{
  "best_metric": 6.048320913895545,
  "best_model_checkpoint": "./checkpoint-4000",
  "epoch": 1.0,
  "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": 4.6000000000000004e-07,
      "loss": 0.6908,
      "step": 25
    },
    {
      "epoch": 0.01,
      "learning_rate": 9.600000000000001e-07,
      "loss": 0.4455,
      "step": 50
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.44e-06,
      "loss": 0.343,
      "step": 75
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.94e-06,
      "loss": 0.3054,
      "step": 100
    },
    {
      "epoch": 0.03,
      "learning_rate": 2.4400000000000004e-06,
      "loss": 0.2376,
      "step": 125
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.9400000000000002e-06,
      "loss": 0.1798,
      "step": 150
    },
    {
      "epoch": 0.04,
      "learning_rate": 3.44e-06,
      "loss": 0.1613,
      "step": 175
    },
    {
      "epoch": 0.05,
      "learning_rate": 3.94e-06,
      "loss": 0.1895,
      "step": 200
    },
    {
      "epoch": 0.06,
      "learning_rate": 4.440000000000001e-06,
      "loss": 0.1222,
      "step": 225
    },
    {
      "epoch": 0.06,
      "learning_rate": 4.94e-06,
      "loss": 0.1342,
      "step": 250
    },
    {
      "epoch": 0.07,
      "learning_rate": 5.4400000000000004e-06,
      "loss": 0.1984,
      "step": 275
    },
    {
      "epoch": 0.07,
      "learning_rate": 5.94e-06,
      "loss": 0.2668,
      "step": 300
    },
    {
      "epoch": 0.08,
      "learning_rate": 6.440000000000001e-06,
      "loss": 0.1933,
      "step": 325
    },
    {
      "epoch": 0.09,
      "learning_rate": 6.9400000000000005e-06,
      "loss": 0.1618,
      "step": 350
    },
    {
      "epoch": 0.09,
      "learning_rate": 7.440000000000001e-06,
      "loss": 0.135,
      "step": 375
    },
    {
      "epoch": 0.1,
      "learning_rate": 7.94e-06,
      "loss": 0.1321,
      "step": 400
    },
    {
      "epoch": 0.11,
      "learning_rate": 8.44e-06,
      "loss": 0.1581,
      "step": 425
    },
    {
      "epoch": 0.11,
      "learning_rate": 8.94e-06,
      "loss": 0.172,
      "step": 450
    },
    {
      "epoch": 0.12,
      "learning_rate": 9.440000000000001e-06,
      "loss": 0.0915,
      "step": 475
    },
    {
      "epoch": 0.12,
      "learning_rate": 9.940000000000001e-06,
      "loss": 0.0771,
      "step": 500
    },
    {
      "epoch": 0.13,
      "learning_rate": 9.937142857142858e-06,
      "loss": 0.1758,
      "step": 525
    },
    {
      "epoch": 0.14,
      "learning_rate": 9.865714285714285e-06,
      "loss": 0.096,
      "step": 550
    },
    {
      "epoch": 0.14,
      "learning_rate": 9.794285714285714e-06,
      "loss": 0.0934,
      "step": 575
    },
    {
      "epoch": 0.15,
      "learning_rate": 9.722857142857143e-06,
      "loss": 0.1448,
      "step": 600
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.651428571428572e-06,
      "loss": 0.1245,
      "step": 625
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.58e-06,
      "loss": 0.186,
      "step": 650
    },
    {
      "epoch": 0.17,
      "learning_rate": 9.508571428571429e-06,
      "loss": 0.2251,
      "step": 675
    },
    {
      "epoch": 0.17,
      "learning_rate": 9.437142857142858e-06,
      "loss": 0.1351,
      "step": 700
    },
    {
      "epoch": 0.18,
      "learning_rate": 9.365714285714287e-06,
      "loss": 0.1062,
      "step": 725
    },
    {
      "epoch": 0.19,
      "learning_rate": 9.294285714285714e-06,
      "loss": 0.1261,
      "step": 750
    },
    {
      "epoch": 0.19,
      "learning_rate": 9.222857142857143e-06,
      "loss": 0.2465,
      "step": 775
    },
    {
      "epoch": 0.2,
      "learning_rate": 9.151428571428572e-06,
      "loss": 0.2123,
      "step": 800
    },
    {
      "epoch": 0.21,
      "learning_rate": 9.080000000000001e-06,
      "loss": 0.104,
      "step": 825
    },
    {
      "epoch": 0.21,
      "learning_rate": 9.00857142857143e-06,
      "loss": 0.1063,
      "step": 850
    },
    {
      "epoch": 0.22,
      "learning_rate": 8.937142857142857e-06,
      "loss": 0.0935,
      "step": 875
    },
    {
      "epoch": 0.23,
      "learning_rate": 8.865714285714287e-06,
      "loss": 0.1954,
      "step": 900
    },
    {
      "epoch": 0.23,
      "learning_rate": 8.794285714285716e-06,
      "loss": 0.1571,
      "step": 925
    },
    {
      "epoch": 0.24,
      "learning_rate": 8.722857142857145e-06,
      "loss": 0.1278,
      "step": 950
    },
    {
      "epoch": 0.24,
      "learning_rate": 8.651428571428572e-06,
      "loss": 0.1523,
      "step": 975
    },
    {
      "epoch": 0.25,
      "learning_rate": 8.580000000000001e-06,
      "loss": 0.1755,
      "step": 1000
    },
    {
      "epoch": 0.25,
      "eval_loss": 0.18439778685569763,
      "eval_runtime": 8848.8908,
      "eval_samples_per_second": 0.384,
      "eval_steps_per_second": 0.048,
      "eval_wer": 7.711814332140629,
      "step": 1000
    },
    {
      "epoch": 0.26,
      "learning_rate": 8.514285714285715e-06,
      "loss": 0.1432,
      "step": 1025
    },
    {
      "epoch": 0.26,
      "learning_rate": 8.442857142857144e-06,
      "loss": 0.1811,
      "step": 1050
    },
    {
      "epoch": 0.27,
      "learning_rate": 8.371428571428573e-06,
      "loss": 0.1765,
      "step": 1075
    },
    {
      "epoch": 0.28,
      "learning_rate": 8.3e-06,
      "loss": 0.0953,
      "step": 1100
    },
    {
      "epoch": 0.28,
      "learning_rate": 8.22857142857143e-06,
      "loss": 0.1404,
      "step": 1125
    },
    {
      "epoch": 0.29,
      "learning_rate": 8.157142857142858e-06,
      "loss": 0.1166,
      "step": 1150
    },
    {
      "epoch": 0.29,
      "learning_rate": 8.085714285714287e-06,
      "loss": 0.3181,
      "step": 1175
    },
    {
      "epoch": 0.3,
      "learning_rate": 8.014285714285715e-06,
      "loss": 0.1414,
      "step": 1200
    },
    {
      "epoch": 0.31,
      "learning_rate": 7.942857142857144e-06,
      "loss": 0.1416,
      "step": 1225
    },
    {
      "epoch": 0.31,
      "learning_rate": 7.871428571428573e-06,
      "loss": 0.1031,
      "step": 1250
    },
    {
      "epoch": 0.32,
      "learning_rate": 7.800000000000002e-06,
      "loss": 0.1126,
      "step": 1275
    },
    {
      "epoch": 0.33,
      "learning_rate": 7.72857142857143e-06,
      "loss": 0.1277,
      "step": 1300
    },
    {
      "epoch": 0.33,
      "learning_rate": 7.657142857142858e-06,
      "loss": 0.1504,
      "step": 1325
    },
    {
      "epoch": 0.34,
      "learning_rate": 7.585714285714286e-06,
      "loss": 0.1097,
      "step": 1350
    },
    {
      "epoch": 0.34,
      "learning_rate": 7.514285714285715e-06,
      "loss": 0.0795,
      "step": 1375
    },
    {
      "epoch": 0.35,
      "learning_rate": 7.442857142857144e-06,
      "loss": 0.109,
      "step": 1400
    },
    {
      "epoch": 0.36,
      "learning_rate": 7.371428571428571e-06,
      "loss": 0.2379,
      "step": 1425
    },
    {
      "epoch": 0.36,
      "learning_rate": 7.3e-06,
      "loss": 0.182,
      "step": 1450
    },
    {
      "epoch": 0.37,
      "learning_rate": 7.2285714285714294e-06,
      "loss": 0.3675,
      "step": 1475
    },
    {
      "epoch": 0.38,
      "learning_rate": 7.1571428571428584e-06,
      "loss": 0.2873,
      "step": 1500
    },
    {
      "epoch": 0.38,
      "learning_rate": 7.085714285714286e-06,
      "loss": 0.0965,
      "step": 1525
    },
    {
      "epoch": 0.39,
      "learning_rate": 7.014285714285715e-06,
      "loss": 0.1017,
      "step": 1550
    },
    {
      "epoch": 0.39,
      "learning_rate": 6.942857142857144e-06,
      "loss": 0.1058,
      "step": 1575
    },
    {
      "epoch": 0.4,
      "learning_rate": 6.871428571428572e-06,
      "loss": 0.0693,
      "step": 1600
    },
    {
      "epoch": 0.41,
      "learning_rate": 6.800000000000001e-06,
      "loss": 0.0821,
      "step": 1625
    },
    {
      "epoch": 0.41,
      "learning_rate": 6.728571428571429e-06,
      "loss": 0.0643,
      "step": 1650
    },
    {
      "epoch": 0.42,
      "learning_rate": 6.657142857142857e-06,
      "loss": 0.0846,
      "step": 1675
    },
    {
      "epoch": 0.42,
      "learning_rate": 6.585714285714286e-06,
      "loss": 0.1024,
      "step": 1700
    },
    {
      "epoch": 0.43,
      "learning_rate": 6.514285714285715e-06,
      "loss": 0.1094,
      "step": 1725
    },
    {
      "epoch": 0.44,
      "learning_rate": 6.442857142857143e-06,
      "loss": 0.1173,
      "step": 1750
    },
    {
      "epoch": 0.44,
      "learning_rate": 6.371428571428572e-06,
      "loss": 0.0973,
      "step": 1775
    },
    {
      "epoch": 0.45,
      "learning_rate": 6.300000000000001e-06,
      "loss": 0.0799,
      "step": 1800
    },
    {
      "epoch": 0.46,
      "learning_rate": 6.22857142857143e-06,
      "loss": 0.1528,
      "step": 1825
    },
    {
      "epoch": 0.46,
      "learning_rate": 6.157142857142858e-06,
      "loss": 0.1452,
      "step": 1850
    },
    {
      "epoch": 0.47,
      "learning_rate": 6.085714285714286e-06,
      "loss": 0.1264,
      "step": 1875
    },
    {
      "epoch": 0.47,
      "learning_rate": 6.014285714285715e-06,
      "loss": 0.1084,
      "step": 1900
    },
    {
      "epoch": 0.48,
      "learning_rate": 5.942857142857143e-06,
      "loss": 0.1454,
      "step": 1925
    },
    {
      "epoch": 0.49,
      "learning_rate": 5.871428571428572e-06,
      "loss": 0.0913,
      "step": 1950
    },
    {
      "epoch": 0.49,
      "learning_rate": 5.8e-06,
      "loss": 0.0881,
      "step": 1975
    },
    {
      "epoch": 0.5,
      "learning_rate": 5.7285714285714285e-06,
      "loss": 0.1185,
      "step": 2000
    },
    {
      "epoch": 0.5,
      "eval_loss": 0.16357915103435516,
      "eval_runtime": 8662.0186,
      "eval_samples_per_second": 0.392,
      "eval_steps_per_second": 0.049,
      "eval_wer": 7.065948855989233,
      "step": 2000
    },
    {
      "epoch": 0.51,
      "learning_rate": 5.6571428571428576e-06,
      "loss": 0.0804,
      "step": 2025
    },
    {
      "epoch": 0.51,
      "learning_rate": 5.5857142857142866e-06,
      "loss": 0.0799,
      "step": 2050
    },
    {
      "epoch": 0.52,
      "learning_rate": 5.514285714285714e-06,
      "loss": 0.2222,
      "step": 2075
    },
    {
      "epoch": 0.53,
      "learning_rate": 5.442857142857143e-06,
      "loss": 0.142,
      "step": 2100
    },
    {
      "epoch": 0.53,
      "learning_rate": 5.371428571428572e-06,
      "loss": 0.1169,
      "step": 2125
    },
    {
      "epoch": 0.54,
      "learning_rate": 5.300000000000001e-06,
      "loss": 0.0686,
      "step": 2150
    },
    {
      "epoch": 0.54,
      "learning_rate": 5.22857142857143e-06,
      "loss": 0.0817,
      "step": 2175
    },
    {
      "epoch": 0.55,
      "learning_rate": 5.157142857142857e-06,
      "loss": 0.0983,
      "step": 2200
    },
    {
      "epoch": 0.56,
      "learning_rate": 5.085714285714286e-06,
      "loss": 0.0901,
      "step": 2225
    },
    {
      "epoch": 0.56,
      "learning_rate": 5.014285714285715e-06,
      "loss": 0.0679,
      "step": 2250
    },
    {
      "epoch": 0.57,
      "learning_rate": 4.9428571428571435e-06,
      "loss": 0.109,
      "step": 2275
    },
    {
      "epoch": 0.57,
      "learning_rate": 4.871428571428572e-06,
      "loss": 0.1129,
      "step": 2300
    },
    {
      "epoch": 0.58,
      "learning_rate": 4.800000000000001e-06,
      "loss": 0.1413,
      "step": 2325
    },
    {
      "epoch": 0.59,
      "learning_rate": 4.728571428571429e-06,
      "loss": 0.0921,
      "step": 2350
    },
    {
      "epoch": 0.59,
      "learning_rate": 4.657142857142857e-06,
      "loss": 0.089,
      "step": 2375
    },
    {
      "epoch": 0.6,
      "learning_rate": 4.585714285714286e-06,
      "loss": 0.154,
      "step": 2400
    },
    {
      "epoch": 0.61,
      "learning_rate": 4.514285714285714e-06,
      "loss": 0.1316,
      "step": 2425
    },
    {
      "epoch": 0.61,
      "learning_rate": 4.442857142857143e-06,
      "loss": 0.1237,
      "step": 2450
    },
    {
      "epoch": 0.62,
      "learning_rate": 4.371428571428572e-06,
      "loss": 0.1243,
      "step": 2475
    },
    {
      "epoch": 0.62,
      "learning_rate": 4.3e-06,
      "loss": 0.1017,
      "step": 2500
    },
    {
      "epoch": 0.63,
      "learning_rate": 4.228571428571429e-06,
      "loss": 0.122,
      "step": 2525
    },
    {
      "epoch": 0.64,
      "learning_rate": 4.1571428571428575e-06,
      "loss": 0.1323,
      "step": 2550
    },
    {
      "epoch": 0.64,
      "learning_rate": 4.0857142857142865e-06,
      "loss": 0.157,
      "step": 2575
    },
    {
      "epoch": 0.65,
      "learning_rate": 4.014285714285715e-06,
      "loss": 0.1428,
      "step": 2600
    },
    {
      "epoch": 0.66,
      "learning_rate": 3.942857142857143e-06,
      "loss": 0.0626,
      "step": 2625
    },
    {
      "epoch": 0.66,
      "learning_rate": 3.871428571428572e-06,
      "loss": 0.1171,
      "step": 2650
    },
    {
      "epoch": 0.67,
      "learning_rate": 3.8000000000000005e-06,
      "loss": 0.0734,
      "step": 2675
    },
    {
      "epoch": 0.68,
      "learning_rate": 3.7285714285714286e-06,
      "loss": 0.0516,
      "step": 2700
    },
    {
      "epoch": 0.68,
      "learning_rate": 3.6571428571428576e-06,
      "loss": 0.0929,
      "step": 2725
    },
    {
      "epoch": 0.69,
      "learning_rate": 3.5857142857142862e-06,
      "loss": 0.1353,
      "step": 2750
    },
    {
      "epoch": 0.69,
      "learning_rate": 3.5142857142857144e-06,
      "loss": 0.0814,
      "step": 2775
    },
    {
      "epoch": 0.7,
      "learning_rate": 3.4428571428571434e-06,
      "loss": 0.1041,
      "step": 2800
    },
    {
      "epoch": 0.71,
      "learning_rate": 3.3714285714285716e-06,
      "loss": 0.0983,
      "step": 2825
    },
    {
      "epoch": 0.71,
      "learning_rate": 3.3000000000000006e-06,
      "loss": 0.0998,
      "step": 2850
    },
    {
      "epoch": 0.72,
      "learning_rate": 3.2285714285714288e-06,
      "loss": 0.1144,
      "step": 2875
    },
    {
      "epoch": 0.72,
      "learning_rate": 3.1571428571428573e-06,
      "loss": 0.1186,
      "step": 2900
    },
    {
      "epoch": 0.73,
      "learning_rate": 3.085714285714286e-06,
      "loss": 0.1283,
      "step": 2925
    },
    {
      "epoch": 0.74,
      "learning_rate": 3.0142857142857145e-06,
      "loss": 0.0761,
      "step": 2950
    },
    {
      "epoch": 0.74,
      "learning_rate": 2.9428571428571427e-06,
      "loss": 0.0605,
      "step": 2975
    },
    {
      "epoch": 0.75,
      "learning_rate": 2.8714285714285717e-06,
      "loss": 0.1081,
      "step": 3000
    },
    {
      "epoch": 0.75,
      "eval_loss": 0.1396336704492569,
      "eval_runtime": 8597.5423,
      "eval_samples_per_second": 0.395,
      "eval_steps_per_second": 0.049,
      "eval_wer": 6.084430292485966,
      "step": 3000
    },
    {
      "epoch": 0.76,
      "learning_rate": 2.8000000000000003e-06,
      "loss": 0.1712,
      "step": 3025
    },
    {
      "epoch": 0.76,
      "learning_rate": 2.728571428571429e-06,
      "loss": 0.1508,
      "step": 3050
    },
    {
      "epoch": 0.77,
      "learning_rate": 2.6571428571428575e-06,
      "loss": 0.1016,
      "step": 3075
    },
    {
      "epoch": 0.78,
      "learning_rate": 2.5857142857142856e-06,
      "loss": 0.0623,
      "step": 3100
    },
    {
      "epoch": 0.78,
      "learning_rate": 2.5142857142857147e-06,
      "loss": 0.0629,
      "step": 3125
    },
    {
      "epoch": 0.79,
      "learning_rate": 2.442857142857143e-06,
      "loss": 0.1124,
      "step": 3150
    },
    {
      "epoch": 0.79,
      "learning_rate": 2.371428571428572e-06,
      "loss": 0.1175,
      "step": 3175
    },
    {
      "epoch": 0.8,
      "learning_rate": 2.3000000000000004e-06,
      "loss": 0.1584,
      "step": 3200
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.228571428571429e-06,
      "loss": 0.2534,
      "step": 3225
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.157142857142857e-06,
      "loss": 0.1813,
      "step": 3250
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.0857142857142858e-06,
      "loss": 0.0968,
      "step": 3275
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.0142857142857144e-06,
      "loss": 0.0906,
      "step": 3300
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.942857142857143e-06,
      "loss": 0.1255,
      "step": 3325
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.8714285714285715e-06,
      "loss": 0.0677,
      "step": 3350
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.8000000000000001e-06,
      "loss": 0.0705,
      "step": 3375
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.7285714285714287e-06,
      "loss": 0.0884,
      "step": 3400
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.657142857142857e-06,
      "loss": 0.0658,
      "step": 3425
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.5857142857142857e-06,
      "loss": 0.0749,
      "step": 3450
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.5142857142857145e-06,
      "loss": 0.0737,
      "step": 3475
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.442857142857143e-06,
      "loss": 0.2422,
      "step": 3500
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.3714285714285717e-06,
      "loss": 0.1487,
      "step": 3525
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.3e-06,
      "loss": 0.0917,
      "step": 3550
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.2285714285714286e-06,
      "loss": 0.0937,
      "step": 3575
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.1571428571428572e-06,
      "loss": 0.1092,
      "step": 3600
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.0857142857142858e-06,
      "loss": 0.0945,
      "step": 3625
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.0142857142857144e-06,
      "loss": 0.08,
      "step": 3650
    },
    {
      "epoch": 0.92,
      "learning_rate": 9.42857142857143e-07,
      "loss": 0.0898,
      "step": 3675
    },
    {
      "epoch": 0.93,
      "learning_rate": 8.714285714285716e-07,
      "loss": 0.0821,
      "step": 3700
    },
    {
      "epoch": 0.93,
      "learning_rate": 8.000000000000001e-07,
      "loss": 0.0762,
      "step": 3725
    },
    {
      "epoch": 0.94,
      "learning_rate": 7.285714285714287e-07,
      "loss": 0.0619,
      "step": 3750
    },
    {
      "epoch": 0.94,
      "learning_rate": 6.571428571428571e-07,
      "loss": 0.0972,
      "step": 3775
    },
    {
      "epoch": 0.95,
      "learning_rate": 5.857142857142857e-07,
      "loss": 0.0923,
      "step": 3800
    },
    {
      "epoch": 0.96,
      "learning_rate": 5.142857142857143e-07,
      "loss": 0.1006,
      "step": 3825
    },
    {
      "epoch": 0.96,
      "learning_rate": 4.4285714285714286e-07,
      "loss": 0.1109,
      "step": 3850
    },
    {
      "epoch": 0.97,
      "learning_rate": 3.7142857142857145e-07,
      "loss": 0.0986,
      "step": 3875
    },
    {
      "epoch": 0.97,
      "learning_rate": 3.0000000000000004e-07,
      "loss": 0.0665,
      "step": 3900
    },
    {
      "epoch": 0.98,
      "learning_rate": 2.285714285714286e-07,
      "loss": 0.0962,
      "step": 3925
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.5714285714285717e-07,
      "loss": 0.1476,
      "step": 3950
    },
    {
      "epoch": 0.99,
      "learning_rate": 8.571428571428573e-08,
      "loss": 0.2288,
      "step": 3975
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.4285714285714288e-08,
      "loss": 0.1222,
      "step": 4000
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.13695605099201202,
      "eval_runtime": 8582.9213,
      "eval_samples_per_second": 0.395,
      "eval_steps_per_second": 0.05,
      "eval_wer": 6.048320913895545,
      "step": 4000
    },
    {
      "epoch": 1.0,
      "step": 4000,
      "total_flos": 5.435317813248e+20,
      "train_loss": 0.1346710950732231,
      "train_runtime": 109366.7192,
      "train_samples_per_second": 2.341,
      "train_steps_per_second": 0.037
    }
  ],
  "max_steps": 4000,
  "num_train_epochs": 9223372036854775807,
  "total_flos": 5.435317813248e+20,
  "trial_name": null,
  "trial_params": null
}