{
  "best_metric": 53.265028208287404,
  "best_model_checkpoint": "./output/small/yt-special-batch12-lr6-small/checkpoint-4000",
  "epoch": 1.2883277505797475,
  "global_step": 5000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 4.2e-08,
      "loss": 2.3178,
      "step": 25
    },
    {
      "epoch": 0.01,
      "learning_rate": 9.199999999999999e-08,
      "loss": 2.4286,
      "step": 50
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.4199999999999997e-07,
      "loss": 2.2344,
      "step": 75
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.92e-07,
      "loss": 2.2967,
      "step": 100
    },
    {
      "epoch": 0.03,
      "learning_rate": 2.4199999999999997e-07,
      "loss": 2.0184,
      "step": 125
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.9e-07,
      "loss": 2.0577,
      "step": 150
    },
    {
      "epoch": 0.05,
      "learning_rate": 3.4000000000000003e-07,
      "loss": 1.9035,
      "step": 175
    },
    {
      "epoch": 0.05,
      "learning_rate": 3.8999999999999997e-07,
      "loss": 1.8914,
      "step": 200
    },
    {
      "epoch": 0.06,
      "learning_rate": 4.3999999999999997e-07,
      "loss": 1.7214,
      "step": 225
    },
    {
      "epoch": 0.06,
      "learning_rate": 4.9e-07,
      "loss": 1.6912,
      "step": 250
    },
    {
      "epoch": 0.07,
      "learning_rate": 5.4e-07,
      "loss": 1.4595,
      "step": 275
    },
    {
      "epoch": 0.08,
      "learning_rate": 5.9e-07,
      "loss": 1.4545,
      "step": 300
    },
    {
      "epoch": 0.08,
      "learning_rate": 6.4e-07,
      "loss": 1.2783,
      "step": 325
    },
    {
      "epoch": 0.09,
      "learning_rate": 6.9e-07,
      "loss": 1.2624,
      "step": 350
    },
    {
      "epoch": 0.1,
      "learning_rate": 7.4e-07,
      "loss": 1.1341,
      "step": 375
    },
    {
      "epoch": 0.1,
      "learning_rate": 7.9e-07,
      "loss": 1.184,
      "step": 400
    },
    {
      "epoch": 0.11,
      "learning_rate": 8.399999999999999e-07,
      "loss": 1.1445,
      "step": 425
    },
    {
      "epoch": 0.12,
      "learning_rate": 8.9e-07,
      "loss": 1.0971,
      "step": 450
    },
    {
      "epoch": 0.12,
      "learning_rate": 9.399999999999999e-07,
      "loss": 0.9895,
      "step": 475
    },
    {
      "epoch": 0.13,
      "learning_rate": 9.9e-07,
      "loss": 1.0503,
      "step": 500
    },
    {
      "epoch": 0.14,
      "learning_rate": 9.955555555555554e-07,
      "loss": 1.0095,
      "step": 525
    },
    {
      "epoch": 0.14,
      "learning_rate": 9.9e-07,
      "loss": 0.9494,
      "step": 550
    },
    {
      "epoch": 0.15,
      "learning_rate": 9.844444444444445e-07,
      "loss": 0.954,
      "step": 575
    },
    {
      "epoch": 0.15,
      "learning_rate": 9.788888888888889e-07,
      "loss": 1.0283,
      "step": 600
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.733333333333333e-07,
      "loss": 1.0035,
      "step": 625
    },
    {
      "epoch": 0.17,
      "learning_rate": 9.677777777777777e-07,
      "loss": 0.9427,
      "step": 650
    },
    {
      "epoch": 0.17,
      "learning_rate": 9.62222222222222e-07,
      "loss": 0.9594,
      "step": 675
    },
    {
      "epoch": 0.18,
      "learning_rate": 9.566666666666667e-07,
      "loss": 0.9726,
      "step": 700
    },
    {
      "epoch": 0.19,
      "learning_rate": 9.511111111111111e-07,
      "loss": 1.012,
      "step": 725
    },
    {
      "epoch": 0.19,
      "learning_rate": 9.455555555555556e-07,
      "loss": 0.9354,
      "step": 750
    },
    {
      "epoch": 0.2,
      "learning_rate": 9.399999999999999e-07,
      "loss": 1.0653,
      "step": 775
    },
    {
      "epoch": 0.21,
      "learning_rate": 9.344444444444444e-07,
      "loss": 1.0245,
      "step": 800
    },
    {
      "epoch": 0.21,
      "learning_rate": 9.288888888888888e-07,
      "loss": 0.9691,
      "step": 825
    },
    {
      "epoch": 0.22,
      "learning_rate": 9.233333333333333e-07,
      "loss": 0.915,
      "step": 850
    },
    {
      "epoch": 0.23,
      "learning_rate": 9.177777777777777e-07,
      "loss": 0.9808,
      "step": 875
    },
    {
      "epoch": 0.23,
      "learning_rate": 9.122222222222222e-07,
      "loss": 0.9037,
      "step": 900
    },
    {
      "epoch": 0.24,
      "learning_rate": 9.066666666666665e-07,
      "loss": 0.9195,
      "step": 925
    },
    {
      "epoch": 0.24,
      "learning_rate": 9.01111111111111e-07,
      "loss": 0.9472,
      "step": 950
    },
    {
      "epoch": 0.25,
      "learning_rate": 8.955555555555555e-07,
      "loss": 0.9243,
      "step": 975
    },
    {
      "epoch": 0.26,
      "learning_rate": 8.9e-07,
      "loss": 0.923,
      "step": 1000
    },
    {
      "epoch": 0.26,
      "eval_loss": 0.9312085509300232,
      "eval_runtime": 468.6663,
      "eval_samples_per_second": 5.23,
      "eval_steps_per_second": 0.873,
      "eval_wer": 66.94118409960443,
      "step": 1000
    },
    {
      "epoch": 0.26,
      "learning_rate": 8.844444444444445e-07,
      "loss": 0.9581,
      "step": 1025
    },
    {
      "epoch": 0.27,
      "learning_rate": 8.788888888888889e-07,
      "loss": 0.9608,
      "step": 1050
    },
    {
      "epoch": 0.28,
      "learning_rate": 8.733333333333333e-07,
      "loss": 0.8676,
      "step": 1075
    },
    {
      "epoch": 0.28,
      "learning_rate": 8.677777777777777e-07,
      "loss": 0.9396,
      "step": 1100
    },
    {
      "epoch": 0.29,
      "learning_rate": 8.622222222222222e-07,
      "loss": 0.9186,
      "step": 1125
    },
    {
      "epoch": 0.3,
      "learning_rate": 8.566666666666667e-07,
      "loss": 0.8423,
      "step": 1150
    },
    {
      "epoch": 0.3,
      "learning_rate": 8.511111111111111e-07,
      "loss": 0.9042,
      "step": 1175
    },
    {
      "epoch": 0.31,
      "learning_rate": 8.455555555555555e-07,
      "loss": 0.8735,
      "step": 1200
    },
    {
      "epoch": 0.32,
      "learning_rate": 8.399999999999999e-07,
      "loss": 0.9525,
      "step": 1225
    },
    {
      "epoch": 0.32,
      "learning_rate": 8.344444444444444e-07,
      "loss": 0.8673,
      "step": 1250
    },
    {
      "epoch": 0.33,
      "learning_rate": 8.288888888888888e-07,
      "loss": 0.9351,
      "step": 1275
    },
    {
      "epoch": 0.33,
      "learning_rate": 8.233333333333333e-07,
      "loss": 0.9544,
      "step": 1300
    },
    {
      "epoch": 0.34,
      "learning_rate": 8.177777777777778e-07,
      "loss": 0.8801,
      "step": 1325
    },
    {
      "epoch": 0.35,
      "learning_rate": 8.122222222222221e-07,
      "loss": 0.8719,
      "step": 1350
    },
    {
      "epoch": 0.35,
      "learning_rate": 8.066666666666666e-07,
      "loss": 0.9392,
      "step": 1375
    },
    {
      "epoch": 0.36,
      "learning_rate": 8.01111111111111e-07,
      "loss": 0.8593,
      "step": 1400
    },
    {
      "epoch": 0.37,
      "learning_rate": 7.955555555555556e-07,
      "loss": 0.8251,
      "step": 1425
    },
    {
      "epoch": 0.37,
      "learning_rate": 7.9e-07,
      "loss": 0.9402,
      "step": 1450
    },
    {
      "epoch": 0.38,
      "learning_rate": 7.844444444444445e-07,
      "loss": 0.9226,
      "step": 1475
    },
    {
      "epoch": 0.39,
      "learning_rate": 7.788888888888889e-07,
      "loss": 0.7802,
      "step": 1500
    },
    {
      "epoch": 0.39,
      "learning_rate": 7.733333333333333e-07,
      "loss": 0.8417,
      "step": 1525
    },
    {
      "epoch": 0.4,
      "learning_rate": 7.677777777777778e-07,
      "loss": 0.9365,
      "step": 1550
    },
    {
      "epoch": 0.41,
      "learning_rate": 7.622222222222222e-07,
      "loss": 0.9738,
      "step": 1575
    },
    {
      "epoch": 0.41,
      "learning_rate": 7.566666666666667e-07,
      "loss": 0.8712,
      "step": 1600
    },
    {
      "epoch": 0.42,
      "learning_rate": 7.51111111111111e-07,
      "loss": 0.894,
      "step": 1625
    },
    {
      "epoch": 0.43,
      "learning_rate": 7.455555555555555e-07,
      "loss": 0.8608,
      "step": 1650
    },
    {
      "epoch": 0.43,
      "learning_rate": 7.4e-07,
      "loss": 0.8751,
      "step": 1675
    },
    {
      "epoch": 0.44,
      "learning_rate": 7.344444444444444e-07,
      "loss": 0.8743,
      "step": 1700
    },
    {
      "epoch": 0.44,
      "learning_rate": 7.288888888888889e-07,
      "loss": 0.8679,
      "step": 1725
    },
    {
      "epoch": 0.45,
      "learning_rate": 7.233333333333333e-07,
      "loss": 0.8558,
      "step": 1750
    },
    {
      "epoch": 0.46,
      "learning_rate": 7.177777777777777e-07,
      "loss": 0.8798,
      "step": 1775
    },
    {
      "epoch": 0.46,
      "learning_rate": 7.122222222222221e-07,
      "loss": 0.9412,
      "step": 1800
    },
    {
      "epoch": 0.47,
      "learning_rate": 7.066666666666666e-07,
      "loss": 0.9113,
      "step": 1825
    },
    {
      "epoch": 0.48,
      "learning_rate": 7.01111111111111e-07,
      "loss": 0.9288,
      "step": 1850
    },
    {
      "epoch": 0.48,
      "learning_rate": 6.955555555555556e-07,
      "loss": 0.7912,
      "step": 1875
    },
    {
      "epoch": 0.49,
      "learning_rate": 6.9e-07,
      "loss": 0.8788,
      "step": 1900
    },
    {
      "epoch": 0.5,
      "learning_rate": 6.844444444444444e-07,
      "loss": 0.8745,
      "step": 1925
    },
    {
      "epoch": 0.5,
      "learning_rate": 6.788888888888889e-07,
      "loss": 0.9003,
      "step": 1950
    },
    {
      "epoch": 0.51,
      "learning_rate": 6.733333333333333e-07,
      "loss": 0.9637,
      "step": 1975
    },
    {
      "epoch": 0.52,
      "learning_rate": 6.677777777777778e-07,
      "loss": 0.8161,
      "step": 2000
    },
    {
      "epoch": 0.52,
      "eval_loss": 0.864041268825531,
      "eval_runtime": 449.2777,
      "eval_samples_per_second": 5.455,
      "eval_steps_per_second": 0.91,
      "eval_wer": 58.569483172297524,
      "step": 2000
    },
    {
      "epoch": 0.52,
      "learning_rate": 6.622222222222222e-07,
      "loss": 0.8642,
      "step": 2025
    },
    {
      "epoch": 0.53,
      "learning_rate": 6.566666666666666e-07,
      "loss": 0.7682,
      "step": 2050
    },
    {
      "epoch": 0.53,
      "learning_rate": 6.511111111111111e-07,
      "loss": 0.8596,
      "step": 2075
    },
    {
      "epoch": 0.54,
      "learning_rate": 6.455555555555555e-07,
      "loss": 0.9109,
      "step": 2100
    },
    {
      "epoch": 0.55,
      "learning_rate": 6.4e-07,
      "loss": 0.8608,
      "step": 2125
    },
    {
      "epoch": 0.55,
      "learning_rate": 6.344444444444444e-07,
      "loss": 0.902,
      "step": 2150
    },
    {
      "epoch": 0.56,
      "learning_rate": 6.288888888888889e-07,
      "loss": 0.8274,
      "step": 2175
    },
    {
      "epoch": 0.57,
      "learning_rate": 6.233333333333332e-07,
      "loss": 0.8097,
      "step": 2200
    },
    {
      "epoch": 0.57,
      "learning_rate": 6.177777777777777e-07,
      "loss": 0.8095,
      "step": 2225
    },
    {
      "epoch": 0.58,
      "learning_rate": 6.122222222222222e-07,
      "loss": 0.8014,
      "step": 2250
    },
    {
      "epoch": 0.59,
      "learning_rate": 6.066666666666666e-07,
      "loss": 0.8092,
      "step": 2275
    },
    {
      "epoch": 0.59,
      "learning_rate": 6.011111111111112e-07,
      "loss": 0.9165,
      "step": 2300
    },
    {
      "epoch": 0.6,
      "learning_rate": 5.955555555555555e-07,
      "loss": 0.9182,
      "step": 2325
    },
    {
      "epoch": 0.61,
      "learning_rate": 5.9e-07,
      "loss": 0.8517,
      "step": 2350
    },
    {
      "epoch": 0.61,
      "learning_rate": 5.844444444444444e-07,
      "loss": 0.7592,
      "step": 2375
    },
    {
      "epoch": 0.62,
      "learning_rate": 5.788888888888889e-07,
      "loss": 0.8024,
      "step": 2400
    },
    {
      "epoch": 0.62,
      "learning_rate": 5.733333333333334e-07,
      "loss": 0.8586,
      "step": 2425
    },
    {
      "epoch": 0.63,
      "learning_rate": 5.677777777777778e-07,
      "loss": 0.8403,
      "step": 2450
    },
    {
      "epoch": 0.64,
      "learning_rate": 5.622222222222222e-07,
      "loss": 0.9014,
      "step": 2475
    },
    {
      "epoch": 0.64,
      "learning_rate": 5.566666666666666e-07,
      "loss": 0.8146,
      "step": 2500
    },
    {
      "epoch": 0.65,
      "learning_rate": 5.511111111111111e-07,
      "loss": 0.8347,
      "step": 2525
    },
    {
      "epoch": 0.66,
      "learning_rate": 5.455555555555555e-07,
      "loss": 0.8358,
      "step": 2550
    },
    {
      "epoch": 0.66,
      "learning_rate": 5.4e-07,
      "loss": 0.838,
      "step": 2575
    },
    {
      "epoch": 0.67,
      "learning_rate": 5.344444444444445e-07,
      "loss": 0.8847,
      "step": 2600
    },
    {
      "epoch": 0.68,
      "learning_rate": 5.288888888888888e-07,
      "loss": 0.7753,
      "step": 2625
    },
    {
      "epoch": 0.68,
      "learning_rate": 5.233333333333333e-07,
      "loss": 0.8078,
      "step": 2650
    },
    {
      "epoch": 0.69,
      "learning_rate": 5.177777777777777e-07,
      "loss": 0.8332,
      "step": 2675
    },
    {
      "epoch": 0.7,
      "learning_rate": 5.122222222222222e-07,
      "loss": 0.8811,
      "step": 2700
    },
    {
      "epoch": 0.7,
      "learning_rate": 5.066666666666667e-07,
      "loss": 0.7826,
      "step": 2725
    },
    {
      "epoch": 0.71,
      "learning_rate": 5.011111111111111e-07,
      "loss": 0.8094,
      "step": 2750
    },
    {
      "epoch": 0.72,
      "learning_rate": 4.955555555555556e-07,
      "loss": 0.8297,
      "step": 2775
    },
    {
      "epoch": 0.72,
      "learning_rate": 4.9e-07,
      "loss": 0.8302,
      "step": 2800
    },
    {
      "epoch": 0.73,
      "learning_rate": 4.844444444444445e-07,
      "loss": 0.839,
      "step": 2825
    },
    {
      "epoch": 0.73,
      "learning_rate": 4.788888888888889e-07,
      "loss": 0.9498,
      "step": 2850
    },
    {
      "epoch": 0.74,
      "learning_rate": 4.733333333333333e-07,
      "loss": 0.7662,
      "step": 2875
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.677777777777778e-07,
      "loss": 0.873,
      "step": 2900
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.622222222222222e-07,
      "loss": 0.8401,
      "step": 2925
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.5666666666666665e-07,
      "loss": 0.9097,
      "step": 2950
    },
    {
      "epoch": 0.77,
      "learning_rate": 4.511111111111111e-07,
      "loss": 0.8763,
      "step": 2975
    },
    {
      "epoch": 0.77,
      "learning_rate": 4.455555555555555e-07,
      "loss": 0.8542,
      "step": 3000
    },
    {
      "epoch": 0.77,
      "eval_loss": 0.8354777097702026,
      "eval_runtime": 440.7806,
      "eval_samples_per_second": 5.561,
      "eval_steps_per_second": 0.928,
      "eval_wer": 57.70053822709292,
      "step": 3000
    },
    {
      "epoch": 0.78,
      "learning_rate": 4.3999999999999997e-07,
      "loss": 0.7738,
      "step": 3025
    },
    {
      "epoch": 0.79,
      "learning_rate": 4.344444444444444e-07,
      "loss": 0.8802,
      "step": 3050
    },
    {
      "epoch": 0.79,
      "learning_rate": 4.2888888888888883e-07,
      "loss": 0.8181,
      "step": 3075
    },
    {
      "epoch": 0.8,
      "learning_rate": 4.2333333333333334e-07,
      "loss": 0.8766,
      "step": 3100
    },
    {
      "epoch": 0.81,
      "learning_rate": 4.177777777777778e-07,
      "loss": 0.8512,
      "step": 3125
    },
    {
      "epoch": 0.81,
      "learning_rate": 4.122222222222222e-07,
      "loss": 0.8954,
      "step": 3150
    },
    {
      "epoch": 0.82,
      "learning_rate": 4.0666666666666666e-07,
      "loss": 0.7475,
      "step": 3175
    },
    {
      "epoch": 0.82,
      "learning_rate": 4.0111111111111106e-07,
      "loss": 0.8981,
      "step": 3200
    },
    {
      "epoch": 0.83,
      "learning_rate": 3.955555555555555e-07,
      "loss": 0.8088,
      "step": 3225
    },
    {
      "epoch": 0.84,
      "learning_rate": 3.8999999999999997e-07,
      "loss": 0.7649,
      "step": 3250
    },
    {
      "epoch": 0.84,
      "learning_rate": 3.8444444444444443e-07,
      "loss": 0.8483,
      "step": 3275
    },
    {
      "epoch": 0.85,
      "learning_rate": 3.788888888888889e-07,
      "loss": 0.7805,
      "step": 3300
    },
    {
      "epoch": 0.86,
      "learning_rate": 3.7333333333333334e-07,
      "loss": 0.7201,
      "step": 3325
    },
    {
      "epoch": 0.86,
      "learning_rate": 3.6777777777777774e-07,
      "loss": 0.8282,
      "step": 3350
    },
    {
      "epoch": 0.87,
      "learning_rate": 3.622222222222222e-07,
      "loss": 0.8194,
      "step": 3375
    },
    {
      "epoch": 0.88,
      "learning_rate": 3.5666666666666666e-07,
      "loss": 0.867,
      "step": 3400
    },
    {
      "epoch": 0.88,
      "learning_rate": 3.5111111111111106e-07,
      "loss": 0.8162,
      "step": 3425
    },
    {
      "epoch": 0.89,
      "learning_rate": 3.4555555555555557e-07,
      "loss": 0.8144,
      "step": 3450
    },
    {
      "epoch": 0.9,
      "learning_rate": 3.4000000000000003e-07,
      "loss": 0.7685,
      "step": 3475
    },
    {
      "epoch": 0.9,
      "learning_rate": 3.3444444444444443e-07,
      "loss": 0.7935,
      "step": 3500
    },
    {
      "epoch": 0.91,
      "learning_rate": 3.288888888888889e-07,
      "loss": 0.835,
      "step": 3525
    },
    {
      "epoch": 0.91,
      "learning_rate": 3.233333333333333e-07,
      "loss": 0.7847,
      "step": 3550
    },
    {
      "epoch": 0.92,
      "learning_rate": 3.1777777777777775e-07,
      "loss": 0.7997,
      "step": 3575
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.122222222222222e-07,
      "loss": 0.8047,
      "step": 3600
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.066666666666666e-07,
      "loss": 0.8634,
      "step": 3625
    },
    {
      "epoch": 0.94,
      "learning_rate": 3.011111111111111e-07,
      "loss": 0.8143,
      "step": 3650
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.9555555555555557e-07,
      "loss": 0.8192,
      "step": 3675
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.9e-07,
      "loss": 0.6708,
      "step": 3700
    },
    {
      "epoch": 0.96,
      "learning_rate": 2.8444444444444443e-07,
      "loss": 0.8038,
      "step": 3725
    },
    {
      "epoch": 0.97,
      "learning_rate": 2.788888888888889e-07,
      "loss": 0.8432,
      "step": 3750
    },
    {
      "epoch": 0.97,
      "learning_rate": 2.733333333333333e-07,
      "loss": 0.776,
      "step": 3775
    },
    {
      "epoch": 0.98,
      "learning_rate": 2.6777777777777775e-07,
      "loss": 0.7459,
      "step": 3800
    },
    {
      "epoch": 0.99,
      "learning_rate": 2.6222222222222226e-07,
      "loss": 0.8695,
      "step": 3825
    },
    {
      "epoch": 0.99,
      "learning_rate": 2.5666666666666666e-07,
      "loss": 0.8002,
      "step": 3850
    },
    {
      "epoch": 1.0,
      "learning_rate": 2.511111111111111e-07,
      "loss": 0.8695,
      "step": 3875
    },
    {
      "epoch": 1.0,
      "learning_rate": 2.455555555555555e-07,
      "loss": 0.8251,
      "step": 3900
    },
    {
      "epoch": 1.01,
      "learning_rate": 2.4e-07,
      "loss": 0.813,
      "step": 3925
    },
    {
      "epoch": 1.02,
      "learning_rate": 2.3444444444444444e-07,
      "loss": 0.675,
      "step": 3950
    },
    {
      "epoch": 1.02,
      "learning_rate": 2.288888888888889e-07,
      "loss": 0.7864,
      "step": 3975
    },
    {
      "epoch": 1.03,
      "learning_rate": 2.2333333333333332e-07,
      "loss": 0.8727,
      "step": 4000
    },
    {
      "epoch": 1.03,
      "eval_loss": 0.8224001526832581,
      "eval_runtime": 431.8493,
      "eval_samples_per_second": 5.676,
      "eval_steps_per_second": 0.947,
      "eval_wer": 53.265028208287404,
      "step": 4000
    },
    {
      "epoch": 1.04,
      "learning_rate": 2.1777777777777775e-07,
      "loss": 0.751,
      "step": 4025
    },
    {
      "epoch": 1.04,
      "learning_rate": 2.122222222222222e-07,
      "loss": 0.7651,
      "step": 4050
    },
    {
      "epoch": 1.05,
      "learning_rate": 2.0666666666666666e-07,
      "loss": 0.7363,
      "step": 4075
    },
    {
      "epoch": 1.06,
      "learning_rate": 2.011111111111111e-07,
      "loss": 0.795,
      "step": 4100
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.9555555555555555e-07,
      "loss": 0.8456,
      "step": 4125
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.8999999999999998e-07,
      "loss": 0.9163,
      "step": 4150
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.8444444444444444e-07,
      "loss": 0.8098,
      "step": 4175
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.7888888888888887e-07,
      "loss": 0.7636,
      "step": 4200
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.7333333333333332e-07,
      "loss": 0.7904,
      "step": 4225
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.6777777777777778e-07,
      "loss": 0.756,
      "step": 4250
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.622222222222222e-07,
      "loss": 0.7541,
      "step": 4275
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.5666666666666667e-07,
      "loss": 0.7267,
      "step": 4300
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.511111111111111e-07,
      "loss": 0.8279,
      "step": 4325
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.4555555555555555e-07,
      "loss": 0.7165,
      "step": 4350
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.4e-07,
      "loss": 0.7547,
      "step": 4375
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.3444444444444444e-07,
      "loss": 0.7647,
      "step": 4400
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2888888888888887e-07,
      "loss": 0.8198,
      "step": 4425
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2333333333333333e-07,
      "loss": 0.7377,
      "step": 4450
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.1777777777777777e-07,
      "loss": 0.7154,
      "step": 4475
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.1222222222222221e-07,
      "loss": 0.6821,
      "step": 4500
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.0666666666666667e-07,
      "loss": 0.7268,
      "step": 4525
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.011111111111111e-07,
      "loss": 0.8093,
      "step": 4550
    },
    {
      "epoch": 1.18,
      "learning_rate": 9.555555555555556e-08,
      "loss": 0.7444,
      "step": 4575
    },
    {
      "epoch": 1.19,
      "learning_rate": 9e-08,
      "loss": 0.7889,
      "step": 4600
    },
    {
      "epoch": 1.19,
      "learning_rate": 8.444444444444444e-08,
      "loss": 0.7876,
      "step": 4625
    },
    {
      "epoch": 1.2,
      "learning_rate": 7.888888888888889e-08,
      "loss": 0.8238,
      "step": 4650
    },
    {
      "epoch": 1.2,
      "learning_rate": 7.333333333333333e-08,
      "loss": 0.7764,
      "step": 4675
    },
    {
      "epoch": 1.21,
      "learning_rate": 6.777777777777778e-08,
      "loss": 0.7734,
      "step": 4700
    },
    {
      "epoch": 1.22,
      "learning_rate": 6.222222222222221e-08,
      "loss": 0.9015,
      "step": 4725
    },
    {
      "epoch": 1.22,
      "learning_rate": 5.666666666666666e-08,
      "loss": 0.7226,
      "step": 4750
    },
    {
      "epoch": 1.23,
      "learning_rate": 5.1111111111111114e-08,
      "loss": 0.7967,
      "step": 4775
    },
    {
      "epoch": 1.24,
      "learning_rate": 4.555555555555556e-08,
      "loss": 0.75,
      "step": 4800
    },
    {
      "epoch": 1.24,
      "learning_rate": 4e-08,
      "loss": 0.7688,
      "step": 4825
    },
    {
      "epoch": 1.25,
      "learning_rate": 3.4444444444444444e-08,
      "loss": 0.7322,
      "step": 4850
    },
    {
      "epoch": 1.26,
      "learning_rate": 2.8888888888888887e-08,
      "loss": 0.8172,
      "step": 4875
    },
    {
      "epoch": 1.26,
      "learning_rate": 2.3333333333333334e-08,
      "loss": 0.8094,
      "step": 4900
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.7777777777777777e-08,
      "loss": 0.8129,
      "step": 4925
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.2222222222222222e-08,
      "loss": 0.7378,
      "step": 4950
    },
    {
      "epoch": 1.28,
      "learning_rate": 6.666666666666667e-09,
      "loss": 0.6779,
      "step": 4975
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.111111111111111e-09,
      "loss": 0.7896,
      "step": 5000
    },
    {
      "epoch": 1.29,
      "eval_loss": 0.8183886408805847,
      "eval_runtime": 548.5151,
      "eval_samples_per_second": 4.468,
      "eval_steps_per_second": 0.746,
      "eval_wer": 53.569807405486024,
      "step": 5000
    },
    {
      "epoch": 1.29,
      "step": 5000,
      "total_flos": 1.731454684839936e+19,
      "train_loss": 0.9241970775604248,
      "train_runtime": 9377.2484,
      "train_samples_per_second": 6.398,
      "train_steps_per_second": 0.533
    }
  ],
  "max_steps": 5000,
  "num_train_epochs": 2,
  "total_flos": 1.731454684839936e+19,
  "trial_name": null,
  "trial_params": null
}