{
  "best_metric": 28.965267965695134,
  "best_model_checkpoint": "./checkpoint-5000",
  "epoch": 70.4225352112676,
  "eval_steps": 1000,
  "global_step": 5000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.35,
      "learning_rate": 4.4e-08,
      "loss": 1.4431,
      "step": 25
    },
    {
      "epoch": 0.7,
      "learning_rate": 9.4e-08,
      "loss": 1.4443,
      "step": 50
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.44e-07,
      "loss": 1.4288,
      "step": 75
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.94e-07,
      "loss": 1.3802,
      "step": 100
    },
    {
      "epoch": 1.76,
      "learning_rate": 2.4399999999999996e-07,
      "loss": 1.311,
      "step": 125
    },
    {
      "epoch": 2.11,
      "learning_rate": 2.9399999999999996e-07,
      "loss": 1.2618,
      "step": 150
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.4399999999999996e-07,
      "loss": 1.1683,
      "step": 175
    },
    {
      "epoch": 2.82,
      "learning_rate": 3.94e-07,
      "loss": 1.1132,
      "step": 200
    },
    {
      "epoch": 3.17,
      "learning_rate": 4.44e-07,
      "loss": 1.0068,
      "step": 225
    },
    {
      "epoch": 3.52,
      "learning_rate": 4.94e-07,
      "loss": 0.8763,
      "step": 250
    },
    {
      "epoch": 3.87,
      "learning_rate": 5.44e-07,
      "loss": 0.7522,
      "step": 275
    },
    {
      "epoch": 4.23,
      "learning_rate": 5.939999999999999e-07,
      "loss": 0.6957,
      "step": 300
    },
    {
      "epoch": 4.58,
      "learning_rate": 6.44e-07,
      "loss": 0.6593,
      "step": 325
    },
    {
      "epoch": 4.93,
      "learning_rate": 6.939999999999999e-07,
      "loss": 0.6523,
      "step": 350
    },
    {
      "epoch": 5.28,
      "learning_rate": 7.44e-07,
      "loss": 0.6342,
      "step": 375
    },
    {
      "epoch": 5.63,
      "learning_rate": 7.94e-07,
      "loss": 0.6159,
      "step": 400
    },
    {
      "epoch": 5.99,
      "learning_rate": 8.439999999999999e-07,
      "loss": 0.6051,
      "step": 425
    },
    {
      "epoch": 6.34,
      "learning_rate": 8.939999999999999e-07,
      "loss": 0.6003,
      "step": 450
    },
    {
      "epoch": 6.69,
      "learning_rate": 9.439999999999999e-07,
      "loss": 0.586,
      "step": 475
    },
    {
      "epoch": 7.04,
      "learning_rate": 9.94e-07,
      "loss": 0.5745,
      "step": 500
    },
    {
      "epoch": 7.39,
      "learning_rate": 9.95111111111111e-07,
      "loss": 0.5695,
      "step": 525
    },
    {
      "epoch": 7.75,
      "learning_rate": 9.895555555555554e-07,
      "loss": 0.5616,
      "step": 550
    },
    {
      "epoch": 8.1,
      "learning_rate": 9.84e-07,
      "loss": 0.549,
      "step": 575
    },
    {
      "epoch": 8.45,
      "learning_rate": 9.784444444444444e-07,
      "loss": 0.5459,
      "step": 600
    },
    {
      "epoch": 8.8,
      "learning_rate": 9.728888888888888e-07,
      "loss": 0.5385,
      "step": 625
    },
    {
      "epoch": 9.15,
      "learning_rate": 9.673333333333332e-07,
      "loss": 0.5365,
      "step": 650
    },
    {
      "epoch": 9.51,
      "learning_rate": 9.617777777777776e-07,
      "loss": 0.5282,
      "step": 675
    },
    {
      "epoch": 9.86,
      "learning_rate": 9.562222222222223e-07,
      "loss": 0.5247,
      "step": 700
    },
    {
      "epoch": 10.21,
      "learning_rate": 9.506666666666667e-07,
      "loss": 0.5142,
      "step": 725
    },
    {
      "epoch": 10.56,
      "learning_rate": 9.451111111111111e-07,
      "loss": 0.5033,
      "step": 750
    },
    {
      "epoch": 10.92,
      "learning_rate": 9.395555555555556e-07,
      "loss": 0.5205,
      "step": 775
    },
    {
      "epoch": 11.27,
      "learning_rate": 9.34e-07,
      "loss": 0.4991,
      "step": 800
    },
    {
      "epoch": 11.62,
      "learning_rate": 9.284444444444444e-07,
      "loss": 0.4977,
      "step": 825
    },
    {
      "epoch": 11.97,
      "learning_rate": 9.228888888888888e-07,
      "loss": 0.5025,
      "step": 850
    },
    {
      "epoch": 12.32,
      "learning_rate": 9.173333333333333e-07,
      "loss": 0.4888,
      "step": 875
    },
    {
      "epoch": 12.68,
      "learning_rate": 9.117777777777778e-07,
      "loss": 0.4866,
      "step": 900
    },
    {
      "epoch": 13.03,
      "learning_rate": 9.062222222222222e-07,
      "loss": 0.4818,
      "step": 925
    },
    {
      "epoch": 13.38,
      "learning_rate": 9.006666666666666e-07,
      "loss": 0.4753,
      "step": 950
    },
    {
      "epoch": 13.73,
      "learning_rate": 8.95111111111111e-07,
      "loss": 0.4743,
      "step": 975
    },
    {
      "epoch": 14.08,
      "learning_rate": 8.895555555555555e-07,
      "loss": 0.4763,
      "step": 1000
    },
    {
      "epoch": 14.08,
      "eval_loss": 0.5686116814613342,
      "eval_runtime": 236.519,
      "eval_samples_per_second": 38.356,
      "eval_steps_per_second": 0.3,
      "eval_wer": 31.31140538231525,
      "step": 1000
    },
    {
      "epoch": 14.44,
      "learning_rate": 8.839999999999999e-07,
      "loss": 0.4698,
      "step": 1025
    },
    {
      "epoch": 14.79,
      "learning_rate": 8.784444444444444e-07,
      "loss": 0.4673,
      "step": 1050
    },
    {
      "epoch": 15.14,
      "learning_rate": 8.728888888888889e-07,
      "loss": 0.4527,
      "step": 1075
    },
    {
      "epoch": 15.49,
      "learning_rate": 8.673333333333332e-07,
      "loss": 0.4653,
      "step": 1100
    },
    {
      "epoch": 15.85,
      "learning_rate": 8.617777777777777e-07,
      "loss": 0.457,
      "step": 1125
    },
    {
      "epoch": 16.2,
      "learning_rate": 8.562222222222222e-07,
      "loss": 0.4479,
      "step": 1150
    },
    {
      "epoch": 16.55,
      "learning_rate": 8.506666666666667e-07,
      "loss": 0.4474,
      "step": 1175
    },
    {
      "epoch": 16.9,
      "learning_rate": 8.451111111111111e-07,
      "loss": 0.4468,
      "step": 1200
    },
    {
      "epoch": 17.25,
      "learning_rate": 8.395555555555556e-07,
      "loss": 0.4468,
      "step": 1225
    },
    {
      "epoch": 17.61,
      "learning_rate": 8.34e-07,
      "loss": 0.438,
      "step": 1250
    },
    {
      "epoch": 17.96,
      "learning_rate": 8.284444444444444e-07,
      "loss": 0.4447,
      "step": 1275
    },
    {
      "epoch": 18.31,
      "learning_rate": 8.228888888888889e-07,
      "loss": 0.4327,
      "step": 1300
    },
    {
      "epoch": 18.66,
      "learning_rate": 8.173333333333333e-07,
      "loss": 0.4316,
      "step": 1325
    },
    {
      "epoch": 19.01,
      "learning_rate": 8.117777777777778e-07,
      "loss": 0.4351,
      "step": 1350
    },
    {
      "epoch": 19.37,
      "learning_rate": 8.062222222222221e-07,
      "loss": 0.4282,
      "step": 1375
    },
    {
      "epoch": 19.72,
      "learning_rate": 8.006666666666666e-07,
      "loss": 0.4253,
      "step": 1400
    },
    {
      "epoch": 20.07,
      "learning_rate": 7.95111111111111e-07,
      "loss": 0.4213,
      "step": 1425
    },
    {
      "epoch": 20.42,
      "learning_rate": 7.895555555555555e-07,
      "loss": 0.4166,
      "step": 1450
    },
    {
      "epoch": 20.77,
      "learning_rate": 7.84e-07,
      "loss": 0.4257,
      "step": 1475
    },
    {
      "epoch": 21.13,
      "learning_rate": 7.784444444444444e-07,
      "loss": 0.4163,
      "step": 1500
    },
    {
      "epoch": 21.48,
      "learning_rate": 7.728888888888888e-07,
      "loss": 0.4112,
      "step": 1525
    },
    {
      "epoch": 21.83,
      "learning_rate": 7.673333333333332e-07,
      "loss": 0.4171,
      "step": 1550
    },
    {
      "epoch": 22.18,
      "learning_rate": 7.617777777777778e-07,
      "loss": 0.4077,
      "step": 1575
    },
    {
      "epoch": 22.54,
      "learning_rate": 7.562222222222222e-07,
      "loss": 0.4052,
      "step": 1600
    },
    {
      "epoch": 22.89,
      "learning_rate": 7.506666666666667e-07,
      "loss": 0.4077,
      "step": 1625
    },
    {
      "epoch": 23.24,
      "learning_rate": 7.451111111111111e-07,
      "loss": 0.402,
      "step": 1650
    },
    {
      "epoch": 23.59,
      "learning_rate": 7.395555555555555e-07,
      "loss": 0.4024,
      "step": 1675
    },
    {
      "epoch": 23.94,
      "learning_rate": 7.34e-07,
      "loss": 0.4018,
      "step": 1700
    },
    {
      "epoch": 24.3,
      "learning_rate": 7.284444444444444e-07,
      "loss": 0.4014,
      "step": 1725
    },
    {
      "epoch": 24.65,
      "learning_rate": 7.228888888888889e-07,
      "loss": 0.3975,
      "step": 1750
    },
    {
      "epoch": 25.0,
      "learning_rate": 7.173333333333333e-07,
      "loss": 0.3902,
      "step": 1775
    },
    {
      "epoch": 25.35,
      "learning_rate": 7.117777777777777e-07,
      "loss": 0.3928,
      "step": 1800
    },
    {
      "epoch": 25.7,
      "learning_rate": 7.062222222222222e-07,
      "loss": 0.388,
      "step": 1825
    },
    {
      "epoch": 26.06,
      "learning_rate": 7.006666666666666e-07,
      "loss": 0.387,
      "step": 1850
    },
    {
      "epoch": 26.41,
      "learning_rate": 6.951111111111111e-07,
      "loss": 0.389,
      "step": 1875
    },
    {
      "epoch": 26.76,
      "learning_rate": 6.895555555555555e-07,
      "loss": 0.3828,
      "step": 1900
    },
    {
      "epoch": 27.11,
      "learning_rate": 6.84e-07,
      "loss": 0.3846,
      "step": 1925
    },
    {
      "epoch": 27.46,
      "learning_rate": 6.784444444444443e-07,
      "loss": 0.3796,
      "step": 1950
    },
    {
      "epoch": 27.82,
      "learning_rate": 6.728888888888888e-07,
      "loss": 0.3855,
      "step": 1975
    },
    {
      "epoch": 28.17,
      "learning_rate": 6.673333333333334e-07,
      "loss": 0.3784,
      "step": 2000
    },
    {
      "epoch": 28.17,
      "eval_loss": 0.5349831581115723,
      "eval_runtime": 76.3989,
      "eval_samples_per_second": 118.745,
      "eval_steps_per_second": 0.929,
      "eval_wer": 30.069332632339897,
      "step": 2000
    },
    {
      "epoch": 28.52,
      "learning_rate": 6.617777777777778e-07,
      "loss": 0.3766,
      "step": 2025
    },
    {
      "epoch": 28.87,
      "learning_rate": 6.562222222222223e-07,
      "loss": 0.3725,
      "step": 2050
    },
    {
      "epoch": 29.23,
      "learning_rate": 6.506666666666666e-07,
      "loss": 0.3804,
      "step": 2075
    },
    {
      "epoch": 29.58,
      "learning_rate": 6.451111111111111e-07,
      "loss": 0.3759,
      "step": 2100
    },
    {
      "epoch": 29.93,
      "learning_rate": 6.395555555555555e-07,
      "loss": 0.3679,
      "step": 2125
    },
    {
      "epoch": 30.28,
      "learning_rate": 6.34e-07,
      "loss": 0.3644,
      "step": 2150
    },
    {
      "epoch": 30.63,
      "learning_rate": 6.284444444444445e-07,
      "loss": 0.3696,
      "step": 2175
    },
    {
      "epoch": 30.99,
      "learning_rate": 6.228888888888889e-07,
      "loss": 0.3699,
      "step": 2200
    },
    {
      "epoch": 31.34,
      "learning_rate": 6.173333333333333e-07,
      "loss": 0.3638,
      "step": 2225
    },
    {
      "epoch": 31.69,
      "learning_rate": 6.117777777777777e-07,
      "loss": 0.3619,
      "step": 2250
    },
    {
      "epoch": 32.04,
      "learning_rate": 6.062222222222222e-07,
      "loss": 0.3659,
      "step": 2275
    },
    {
      "epoch": 32.39,
      "learning_rate": 6.006666666666666e-07,
      "loss": 0.3584,
      "step": 2300
    },
    {
      "epoch": 32.75,
      "learning_rate": 5.951111111111111e-07,
      "loss": 0.362,
      "step": 2325
    },
    {
      "epoch": 33.1,
      "learning_rate": 5.895555555555555e-07,
      "loss": 0.3655,
      "step": 2350
    },
    {
      "epoch": 33.45,
      "learning_rate": 5.839999999999999e-07,
      "loss": 0.3552,
      "step": 2375
    },
    {
      "epoch": 33.8,
      "learning_rate": 5.784444444444444e-07,
      "loss": 0.3518,
      "step": 2400
    },
    {
      "epoch": 34.15,
      "learning_rate": 5.728888888888888e-07,
      "loss": 0.3511,
      "step": 2425
    },
    {
      "epoch": 34.51,
      "learning_rate": 5.673333333333334e-07,
      "loss": 0.3578,
      "step": 2450
    },
    {
      "epoch": 34.86,
      "learning_rate": 5.617777777777778e-07,
      "loss": 0.3536,
      "step": 2475
    },
    {
      "epoch": 35.21,
      "learning_rate": 5.562222222222222e-07,
      "loss": 0.3476,
      "step": 2500
    },
    {
      "epoch": 35.56,
      "learning_rate": 5.506666666666666e-07,
      "loss": 0.3422,
      "step": 2525
    },
    {
      "epoch": 35.92,
      "learning_rate": 5.451111111111111e-07,
      "loss": 0.3547,
      "step": 2550
    },
    {
      "epoch": 36.27,
      "learning_rate": 5.395555555555556e-07,
      "loss": 0.3524,
      "step": 2575
    },
    {
      "epoch": 36.62,
      "learning_rate": 5.34e-07,
      "loss": 0.3481,
      "step": 2600
    },
    {
      "epoch": 36.97,
      "learning_rate": 5.284444444444445e-07,
      "loss": 0.3427,
      "step": 2625
    },
    {
      "epoch": 37.32,
      "learning_rate": 5.228888888888888e-07,
      "loss": 0.3436,
      "step": 2650
    },
    {
      "epoch": 37.68,
      "learning_rate": 5.173333333333333e-07,
      "loss": 0.3431,
      "step": 2675
    },
    {
      "epoch": 38.03,
      "learning_rate": 5.117777777777777e-07,
      "loss": 0.3404,
      "step": 2700
    },
    {
      "epoch": 38.38,
      "learning_rate": 5.062222222222222e-07,
      "loss": 0.3384,
      "step": 2725
    },
    {
      "epoch": 38.73,
      "learning_rate": 5.006666666666667e-07,
      "loss": 0.3465,
      "step": 2750
    },
    {
      "epoch": 39.08,
      "learning_rate": 4.951111111111111e-07,
      "loss": 0.3373,
      "step": 2775
    },
    {
      "epoch": 39.44,
      "learning_rate": 4.895555555555555e-07,
      "loss": 0.3346,
      "step": 2800
    },
    {
      "epoch": 39.79,
      "learning_rate": 4.839999999999999e-07,
      "loss": 0.3366,
      "step": 2825
    },
    {
      "epoch": 40.14,
      "learning_rate": 4.784444444444444e-07,
      "loss": 0.3401,
      "step": 2850
    },
    {
      "epoch": 40.49,
      "learning_rate": 4.728888888888889e-07,
      "loss": 0.3309,
      "step": 2875
    },
    {
      "epoch": 40.85,
      "learning_rate": 4.673333333333333e-07,
      "loss": 0.3399,
      "step": 2900
    },
    {
      "epoch": 41.2,
      "learning_rate": 4.6177777777777777e-07,
      "loss": 0.3296,
      "step": 2925
    },
    {
      "epoch": 41.55,
      "learning_rate": 4.5622222222222217e-07,
      "loss": 0.3325,
      "step": 2950
    },
    {
      "epoch": 41.9,
      "learning_rate": 4.506666666666666e-07,
      "loss": 0.3366,
      "step": 2975
    },
    {
      "epoch": 42.25,
      "learning_rate": 4.451111111111111e-07,
      "loss": 0.3286,
      "step": 3000
    },
    {
      "epoch": 42.25,
      "eval_loss": 0.5238807797431946,
      "eval_runtime": 76.2832,
      "eval_samples_per_second": 118.925,
      "eval_steps_per_second": 0.931,
      "eval_wer": 29.24128413235632,
      "step": 3000
    },
    {
      "epoch": 42.61,
      "learning_rate": 4.3955555555555554e-07,
      "loss": 0.3262,
      "step": 3025
    },
    {
      "epoch": 42.96,
      "learning_rate": 4.34e-07,
      "loss": 0.3296,
      "step": 3050
    },
    {
      "epoch": 43.31,
      "learning_rate": 4.2844444444444445e-07,
      "loss": 0.328,
      "step": 3075
    },
    {
      "epoch": 43.66,
      "learning_rate": 4.2288888888888886e-07,
      "loss": 0.3281,
      "step": 3100
    },
    {
      "epoch": 44.01,
      "learning_rate": 4.173333333333333e-07,
      "loss": 0.3267,
      "step": 3125
    },
    {
      "epoch": 44.37,
      "learning_rate": 4.1177777777777777e-07,
      "loss": 0.3241,
      "step": 3150
    },
    {
      "epoch": 44.72,
      "learning_rate": 4.0622222222222217e-07,
      "loss": 0.3257,
      "step": 3175
    },
    {
      "epoch": 45.07,
      "learning_rate": 4.0066666666666663e-07,
      "loss": 0.3238,
      "step": 3200
    },
    {
      "epoch": 45.42,
      "learning_rate": 3.9511111111111114e-07,
      "loss": 0.3222,
      "step": 3225
    },
    {
      "epoch": 45.77,
      "learning_rate": 3.8955555555555554e-07,
      "loss": 0.3244,
      "step": 3250
    },
    {
      "epoch": 46.13,
      "learning_rate": 3.84e-07,
      "loss": 0.3244,
      "step": 3275
    },
    {
      "epoch": 46.48,
      "learning_rate": 3.7844444444444445e-07,
      "loss": 0.3195,
      "step": 3300
    },
    {
      "epoch": 46.83,
      "learning_rate": 3.7288888888888886e-07,
      "loss": 0.3204,
      "step": 3325
    },
    {
      "epoch": 47.18,
      "learning_rate": 3.673333333333333e-07,
      "loss": 0.319,
      "step": 3350
    },
    {
      "epoch": 47.54,
      "learning_rate": 3.617777777777777e-07,
      "loss": 0.3191,
      "step": 3375
    },
    {
      "epoch": 47.89,
      "learning_rate": 3.5622222222222223e-07,
      "loss": 0.3155,
      "step": 3400
    },
    {
      "epoch": 48.24,
      "learning_rate": 3.506666666666667e-07,
      "loss": 0.3172,
      "step": 3425
    },
    {
      "epoch": 48.59,
      "learning_rate": 3.451111111111111e-07,
      "loss": 0.3127,
      "step": 3450
    },
    {
      "epoch": 48.94,
      "learning_rate": 3.3955555555555554e-07,
      "loss": 0.3217,
      "step": 3475
    },
    {
      "epoch": 49.3,
      "learning_rate": 3.34e-07,
      "loss": 0.3158,
      "step": 3500
    },
    {
      "epoch": 49.65,
      "learning_rate": 3.284444444444444e-07,
      "loss": 0.3149,
      "step": 3525
    },
    {
      "epoch": 50.0,
      "learning_rate": 3.2288888888888886e-07,
      "loss": 0.3137,
      "step": 3550
    },
    {
      "epoch": 50.35,
      "learning_rate": 3.173333333333333e-07,
      "loss": 0.3124,
      "step": 3575
    },
    {
      "epoch": 50.7,
      "learning_rate": 3.1177777777777777e-07,
      "loss": 0.3127,
      "step": 3600
    },
    {
      "epoch": 51.06,
      "learning_rate": 3.0622222222222223e-07,
      "loss": 0.3146,
      "step": 3625
    },
    {
      "epoch": 51.41,
      "learning_rate": 3.006666666666667e-07,
      "loss": 0.3094,
      "step": 3650
    },
    {
      "epoch": 51.76,
      "learning_rate": 2.951111111111111e-07,
      "loss": 0.3107,
      "step": 3675
    },
    {
      "epoch": 52.11,
      "learning_rate": 2.8955555555555555e-07,
      "loss": 0.3104,
      "step": 3700
    },
    {
      "epoch": 52.46,
      "learning_rate": 2.8399999999999995e-07,
      "loss": 0.3048,
      "step": 3725
    },
    {
      "epoch": 52.82,
      "learning_rate": 2.784444444444444e-07,
      "loss": 0.311,
      "step": 3750
    },
    {
      "epoch": 53.17,
      "learning_rate": 2.728888888888889e-07,
      "loss": 0.3109,
      "step": 3775
    },
    {
      "epoch": 53.52,
      "learning_rate": 2.673333333333333e-07,
      "loss": 0.3081,
      "step": 3800
    },
    {
      "epoch": 53.87,
      "learning_rate": 2.617777777777778e-07,
      "loss": 0.3097,
      "step": 3825
    },
    {
      "epoch": 54.23,
      "learning_rate": 2.5622222222222223e-07,
      "loss": 0.3127,
      "step": 3850
    },
    {
      "epoch": 54.58,
      "learning_rate": 2.5066666666666663e-07,
      "loss": 0.3042,
      "step": 3875
    },
    {
      "epoch": 54.93,
      "learning_rate": 2.451111111111111e-07,
      "loss": 0.3071,
      "step": 3900
    },
    {
      "epoch": 55.28,
      "learning_rate": 2.3955555555555555e-07,
      "loss": 0.3102,
      "step": 3925
    },
    {
      "epoch": 55.63,
      "learning_rate": 2.34e-07,
      "loss": 0.3041,
      "step": 3950
    },
    {
      "epoch": 55.99,
      "learning_rate": 2.2844444444444443e-07,
      "loss": 0.3064,
      "step": 3975
    },
    {
      "epoch": 56.34,
      "learning_rate": 2.2288888888888886e-07,
      "loss": 0.3073,
      "step": 4000
    },
    {
      "epoch": 56.34,
      "eval_loss": 0.5199663639068604,
      "eval_runtime": 77.0033,
      "eval_samples_per_second": 117.813,
      "eval_steps_per_second": 0.922,
      "eval_wer": 29.413794236519568,
      "step": 4000
    },
    {
      "epoch": 56.69,
      "learning_rate": 2.1733333333333332e-07,
      "loss": 0.3075,
      "step": 4025
    },
    {
      "epoch": 57.04,
      "learning_rate": 2.1177777777777778e-07,
      "loss": 0.2995,
      "step": 4050
    },
    {
      "epoch": 57.39,
      "learning_rate": 2.062222222222222e-07,
      "loss": 0.3036,
      "step": 4075
    },
    {
      "epoch": 57.75,
      "learning_rate": 2.0066666666666666e-07,
      "loss": 0.3047,
      "step": 4100
    },
    {
      "epoch": 58.1,
      "learning_rate": 1.9511111111111112e-07,
      "loss": 0.3024,
      "step": 4125
    },
    {
      "epoch": 58.45,
      "learning_rate": 1.8955555555555555e-07,
      "loss": 0.301,
      "step": 4150
    },
    {
      "epoch": 58.8,
      "learning_rate": 1.8399999999999998e-07,
      "loss": 0.3011,
      "step": 4175
    },
    {
      "epoch": 59.15,
      "learning_rate": 1.7844444444444444e-07,
      "loss": 0.3056,
      "step": 4200
    },
    {
      "epoch": 59.51,
      "learning_rate": 1.728888888888889e-07,
      "loss": 0.3008,
      "step": 4225
    },
    {
      "epoch": 59.86,
      "learning_rate": 1.6733333333333332e-07,
      "loss": 0.3041,
      "step": 4250
    },
    {
      "epoch": 60.21,
      "learning_rate": 1.6177777777777775e-07,
      "loss": 0.2984,
      "step": 4275
    },
    {
      "epoch": 60.56,
      "learning_rate": 1.5622222222222224e-07,
      "loss": 0.299,
      "step": 4300
    },
    {
      "epoch": 60.92,
      "learning_rate": 1.5066666666666667e-07,
      "loss": 0.3024,
      "step": 4325
    },
    {
      "epoch": 61.27,
      "learning_rate": 1.451111111111111e-07,
      "loss": 0.3005,
      "step": 4350
    },
    {
      "epoch": 61.62,
      "learning_rate": 1.3955555555555553e-07,
      "loss": 0.3016,
      "step": 4375
    },
    {
      "epoch": 61.97,
      "learning_rate": 1.34e-07,
      "loss": 0.2983,
      "step": 4400
    },
    {
      "epoch": 62.32,
      "learning_rate": 1.2844444444444444e-07,
      "loss": 0.2989,
      "step": 4425
    },
    {
      "epoch": 62.68,
      "learning_rate": 1.228888888888889e-07,
      "loss": 0.2973,
      "step": 4450
    },
    {
      "epoch": 63.03,
      "learning_rate": 1.1733333333333333e-07,
      "loss": 0.3012,
      "step": 4475
    },
    {
      "epoch": 63.38,
      "learning_rate": 1.1177777777777778e-07,
      "loss": 0.299,
      "step": 4500
    },
    {
      "epoch": 63.73,
      "learning_rate": 1.0622222222222221e-07,
      "loss": 0.2978,
      "step": 4525
    },
    {
      "epoch": 64.08,
      "learning_rate": 1.0066666666666667e-07,
      "loss": 0.2992,
      "step": 4550
    },
    {
      "epoch": 64.44,
      "learning_rate": 9.51111111111111e-08,
      "loss": 0.2988,
      "step": 4575
    },
    {
      "epoch": 64.79,
      "learning_rate": 8.955555555555555e-08,
      "loss": 0.2989,
      "step": 4600
    },
    {
      "epoch": 65.14,
      "learning_rate": 8.4e-08,
      "loss": 0.2966,
      "step": 4625
    },
    {
      "epoch": 65.49,
      "learning_rate": 7.844444444444444e-08,
      "loss": 0.3011,
      "step": 4650
    },
    {
      "epoch": 65.85,
      "learning_rate": 7.288888888888888e-08,
      "loss": 0.2934,
      "step": 4675
    },
    {
      "epoch": 66.2,
      "learning_rate": 6.733333333333333e-08,
      "loss": 0.2994,
      "step": 4700
    },
    {
      "epoch": 66.55,
      "learning_rate": 6.177777777777777e-08,
      "loss": 0.2979,
      "step": 4725
    },
    {
      "epoch": 66.9,
      "learning_rate": 5.622222222222222e-08,
      "loss": 0.2966,
      "step": 4750
    },
    {
      "epoch": 67.25,
      "learning_rate": 5.0666666666666664e-08,
      "loss": 0.2977,
      "step": 4775
    },
    {
      "epoch": 67.61,
      "learning_rate": 4.511111111111111e-08,
      "loss": 0.2968,
      "step": 4800
    },
    {
      "epoch": 67.96,
      "learning_rate": 3.955555555555555e-08,
      "loss": 0.2983,
      "step": 4825
    },
    {
      "epoch": 68.31,
      "learning_rate": 3.4e-08,
      "loss": 0.2943,
      "step": 4850
    },
    {
      "epoch": 68.66,
      "learning_rate": 2.8444444444444443e-08,
      "loss": 0.2969,
      "step": 4875
    },
    {
      "epoch": 69.01,
      "learning_rate": 2.2888888888888887e-08,
      "loss": 0.2974,
      "step": 4900
    },
    {
      "epoch": 69.37,
      "learning_rate": 1.7333333333333333e-08,
      "loss": 0.2977,
      "step": 4925
    },
    {
      "epoch": 69.72,
      "learning_rate": 1.1777777777777778e-08,
      "loss": 0.2947,
      "step": 4950
    },
    {
      "epoch": 70.07,
      "learning_rate": 6.222222222222222e-09,
      "loss": 0.297,
      "step": 4975
    },
    {
      "epoch": 70.42,
      "learning_rate": 6.666666666666666e-10,
      "loss": 0.2971,
      "step": 5000
    },
    {
      "epoch": 70.42,
      "eval_loss": 0.5191043019294739,
      "eval_runtime": 76.5999,
      "eval_samples_per_second": 118.434,
      "eval_steps_per_second": 0.927,
      "eval_wer": 28.965267965695134,
      "step": 5000
    },
    {
      "epoch": 70.42,
      "step": 5000,
      "total_flos": 3.14570266509312e+19,
      "train_loss": 0.422480837726593,
      "train_runtime": 11176.3562,
      "train_samples_per_second": 114.527,
      "train_steps_per_second": 0.447
    }
  ],
  "logging_steps": 25,
  "max_steps": 5000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 71,
  "save_steps": 1000,
  "total_flos": 3.14570266509312e+19,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}