{
  "best_metric": 17.94694428111922,
  "best_model_checkpoint": "./checkpoint-5000",
  "epoch": 58.13953488372093,
  "eval_steps": 1000,
  "global_step": 5000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.29,
      "learning_rate": 1.0500000000000001e-06,
      "loss": 10.5572,
      "step": 25
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.3e-06,
      "loss": 8.9397,
      "step": 50
    },
    {
      "epoch": 0.87,
      "learning_rate": 3.55e-06,
      "loss": 7.3274,
      "step": 75
    },
    {
      "epoch": 1.16,
      "learning_rate": 4.800000000000001e-06,
      "loss": 6.2977,
      "step": 100
    },
    {
      "epoch": 1.45,
      "learning_rate": 6.0500000000000005e-06,
      "loss": 5.6656,
      "step": 125
    },
    {
      "epoch": 1.74,
      "learning_rate": 7.2999999999999996e-06,
      "loss": 5.3192,
      "step": 150
    },
    {
      "epoch": 2.03,
      "learning_rate": 8.550000000000001e-06,
      "loss": 5.0393,
      "step": 175
    },
    {
      "epoch": 2.33,
      "learning_rate": 9.800000000000001e-06,
      "loss": 4.7781,
      "step": 200
    },
    {
      "epoch": 2.62,
      "learning_rate": 1.1050000000000001e-05,
      "loss": 4.0512,
      "step": 225
    },
    {
      "epoch": 2.91,
      "learning_rate": 1.23e-05,
      "loss": 3.0152,
      "step": 250
    },
    {
      "epoch": 3.2,
      "learning_rate": 1.3550000000000002e-05,
      "loss": 2.2934,
      "step": 275
    },
    {
      "epoch": 3.49,
      "learning_rate": 1.48e-05,
      "loss": 1.8781,
      "step": 300
    },
    {
      "epoch": 3.78,
      "learning_rate": 1.605e-05,
      "loss": 1.6424,
      "step": 325
    },
    {
      "epoch": 4.07,
      "learning_rate": 1.73e-05,
      "loss": 1.4523,
      "step": 350
    },
    {
      "epoch": 4.36,
      "learning_rate": 1.855e-05,
      "loss": 1.2698,
      "step": 375
    },
    {
      "epoch": 4.65,
      "learning_rate": 1.9800000000000004e-05,
      "loss": 1.1583,
      "step": 400
    },
    {
      "epoch": 4.94,
      "learning_rate": 2.105e-05,
      "loss": 1.0625,
      "step": 425
    },
    {
      "epoch": 5.23,
      "learning_rate": 2.23e-05,
      "loss": 0.9448,
      "step": 450
    },
    {
      "epoch": 5.52,
      "learning_rate": 2.355e-05,
      "loss": 0.883,
      "step": 475
    },
    {
      "epoch": 5.81,
      "learning_rate": 2.48e-05,
      "loss": 0.8449,
      "step": 500
    },
    {
      "epoch": 6.1,
      "learning_rate": 2.4883333333333333e-05,
      "loss": 0.7883,
      "step": 525
    },
    {
      "epoch": 6.4,
      "learning_rate": 2.4744444444444445e-05,
      "loss": 0.6962,
      "step": 550
    },
    {
      "epoch": 6.69,
      "learning_rate": 2.4605555555555558e-05,
      "loss": 0.6835,
      "step": 575
    },
    {
      "epoch": 6.98,
      "learning_rate": 2.4466666666666667e-05,
      "loss": 0.6745,
      "step": 600
    },
    {
      "epoch": 7.27,
      "learning_rate": 2.432777777777778e-05,
      "loss": 0.5938,
      "step": 625
    },
    {
      "epoch": 7.56,
      "learning_rate": 2.418888888888889e-05,
      "loss": 0.573,
      "step": 650
    },
    {
      "epoch": 7.85,
      "learning_rate": 2.4050000000000002e-05,
      "loss": 0.5678,
      "step": 675
    },
    {
      "epoch": 8.14,
      "learning_rate": 2.391111111111111e-05,
      "loss": 0.5257,
      "step": 700
    },
    {
      "epoch": 8.43,
      "learning_rate": 2.3772222222222224e-05,
      "loss": 0.4938,
      "step": 725
    },
    {
      "epoch": 8.72,
      "learning_rate": 2.3633333333333336e-05,
      "loss": 0.4766,
      "step": 750
    },
    {
      "epoch": 9.01,
      "learning_rate": 2.3494444444444446e-05,
      "loss": 0.4733,
      "step": 775
    },
    {
      "epoch": 9.3,
      "learning_rate": 2.3355555555555555e-05,
      "loss": 0.4272,
      "step": 800
    },
    {
      "epoch": 9.59,
      "learning_rate": 2.3216666666666667e-05,
      "loss": 0.4259,
      "step": 825
    },
    {
      "epoch": 9.88,
      "learning_rate": 2.307777777777778e-05,
      "loss": 0.414,
      "step": 850
    },
    {
      "epoch": 10.17,
      "learning_rate": 2.293888888888889e-05,
      "loss": 0.3929,
      "step": 875
    },
    {
      "epoch": 10.47,
      "learning_rate": 2.2800000000000002e-05,
      "loss": 0.3739,
      "step": 900
    },
    {
      "epoch": 10.76,
      "learning_rate": 2.2661111111111115e-05,
      "loss": 0.3643,
      "step": 925
    },
    {
      "epoch": 11.05,
      "learning_rate": 2.2522222222222224e-05,
      "loss": 0.3698,
      "step": 950
    },
    {
      "epoch": 11.34,
      "learning_rate": 2.2383333333333333e-05,
      "loss": 0.3196,
      "step": 975
    },
    {
      "epoch": 11.63,
      "learning_rate": 2.2244444444444446e-05,
      "loss": 0.3275,
      "step": 1000
    },
    {
      "epoch": 11.63,
      "eval_loss": 0.4105360507965088,
      "eval_runtime": 74.8642,
      "eval_samples_per_second": 85.448,
      "eval_steps_per_second": 1.336,
      "eval_wer": 20.540933144260368,
      "step": 1000
    },
    {
      "epoch": 11.92,
      "learning_rate": 2.2105555555555558e-05,
      "loss": 0.3297,
      "step": 1025
    },
    {
      "epoch": 12.21,
      "learning_rate": 2.1966666666666668e-05,
      "loss": 0.2975,
      "step": 1050
    },
    {
      "epoch": 12.5,
      "learning_rate": 2.1827777777777777e-05,
      "loss": 0.2921,
      "step": 1075
    },
    {
      "epoch": 12.79,
      "learning_rate": 2.168888888888889e-05,
      "loss": 0.2965,
      "step": 1100
    },
    {
      "epoch": 13.08,
      "learning_rate": 2.1550000000000002e-05,
      "loss": 0.2763,
      "step": 1125
    },
    {
      "epoch": 13.37,
      "learning_rate": 2.141111111111111e-05,
      "loss": 0.257,
      "step": 1150
    },
    {
      "epoch": 13.66,
      "learning_rate": 2.1272222222222224e-05,
      "loss": 0.26,
      "step": 1175
    },
    {
      "epoch": 13.95,
      "learning_rate": 2.1133333333333337e-05,
      "loss": 0.2629,
      "step": 1200
    },
    {
      "epoch": 14.24,
      "learning_rate": 2.0994444444444446e-05,
      "loss": 0.2406,
      "step": 1225
    },
    {
      "epoch": 14.53,
      "learning_rate": 2.0855555555555555e-05,
      "loss": 0.2324,
      "step": 1250
    },
    {
      "epoch": 14.83,
      "learning_rate": 2.0716666666666668e-05,
      "loss": 0.2347,
      "step": 1275
    },
    {
      "epoch": 15.12,
      "learning_rate": 2.057777777777778e-05,
      "loss": 0.2232,
      "step": 1300
    },
    {
      "epoch": 15.41,
      "learning_rate": 2.043888888888889e-05,
      "loss": 0.2101,
      "step": 1325
    },
    {
      "epoch": 15.7,
      "learning_rate": 2.0300000000000002e-05,
      "loss": 0.2138,
      "step": 1350
    },
    {
      "epoch": 15.99,
      "learning_rate": 2.016111111111111e-05,
      "loss": 0.2106,
      "step": 1375
    },
    {
      "epoch": 16.28,
      "learning_rate": 2.0022222222222224e-05,
      "loss": 0.1863,
      "step": 1400
    },
    {
      "epoch": 16.57,
      "learning_rate": 1.9883333333333333e-05,
      "loss": 0.1975,
      "step": 1425
    },
    {
      "epoch": 16.86,
      "learning_rate": 1.9744444444444446e-05,
      "loss": 0.1903,
      "step": 1450
    },
    {
      "epoch": 17.15,
      "learning_rate": 1.960555555555556e-05,
      "loss": 0.184,
      "step": 1475
    },
    {
      "epoch": 17.44,
      "learning_rate": 1.9466666666666668e-05,
      "loss": 0.1708,
      "step": 1500
    },
    {
      "epoch": 17.73,
      "learning_rate": 1.9327777777777777e-05,
      "loss": 0.1712,
      "step": 1525
    },
    {
      "epoch": 18.02,
      "learning_rate": 1.918888888888889e-05,
      "loss": 0.1784,
      "step": 1550
    },
    {
      "epoch": 18.31,
      "learning_rate": 1.9050000000000002e-05,
      "loss": 0.1563,
      "step": 1575
    },
    {
      "epoch": 18.6,
      "learning_rate": 1.891111111111111e-05,
      "loss": 0.1609,
      "step": 1600
    },
    {
      "epoch": 18.9,
      "learning_rate": 1.8772222222222224e-05,
      "loss": 0.1592,
      "step": 1625
    },
    {
      "epoch": 19.19,
      "learning_rate": 1.8633333333333333e-05,
      "loss": 0.1485,
      "step": 1650
    },
    {
      "epoch": 19.48,
      "learning_rate": 1.8494444444444446e-05,
      "loss": 0.1476,
      "step": 1675
    },
    {
      "epoch": 19.77,
      "learning_rate": 1.8355555555555555e-05,
      "loss": 0.1446,
      "step": 1700
    },
    {
      "epoch": 20.06,
      "learning_rate": 1.8216666666666668e-05,
      "loss": 0.1447,
      "step": 1725
    },
    {
      "epoch": 20.35,
      "learning_rate": 1.807777777777778e-05,
      "loss": 0.1322,
      "step": 1750
    },
    {
      "epoch": 20.64,
      "learning_rate": 1.793888888888889e-05,
      "loss": 0.1326,
      "step": 1775
    },
    {
      "epoch": 20.93,
      "learning_rate": 1.78e-05,
      "loss": 0.1327,
      "step": 1800
    },
    {
      "epoch": 21.22,
      "learning_rate": 1.766111111111111e-05,
      "loss": 0.12,
      "step": 1825
    },
    {
      "epoch": 21.51,
      "learning_rate": 1.7522222222222224e-05,
      "loss": 0.1198,
      "step": 1850
    },
    {
      "epoch": 21.8,
      "learning_rate": 1.7383333333333333e-05,
      "loss": 0.1214,
      "step": 1875
    },
    {
      "epoch": 22.09,
      "learning_rate": 1.7244444444444446e-05,
      "loss": 0.1173,
      "step": 1900
    },
    {
      "epoch": 22.38,
      "learning_rate": 1.7105555555555555e-05,
      "loss": 0.1109,
      "step": 1925
    },
    {
      "epoch": 22.67,
      "learning_rate": 1.6966666666666668e-05,
      "loss": 0.1117,
      "step": 1950
    },
    {
      "epoch": 22.97,
      "learning_rate": 1.6827777777777777e-05,
      "loss": 0.1123,
      "step": 1975
    },
    {
      "epoch": 23.26,
      "learning_rate": 1.668888888888889e-05,
      "loss": 0.1016,
      "step": 2000
    },
    {
      "epoch": 23.26,
      "eval_loss": 0.4036574959754944,
      "eval_runtime": 73.4614,
      "eval_samples_per_second": 87.08,
      "eval_steps_per_second": 1.361,
      "eval_wer": 18.386193061944454,
      "step": 2000
    },
    {
      "epoch": 23.55,
      "learning_rate": 1.6550000000000002e-05,
      "loss": 0.1007,
      "step": 2025
    },
    {
      "epoch": 23.84,
      "learning_rate": 1.6411111111111112e-05,
      "loss": 0.1051,
      "step": 2050
    },
    {
      "epoch": 24.13,
      "learning_rate": 1.627222222222222e-05,
      "loss": 0.1012,
      "step": 2075
    },
    {
      "epoch": 24.42,
      "learning_rate": 1.6133333333333334e-05,
      "loss": 0.093,
      "step": 2100
    },
    {
      "epoch": 24.71,
      "learning_rate": 1.5994444444444446e-05,
      "loss": 0.0956,
      "step": 2125
    },
    {
      "epoch": 25.0,
      "learning_rate": 1.5855555555555555e-05,
      "loss": 0.0971,
      "step": 2150
    },
    {
      "epoch": 25.29,
      "learning_rate": 1.5716666666666668e-05,
      "loss": 0.0837,
      "step": 2175
    },
    {
      "epoch": 25.58,
      "learning_rate": 1.5577777777777777e-05,
      "loss": 0.0884,
      "step": 2200
    },
    {
      "epoch": 25.87,
      "learning_rate": 1.543888888888889e-05,
      "loss": 0.0871,
      "step": 2225
    },
    {
      "epoch": 26.16,
      "learning_rate": 1.53e-05,
      "loss": 0.0826,
      "step": 2250
    },
    {
      "epoch": 26.45,
      "learning_rate": 1.5161111111111112e-05,
      "loss": 0.0807,
      "step": 2275
    },
    {
      "epoch": 26.74,
      "learning_rate": 1.5022222222222224e-05,
      "loss": 0.0808,
      "step": 2300
    },
    {
      "epoch": 27.03,
      "learning_rate": 1.4883333333333335e-05,
      "loss": 0.0807,
      "step": 2325
    },
    {
      "epoch": 27.33,
      "learning_rate": 1.4744444444444445e-05,
      "loss": 0.0737,
      "step": 2350
    },
    {
      "epoch": 27.62,
      "learning_rate": 1.4605555555555556e-05,
      "loss": 0.0739,
      "step": 2375
    },
    {
      "epoch": 27.91,
      "learning_rate": 1.4466666666666667e-05,
      "loss": 0.0748,
      "step": 2400
    },
    {
      "epoch": 28.2,
      "learning_rate": 1.4327777777777779e-05,
      "loss": 0.0691,
      "step": 2425
    },
    {
      "epoch": 28.49,
      "learning_rate": 1.418888888888889e-05,
      "loss": 0.0711,
      "step": 2450
    },
    {
      "epoch": 28.78,
      "learning_rate": 1.4050000000000003e-05,
      "loss": 0.0682,
      "step": 2475
    },
    {
      "epoch": 29.07,
      "learning_rate": 1.391111111111111e-05,
      "loss": 0.067,
      "step": 2500
    },
    {
      "epoch": 29.36,
      "learning_rate": 1.3772222222222223e-05,
      "loss": 0.0655,
      "step": 2525
    },
    {
      "epoch": 29.65,
      "learning_rate": 1.3633333333333334e-05,
      "loss": 0.0635,
      "step": 2550
    },
    {
      "epoch": 29.94,
      "learning_rate": 1.3494444444444446e-05,
      "loss": 0.066,
      "step": 2575
    },
    {
      "epoch": 30.23,
      "learning_rate": 1.3355555555555557e-05,
      "loss": 0.0587,
      "step": 2600
    },
    {
      "epoch": 30.52,
      "learning_rate": 1.3216666666666667e-05,
      "loss": 0.0605,
      "step": 2625
    },
    {
      "epoch": 30.81,
      "learning_rate": 1.3077777777777778e-05,
      "loss": 0.0606,
      "step": 2650
    },
    {
      "epoch": 31.1,
      "learning_rate": 1.2938888888888888e-05,
      "loss": 0.0581,
      "step": 2675
    },
    {
      "epoch": 31.4,
      "learning_rate": 1.2800000000000001e-05,
      "loss": 0.0544,
      "step": 2700
    },
    {
      "epoch": 31.69,
      "learning_rate": 1.2661111111111112e-05,
      "loss": 0.0568,
      "step": 2725
    },
    {
      "epoch": 31.98,
      "learning_rate": 1.2522222222222225e-05,
      "loss": 0.0567,
      "step": 2750
    },
    {
      "epoch": 32.27,
      "learning_rate": 1.2383333333333334e-05,
      "loss": 0.05,
      "step": 2775
    },
    {
      "epoch": 32.56,
      "learning_rate": 1.2244444444444445e-05,
      "loss": 0.0525,
      "step": 2800
    },
    {
      "epoch": 32.85,
      "learning_rate": 1.2105555555555556e-05,
      "loss": 0.0541,
      "step": 2825
    },
    {
      "epoch": 33.14,
      "learning_rate": 1.1966666666666668e-05,
      "loss": 0.0508,
      "step": 2850
    },
    {
      "epoch": 33.43,
      "learning_rate": 1.1827777777777778e-05,
      "loss": 0.0474,
      "step": 2875
    },
    {
      "epoch": 33.72,
      "learning_rate": 1.168888888888889e-05,
      "loss": 0.05,
      "step": 2900
    },
    {
      "epoch": 34.01,
      "learning_rate": 1.1550000000000001e-05,
      "loss": 0.0491,
      "step": 2925
    },
    {
      "epoch": 34.3,
      "learning_rate": 1.141111111111111e-05,
      "loss": 0.045,
      "step": 2950
    },
    {
      "epoch": 34.59,
      "learning_rate": 1.1272222222222223e-05,
      "loss": 0.0476,
      "step": 2975
    },
    {
      "epoch": 34.88,
      "learning_rate": 1.1133333333333334e-05,
      "loss": 0.0444,
      "step": 3000
    },
    {
      "epoch": 34.88,
      "eval_loss": 0.42896273732185364,
      "eval_runtime": 73.302,
      "eval_samples_per_second": 87.269,
      "eval_steps_per_second": 1.364,
      "eval_wer": 18.28589215923633,
      "step": 3000
    },
    {
      "epoch": 35.17,
      "learning_rate": 1.0994444444444445e-05,
      "loss": 0.0444,
      "step": 3025
    },
    {
      "epoch": 35.47,
      "learning_rate": 1.0855555555555556e-05,
      "loss": 0.0436,
      "step": 3050
    },
    {
      "epoch": 35.76,
      "learning_rate": 1.0716666666666667e-05,
      "loss": 0.0436,
      "step": 3075
    },
    {
      "epoch": 36.05,
      "learning_rate": 1.0577777777777778e-05,
      "loss": 0.0442,
      "step": 3100
    },
    {
      "epoch": 36.34,
      "learning_rate": 1.043888888888889e-05,
      "loss": 0.0396,
      "step": 3125
    },
    {
      "epoch": 36.63,
      "learning_rate": 1.03e-05,
      "loss": 0.0407,
      "step": 3150
    },
    {
      "epoch": 36.92,
      "learning_rate": 1.0161111111111112e-05,
      "loss": 0.04,
      "step": 3175
    },
    {
      "epoch": 37.21,
      "learning_rate": 1.0022222222222223e-05,
      "loss": 0.0391,
      "step": 3200
    },
    {
      "epoch": 37.5,
      "learning_rate": 9.883333333333334e-06,
      "loss": 0.0374,
      "step": 3225
    },
    {
      "epoch": 37.79,
      "learning_rate": 9.744444444444445e-06,
      "loss": 0.0399,
      "step": 3250
    },
    {
      "epoch": 38.08,
      "learning_rate": 9.605555555555556e-06,
      "loss": 0.0374,
      "step": 3275
    },
    {
      "epoch": 38.37,
      "learning_rate": 9.466666666666667e-06,
      "loss": 0.0363,
      "step": 3300
    },
    {
      "epoch": 38.66,
      "learning_rate": 9.327777777777778e-06,
      "loss": 0.0374,
      "step": 3325
    },
    {
      "epoch": 38.95,
      "learning_rate": 9.18888888888889e-06,
      "loss": 0.038,
      "step": 3350
    },
    {
      "epoch": 39.24,
      "learning_rate": 9.05e-06,
      "loss": 0.0343,
      "step": 3375
    },
    {
      "epoch": 39.53,
      "learning_rate": 8.911111111111112e-06,
      "loss": 0.0355,
      "step": 3400
    },
    {
      "epoch": 39.83,
      "learning_rate": 8.772222222222222e-06,
      "loss": 0.0372,
      "step": 3425
    },
    {
      "epoch": 40.12,
      "learning_rate": 8.633333333333334e-06,
      "loss": 0.0349,
      "step": 3450
    },
    {
      "epoch": 40.41,
      "learning_rate": 8.494444444444445e-06,
      "loss": 0.0325,
      "step": 3475
    },
    {
      "epoch": 40.7,
      "learning_rate": 8.355555555555556e-06,
      "loss": 0.0334,
      "step": 3500
    },
    {
      "epoch": 40.99,
      "learning_rate": 8.216666666666667e-06,
      "loss": 0.0347,
      "step": 3525
    },
    {
      "epoch": 41.28,
      "learning_rate": 8.077777777777778e-06,
      "loss": 0.0313,
      "step": 3550
    },
    {
      "epoch": 41.57,
      "learning_rate": 7.938888888888889e-06,
      "loss": 0.0321,
      "step": 3575
    },
    {
      "epoch": 41.86,
      "learning_rate": 7.8e-06,
      "loss": 0.0309,
      "step": 3600
    },
    {
      "epoch": 42.15,
      "learning_rate": 7.661111111111112e-06,
      "loss": 0.0313,
      "step": 3625
    },
    {
      "epoch": 42.44,
      "learning_rate": 7.5222222222222226e-06,
      "loss": 0.0293,
      "step": 3650
    },
    {
      "epoch": 42.73,
      "learning_rate": 7.3833333333333335e-06,
      "loss": 0.0319,
      "step": 3675
    },
    {
      "epoch": 43.02,
      "learning_rate": 7.244444444444445e-06,
      "loss": 0.0306,
      "step": 3700
    },
    {
      "epoch": 43.31,
      "learning_rate": 7.105555555555555e-06,
      "loss": 0.03,
      "step": 3725
    },
    {
      "epoch": 43.6,
      "learning_rate": 6.966666666666667e-06,
      "loss": 0.0299,
      "step": 3750
    },
    {
      "epoch": 43.9,
      "learning_rate": 6.827777777777779e-06,
      "loss": 0.03,
      "step": 3775
    },
    {
      "epoch": 44.19,
      "learning_rate": 6.688888888888889e-06,
      "loss": 0.0287,
      "step": 3800
    },
    {
      "epoch": 44.48,
      "learning_rate": 6.550000000000001e-06,
      "loss": 0.0279,
      "step": 3825
    },
    {
      "epoch": 44.77,
      "learning_rate": 6.411111111111111e-06,
      "loss": 0.0281,
      "step": 3850
    },
    {
      "epoch": 45.06,
      "learning_rate": 6.272222222222223e-06,
      "loss": 0.0284,
      "step": 3875
    },
    {
      "epoch": 45.35,
      "learning_rate": 6.133333333333334e-06,
      "loss": 0.027,
      "step": 3900
    },
    {
      "epoch": 45.64,
      "learning_rate": 5.9944444444444446e-06,
      "loss": 0.0261,
      "step": 3925
    },
    {
      "epoch": 45.93,
      "learning_rate": 5.8555555555555555e-06,
      "loss": 0.0276,
      "step": 3950
    },
    {
      "epoch": 46.22,
      "learning_rate": 5.7166666666666664e-06,
      "loss": 0.0264,
      "step": 3975
    },
    {
      "epoch": 46.51,
      "learning_rate": 5.577777777777778e-06,
      "loss": 0.0265,
      "step": 4000
    },
    {
      "epoch": 46.51,
      "eval_loss": 0.4462805688381195,
      "eval_runtime": 73.8032,
      "eval_samples_per_second": 86.676,
      "eval_steps_per_second": 1.355,
      "eval_wer": 18.01438799156089,
      "step": 4000
    },
    {
      "epoch": 46.8,
      "learning_rate": 5.438888888888889e-06,
      "loss": 0.0268,
      "step": 4025
    },
    {
      "epoch": 47.09,
      "learning_rate": 5.3e-06,
      "loss": 0.0262,
      "step": 4050
    },
    {
      "epoch": 47.38,
      "learning_rate": 5.161111111111112e-06,
      "loss": 0.0252,
      "step": 4075
    },
    {
      "epoch": 47.67,
      "learning_rate": 5.022222222222223e-06,
      "loss": 0.026,
      "step": 4100
    },
    {
      "epoch": 47.97,
      "learning_rate": 4.883333333333334e-06,
      "loss": 0.025,
      "step": 4125
    },
    {
      "epoch": 48.26,
      "learning_rate": 4.744444444444445e-06,
      "loss": 0.0244,
      "step": 4150
    },
    {
      "epoch": 48.55,
      "learning_rate": 4.605555555555556e-06,
      "loss": 0.0239,
      "step": 4175
    },
    {
      "epoch": 48.84,
      "learning_rate": 4.4666666666666665e-06,
      "loss": 0.0245,
      "step": 4200
    },
    {
      "epoch": 49.13,
      "learning_rate": 4.3277777777777775e-06,
      "loss": 0.024,
      "step": 4225
    },
    {
      "epoch": 49.42,
      "learning_rate": 4.188888888888889e-06,
      "loss": 0.0241,
      "step": 4250
    },
    {
      "epoch": 49.71,
      "learning_rate": 4.05e-06,
      "loss": 0.0238,
      "step": 4275
    },
    {
      "epoch": 50.0,
      "learning_rate": 3.911111111111111e-06,
      "loss": 0.0246,
      "step": 4300
    },
    {
      "epoch": 50.29,
      "learning_rate": 3.772222222222222e-06,
      "loss": 0.0227,
      "step": 4325
    },
    {
      "epoch": 50.58,
      "learning_rate": 3.633333333333334e-06,
      "loss": 0.0236,
      "step": 4350
    },
    {
      "epoch": 50.87,
      "learning_rate": 3.4944444444444448e-06,
      "loss": 0.0229,
      "step": 4375
    },
    {
      "epoch": 51.16,
      "learning_rate": 3.3555555555555557e-06,
      "loss": 0.0234,
      "step": 4400
    },
    {
      "epoch": 51.45,
      "learning_rate": 3.216666666666667e-06,
      "loss": 0.0229,
      "step": 4425
    },
    {
      "epoch": 51.74,
      "learning_rate": 3.077777777777778e-06,
      "loss": 0.0226,
      "step": 4450
    },
    {
      "epoch": 52.03,
      "learning_rate": 2.938888888888889e-06,
      "loss": 0.0221,
      "step": 4475
    },
    {
      "epoch": 52.33,
      "learning_rate": 2.8000000000000003e-06,
      "loss": 0.0225,
      "step": 4500
    },
    {
      "epoch": 52.62,
      "learning_rate": 2.6611111111111112e-06,
      "loss": 0.0233,
      "step": 4525
    },
    {
      "epoch": 52.91,
      "learning_rate": 2.522222222222222e-06,
      "loss": 0.0214,
      "step": 4550
    },
    {
      "epoch": 53.2,
      "learning_rate": 2.3833333333333335e-06,
      "loss": 0.0221,
      "step": 4575
    },
    {
      "epoch": 53.49,
      "learning_rate": 2.2444444444444445e-06,
      "loss": 0.0227,
      "step": 4600
    },
    {
      "epoch": 53.78,
      "learning_rate": 2.105555555555556e-06,
      "loss": 0.0218,
      "step": 4625
    },
    {
      "epoch": 54.07,
      "learning_rate": 1.9666666666666668e-06,
      "loss": 0.0219,
      "step": 4650
    },
    {
      "epoch": 54.36,
      "learning_rate": 1.827777777777778e-06,
      "loss": 0.0214,
      "step": 4675
    },
    {
      "epoch": 54.65,
      "learning_rate": 1.6888888888888888e-06,
      "loss": 0.0223,
      "step": 4700
    },
    {
      "epoch": 54.94,
      "learning_rate": 1.55e-06,
      "loss": 0.0213,
      "step": 4725
    },
    {
      "epoch": 55.23,
      "learning_rate": 1.4111111111111111e-06,
      "loss": 0.0205,
      "step": 4750
    },
    {
      "epoch": 55.52,
      "learning_rate": 1.2722222222222223e-06,
      "loss": 0.0206,
      "step": 4775
    },
    {
      "epoch": 55.81,
      "learning_rate": 1.1333333333333334e-06,
      "loss": 0.021,
      "step": 4800
    },
    {
      "epoch": 56.1,
      "learning_rate": 9.944444444444446e-07,
      "loss": 0.0211,
      "step": 4825
    },
    {
      "epoch": 56.4,
      "learning_rate": 8.555555555555556e-07,
      "loss": 0.0213,
      "step": 4850
    },
    {
      "epoch": 56.69,
      "learning_rate": 7.166666666666667e-07,
      "loss": 0.021,
      "step": 4875
    },
    {
      "epoch": 56.98,
      "learning_rate": 5.777777777777778e-07,
      "loss": 0.0201,
      "step": 4900
    },
    {
      "epoch": 57.27,
      "learning_rate": 4.3888888888888895e-07,
      "loss": 0.0218,
      "step": 4925
    },
    {
      "epoch": 57.56,
      "learning_rate": 3.0000000000000004e-07,
      "loss": 0.0212,
      "step": 4950
    },
    {
      "epoch": 57.85,
      "learning_rate": 1.611111111111111e-07,
      "loss": 0.0202,
      "step": 4975
    },
    {
      "epoch": 58.14,
      "learning_rate": 2.2222222222222224e-08,
      "loss": 0.0213,
      "step": 5000
    },
    {
      "epoch": 58.14,
      "eval_loss": 0.44943663477897644,
      "eval_runtime": 73.6911,
      "eval_samples_per_second": 86.808,
      "eval_steps_per_second": 1.357,
      "eval_wer": 17.94694428111922,
      "step": 5000
    },
    {
      "epoch": 58.14,
      "step": 5000,
      "total_flos": 4.129597039509504e+19,
      "train_loss": 0.4845314011693001,
      "train_runtime": 4394.9084,
      "train_samples_per_second": 145.623,
      "train_steps_per_second": 1.138
    }
  ],
  "logging_steps": 25,
  "max_steps": 5000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 59,
  "save_steps": 1000,
  "total_flos": 4.129597039509504e+19,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": null
}