 8.8e-06,
      "loss": 0.2355,
      "step": 925
    },
    {
      "epoch": 1.09,
      "learning_rate": 8.72857142857143e-06,
      "loss": 0.2217,
      "step": 950
    },
    {
      "epoch": 1.1,
      "learning_rate": 8.657142857142858e-06,
      "loss": 0.1444,
      "step": 975
    },
    {
      "epoch": 1.1,
      "learning_rate": 8.585714285714286e-06,
      "loss": 0.1332,
      "step": 1000
    },
    {
      "epoch": 1.1,
      "eval_loss": 0.2451411932706833,
      "eval_runtime": 1142.5886,
      "eval_samples_per_second": 3.259,
      "eval_steps_per_second": 0.102,
      "eval_wer": 11.407778404815652,
      "step": 1000
    },
    {
      "epoch": 1.11,
      "learning_rate": 8.514285714285715e-06,
      "loss": 0.1014,
      "step": 1025
    },
    {
      "epoch": 1.12,
      "learning_rate": 8.442857142857144e-06,
      "loss": 0.1186,
      "step": 1050
    },
    {
      "epoch": 1.12,
      "learning_rate": 8.371428571428573e-06,
      "loss": 0.118,
      "step": 1075
    },
    {
      "epoch": 1.13,
      "learning_rate": 8.3e-06,
      "loss": 0.245,
      "step": 1100
    },
    {
      "epoch": 1.14,
      "learning_rate": 8.22857142857143e-06,
      "loss": 0.3207,
      "step": 1125
    },
    {
      "epoch": 1.14,
      "learning_rate": 8.157142857142858e-06,
      "loss": 0.2316,
      "step": 1150
    },
    {
      "epoch": 2.0,
      "learning_rate": 8.085714285714287e-06,
      "loss": 0.1327,
      "step": 1175
    },
    {
      "epoch": 2.01,
      "learning_rate": 8.014285714285715e-06,
      "loss": 0.0643,
      "step": 1200
    },
    {
      "epoch": 2.02,
      "learning_rate": 7.942857142857144e-06,
      "loss": 0.0719,
      "step": 1225
    },
    {
      "epoch": 2.02,
      "learning_rate": 7.871428571428573e-06,
      "loss": 0.0497,
      "step": 1250
    },
    {
      "epoch": 2.03,
      "learning_rate": 7.800000000000002e-06,
      "loss": 0.081,
      "step": 1275
    },
    {
      "epoch": 2.03,
      "learning_rate": 7.72857142857143e-06,
      "loss": 0.1056,
      "step": 1300
    },
    {
      "epoch": 2.04,
      "learning_rate": 7.657142857142858e-06,
      "loss": 0.0932,
      "step": 1325
    },
    {
      "epoch": 2.05,
      "learning_rate": 7.585714285714286e-06,
      "loss": 0.0709,
      "step": 1350
    },
    {
      "epoch": 2.05,
      "learning_rate": 7.514285714285715e-06,
      "loss": 0.0529,
      "step": 1375
    },
    {
      "epoch": 2.06,
      "learning_rate": 7.442857142857144e-06,
      "loss": 0.0409,
      "step": 1400
    },
    {
      "epoch": 2.07,
      "learning_rate": 7.371428571428571e-06,
      "loss": 0.0405,
      "step": 1425
    },
    {
      "epoch": 2.07,
      "learning_rate": 7.3e-06,
      "loss": 0.1962,
      "step": 1450
    },
    {
      "epoch": 2.08,
      "learning_rate": 7.2285714285714294e-06,
      "loss": 0.1471,
      "step": 1475
    },
    {
      "epoch": 2.08,
      "learning_rate": 7.1571428571428584e-06,
      "loss": 0.1097,
      "step": 1500
    },
    {
      "epoch": 2.08,
      "eval_loss": 0.26101914048194885,
      "eval_runtime": 1140.7964,
      "eval_samples_per_second": 3.264,
      "eval_steps_per_second": 0.103,
      "eval_wer": 11.535929270127916,
      "step": 1500
    },
    {
      "epoch": 2.09,
      "learning_rate": 7.085714285714286e-06,
      "loss": 0.1866,
      "step": 1525
    },
    {
      "epoch": 2.1,
      "learning_rate": 7.014285714285715e-06,
      "loss": 0.0951,
      "step": 1550
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.942857142857144e-06,
      "loss": 0.1809,
      "step": 1575
    },
    {
      "epoch": 2.11,
      "learning_rate": 6.871428571428572e-06,
      "loss": 0.3334,
      "step": 1600
    },
    {
      "epoch": 2.12,
      "learning_rate": 6.800000000000001e-06,
      "loss": 0.2182,
      "step": 1625
    },
    {
      "epoch": 2.12,
      "learning_rate": 6.728571428571429e-06,
      "loss": 0.1133,
      "step": 1650
    },
    {
      "epoch": 2.13,
      "learning_rate": 6.657142857142857e-06,
      "loss": 0.1032,
      "step": 1675
    },
    {
      "epoch": 2.13,
      "learning_rate": 6.585714285714286e-06,
      "loss": 0.0871,
      "step": 1700
    },
    {
      "epoch": 2.14,
      "learning_rate": 6.514285714285715e-06,
      "loss": 0.0508,
      "step": 1725
    },
    {
      "epoch": 3.0,
      "learning_rate": 6.442857142857143e-06,
      "loss": 0.0396,
      "step": 1750
    },
    {
      "epoch": 3.01,
      "learning_rate": 6.371428571428572e-06,
      "loss": 0.0651,
      "step": 1775
    },
    {
      "epoch": 3.01,
      "learning_rate": 6.300000000000001e-06,
      "loss": 0.0872,
      "step": 1800
    },
    {
      "epoch": 3.02,
      "learning_rate": 6.22857142857143e-06,
      "loss": 0.0515,
      "step": 1825
    },
    {
      "epoch": 3.03,
      "learning_rate": 6.157142857142858e-06,
      "loss": 0.0264,
      "step": 1850
    },
    {
      "epoch": 3.03,
      "learning_rate": 6.085714285714286e-06,
      "loss": 0.0161,
      "step": 1875
    },
    {
      "epoch": 3.04,
      "learning_rate": 6.014285714285715e-06,
      "loss": 0.0169,
      "step": 1900
    },
    {
      "epoch": 3.05,
      "learning_rate": 5.942857142857143e-06,
      "loss": 0.0451,
      "step": 1925
    },
    {
      "epoch": 3.05,
      "learning_rate": 5.871428571428572e-06,
      "loss": 0.0589,
      "step": 1950
    },
    {
      "epoch": 3.06,
      "learning_rate": 5.8e-06,
      "loss": 0.0307,
      "step": 1975
    },
    {
      "epoch": 3.06,
      "learning_rate": 5.7285714285714285e-06,
      "loss": 0.0412,
      "step": 2000
    },
    {
      "epoch": 3.06,
      "eval_loss": 0.28039729595184326,
      "eval_runtime": 1144.8222,
      "eval_samples_per_second": 3.253,
      "eval_steps_per_second": 0.102,
      "eval_wer": 10.959838224228744,
      "step": 2000
    },
    {
      "epoch": 3.07,
      "learning_rate": 5.6571428571428576e-06,
      "loss": 0.0394,
      "step": 2025
    },
    {
      "epoch": 3.08,
      "learning_rate": 5.5857142857142866e-06,
      "loss": 0.0254,
      "step": 2050
    },
    {
      "epoch": 3.08,
      "learning_rate": 5.514285714285714e-06,
      "loss": 0.033,
      "step": 2075
    },
    {
      "epoch": 3.09,
      "learning_rate": 5.442857142857143e-06,
      "loss": 0.0405,
      "step": 2100
    },
    {
      "epoch": 3.1,
      "learning_rate": 5.371428571428572e-06,
      "loss": 0.0369,
      "step": 2125
    },
    {
      "epoch": 3.1,
      "learning_rate": 5.300000000000001e-06,
      "loss": 0.0594,
      "step": 2150
    },
    {
      "epoch": 3.11,
      "learning_rate": 5.22857142857143e-06,
      "loss": 0.1956,
      "step": 2175
    },
    {
      "epoch": 3.11,
      "learning_rate": 5.157142857142857e-06,
      "loss": 0.11,
      "step": 2200
    },
    {
      "epoch": 3.12,
      "learning_rate": 5.085714285714286e-06,
      "loss": 0.0453,
      "step": 2225
    },
    {
      "epoch": 3.13,
      "learning_rate": 5.014285714285715e-06,
      "loss": 0.0313,
      "step": 2250
    },
    {
      "epoch": 3.13,
      "learning_rate": 4.9428571428571435e-06,
      "loss": 0.0321,
      "step": 2275
    },
    {
      "epoch": 3.14,
      "learning_rate": 4.871428571428572e-06,
      "loss": 0.0777,
      "step": 2300
    },
    {
      "epoch": 4.0,
      "learning_rate": 4.800000000000001e-06,
      "loss": 0.1055,
      "step": 2325
    },
    {
      "epoch": 4.01,
      "learning_rate": 4.728571428571429e-06,
      "loss": 0.0103,
      "step": 2350
    },
    {
      "epoch": 4.01,
      "learning_rate": 4.657142857142857e-06,
      "loss": 0.0082,
      "step": 2375
    },
    {
      "epoch": 4.02,
      "learning_rate": 4.585714285714286e-06,
      "loss": 0.0161,
      "step": 2400
    },
    {
      "epoch": 4.03,
      "learning_rate": 4.514285714285714e-06,
      "loss": 0.0105,
      "step": 2425
    },
    {
      "epoch": 4.03,
      "learning_rate": 4.442857142857143e-06,
      "loss": 0.0085,
      "step": 2450
    },
    {
      "epoch": 4.04,
      "learning_rate": 4.371428571428572e-06,
      "loss": 0.0448,
      "step": 2475
    },
    {
      "epoch": 4.04,
      "learning_rate": 4.3e-06,
      "loss": 0.0219,
      "step": 2500
    },
    {
      "epoch": 4.04,
      "eval_loss": 0.342590868473053,
      "eval_runtime": 1102.6593,
      "eval_samples_per_second": 3.377,
      "eval_steps_per_second": 0.106,
      "eval_wer": 10.96689240030098,
      "step": 2500
    },
    {
      "epoch": 4.05,
      "learning_rate": 4.228571428571429e-06,
      "loss": 0.028,
      "step": 2525
    },
    {
      "epoch": 4.06,
      "learning_rate": 4.1571428571428575e-06,
      "loss": 0.0989,
      "step": 2550
    },
    {
      "epoch": 4.06,
      "learning_rate": 4.0857142857142865e-06,
      "loss": 0.0636,
      "step": 2575
    },
    {
      "epoch": 4.07,
      "learning_rate": 4.014285714285715e-06,
      "loss": 0.0335,
      "step": 2600
    },
    {
      "epoch": 4.08,
      "learning_rate": 3.942857142857143e-06,
      "loss": 0.0533,
      "step": 2625
    },
    {
      "epoch": 4.08,
      "learning_rate": 3.871428571428572e-06,
      "loss": 0.0238,
      "step": 2650
    },
    {
      "epoch": 4.09,
      "learning_rate": 3.8000000000000005e-06,
      "loss": 0.019,
      "step": 2675
    },
    {
      "epoch": 4.09,
      "learning_rate": 3.7285714285714286e-06,
      "loss": 0.0132,
      "step": 2700
    },
    {
      "epoch": 4.1,
      "learning_rate": 3.6571428571428576e-06,
      "loss": 0.0116,
      "step": 2725
    },
    {
      "epoch": 4.11,
      "learning_rate": 3.5857142857142862e-06,
      "loss": 0.0211,
      "step": 2750
    },
    {
      "epoch": 4.11,
      "learning_rate": 3.5142857142857144e-06,
      "loss": 0.0285,
      "step": 2775
    },
    {
      "epoch": 4.12,
      "learning_rate": 3.4428571428571434e-06,
      "loss": 0.0126,
      "step": 2800
    },
    {
      "epoch": 4.13,
      "learning_rate": 3.3714285714285716e-06,
      "loss": 0.0118,
      "step": 2825
    },
    {
      "epoch": 4.13,
      "learning_rate": 3.3000000000000006e-06,
      "loss": 0.0108,
      "step": 2850
    },
    {
      "epoch": 4.14,
      "learning_rate": 3.2285714285714288e-06,
      "loss": 0.0162,
      "step": 2875
    },
    {
      "epoch": 4.14,
      "learning_rate": 3.1571428571428573e-06,
      "loss": 0.0168,
      "step": 2900
    },
    {
      "epoch": 5.0,
      "learning_rate": 3.085714285714286e-06,
      "loss": 0.0584,
      "step": 2925
    },
    {
      "epoch": 5.01,
      "learning_rate": 3.0142857142857145e-06,
      "loss": 0.03,
      "step": 2950
    },
    {
      "epoch": 5.02,
      "learning_rate": 2.9428571428571427e-06,
      "loss": 0.0156,
      "step": 2975
    },
    {
      "epoch": 5.02,
      "learning_rate": 2.8714285714285717e-06,
      "loss": 0.0139,
      "step": 3000
    },
    {
      "epoch": 5.02,
      "eval_loss": 0.35033050179481506,
      "eval_runtime": 1072.8874,
      "eval_samples_per_second": 3.471,
      "eval_steps_per_second": 0.109,
      "eval_wer": 11.332533860045146,
      "step": 3000
    },
    {
      "epoch": 5.03,
      "learning_rate": 2.8000000000000003e-06,
      "loss": 0.0066,
      "step": 3025
    },
    {
      "epoch": 5.04,
      "learning_rate": 2.728571428571429e-06,
      "loss": 0.0044,
      "step": 3050
    },
    {
      "epoch": 5.04,
      "learning_rate": 2.6571428571428575e-06,
      "loss": 0.0043,
      "step": 3075
    },
    {
      "epoch": 5.05,
      "learning_rate": 2.5857142857142856e-06,
      "loss": 0.0028,
      "step": 3100
    },
    {
      "epoch": 5.05,
      "learning_rate": 2.5142857142857147e-06,
      "loss": 0.0026,
      "step": 3125
    },
    {
      "epoch": 5.06,
      "learning_rate": 2.442857142857143e-06,
      "loss": 0.013,
      "step": 3150
    },
    {
      "epoch": 5.07,
      "learning_rate": 2.371428571428572e-06,
      "loss": 0.0118,
      "step": 3175
    },
    {
      "epoch": 5.07,
      "learning_rate": 2.3000000000000004e-06,
      "loss": 0.0036,
      "step": 3200
    },
    {
      "epoch": 5.08,
      "learning_rate": 2.228571428571429e-06,
      "loss": 0.02,
      "step": 3225
    },
    {
      "epoch": 5.09,
      "learning_rate": 2.157142857142857e-06,
      "loss": 0.0202,
      "step": 3250
    },
    {
      "epoch": 5.09,
      "learning_rate": 2.0857142857142858e-06,
      "loss": 0.0096,
      "step": 3275
    },
    {
      "epoch": 5.1,
      "learning_rate": 2.0142857142857144e-06,
      "loss": 0.0063,
      "step": 3300
    },
    {
      "epoch": 5.11,
      "learning_rate": 1.942857142857143e-06,
      "loss": 0.0068,
      "step": 3325
    },
    {
      "epoch": 5.11,
      "learning_rate": 1.8714285714285715e-06,
      "loss": 0.0027,
      "step": 3350
    },
    {
      "epoch": 5.12,
      "learning_rate": 1.8000000000000001e-06,
      "loss": 0.003,
      "step": 3375
    },
    {
      "epoch": 5.12,
      "learning_rate": 1.7285714285714287e-06,
      "loss": 0.0054,
      "step": 3400
    },
    {
      "epoch": 5.13,
      "learning_rate": 1.657142857142857e-06,
      "loss": 0.0059,
      "step": 3425
    },
    {
      "epoch": 5.14,
      "learning_rate": 1.5857142857142857e-06,
      "loss": 0.0041,
      "step": 3450
    },
    {
      "epoch": 5.14,
      "learning_rate": 1.5142857142857145e-06,
      "loss": 0.0051,
      "step": 3475
    },
    {
      "epoch": 6.0,
      "learning_rate": 1.442857142857143e-06,
      "loss": 0.0086,
      "step": 3500
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.3761499524116516,
      "eval_runtime": 1251.4055,
      "eval_samples_per_second": 2.976,
      "eval_steps_per_second": 0.093,
      "eval_wer": 11.022150112866818,
      "step": 3500
    },
    {
      "epoch": 6.01,
      "learning_rate": 1.3714285714285717e-06,
      "loss": 0.0041,
      "step": 3525
    },
    {
      "epoch": 6.02,
      "learning_rate": 1.3e-06,
      "loss": 0.0021,
      "step": 3550
    },
    {
      "epoch": 6.02,
      "learning_rate": 1.2285714285714286e-06,
      "loss": 0.002,
      "step": 3575
    },
    {
      "epoch": 6.03,
      "learning_rate": 1.1571428571428572e-06,
      "loss": 0.003,
      "step": 3600
    },
    {
      "epoch": 6.03,
      "learning_rate": 1.0857142857142858e-06,
      "loss": 0.0028,
      "step": 3625
    },
    {
      "epoch": 6.04,
      "learning_rate": 1.0142857142857144e-06,
      "loss": 0.0023,
      "step": 3650
    },
    {
      "epoch": 6.05,
      "learning_rate": 9.42857142857143e-07,
      "loss": 0.0022,
      "step": 3675
    },
    {
      "epoch": 6.05,
      "learning_rate": 8.714285714285716e-07,
      "loss": 0.0026,
      "step": 3700
    },
    {
      "epoch": 6.06,
      "learning_rate": 8.000000000000001e-07,
      "loss": 0.0027,
      "step": 3725
    },
    {
      "epoch": 6.07,
      "learning_rate": 7.285714285714287e-07,
      "loss": 0.003,
      "step": 3750
    },
    {
      "epoch": 6.07,
      "learning_rate": 6.571428571428571e-07,
      "loss": 0.0016,
      "step": 3775
    },
    {
      "epoch": 6.08,
      "learning_rate": 5.857142857142857e-07,
      "loss": 0.0173,
      "step": 3800
    },
    {
      "epoch": 6.08,
      "learning_rate": 5.142857142857143e-07,
      "loss": 0.0208,
      "step": 3825
    },
    {
      "epoch": 6.09,
      "learning_rate": 4.4285714285714286e-07,
      "loss": 0.0164,
      "step": 3850
    },
    {
      "epoch": 6.1,
      "learning_rate": 3.7142857142857145e-07,
      "loss": 0.0073,
      "step": 3875
    },
    {
      "epoch": 6.1,
      "learning_rate": 3.0000000000000004e-07,
      "loss": 0.0018,
      "step": 3900
    },
    {
      "epoch": 6.11,
      "learning_rate": 2.285714285714286e-07,
      "loss": 0.0014,
      "step": 3925
    },
    {
      "epoch": 6.12,
      "learning_rate": 1.5714285714285717e-07,
      "loss": 0.002,
      "step": 3950
    },
    {
      "epoch": 6.12,
      "learning_rate": 8.571428571428573e-08,
      "loss": 0.0026,
      "step": 3975
    },
    {
      "epoch": 6.13,
      "learning_rate": 1.4285714285714288e-08,
      "loss": 0.0017,
      "step": 4000
    },
    {
      "epoch": 6.13,
      "eval_loss": 0.37628626823425293,
      "eval_runtime": 1164.4141,
      "eval_samples_per_second": 3.198,
      "eval_steps_per_second": 0.1,
      "eval_wer": 11.283154627539503,
      "step": 4000
    },
    {
      "epoch": 6.13,
      "step": 4000,
      "total_flos": 1.3058843912699904e+20,
      "train_loss": 0.16058776646107437,
      "train_runtime": 46179.6845,
      "train_samples_per_second": 2.772,
      "train_steps_per_second": 0.087
    }
  ],
  "max_steps": 4000,
  "num_train_epochs": 9223372036854775807,
  "total_flos": 1.3058843912699904e+20,
  "trial_name": null,
  "trial_params": null
}