{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 14.336917562724015,
  "global_step": 4000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09,
      "learning_rate": 4.558957377820063e-06,
      "loss": 4.7495,
      "step": 25
    },
    {
      "epoch": 0.18,
      "learning_rate": 6.014328754265799e-06,
      "loss": 2.6193,
      "step": 50
    },
    {
      "epoch": 0.27,
      "learning_rate": 6.765821034569313e-06,
      "loss": 1.6513,
      "step": 75
    },
    {
      "epoch": 0.36,
      "learning_rate": 7.276063921387198e-06,
      "loss": 0.9913,
      "step": 100
    },
    {
      "epoch": 0.45,
      "learning_rate": 7.662870867121632e-06,
      "loss": 0.4604,
      "step": 125
    },
    {
      "epoch": 0.54,
      "learning_rate": 7.974480416326626e-06,
      "loss": 0.2504,
      "step": 150
    },
    {
      "epoch": 0.63,
      "learning_rate": 8.235424875329062e-06,
      "loss": 0.1962,
      "step": 175
    },
    {
      "epoch": 0.72,
      "learning_rate": 8.459898498446897e-06,
      "loss": 0.1723,
      "step": 200
    },
    {
      "epoch": 0.81,
      "learning_rate": 8.656856986535236e-06,
      "loss": 0.1621,
      "step": 225
    },
    {
      "epoch": 0.9,
      "learning_rate": 8.832315150411909e-06,
      "loss": 0.1584,
      "step": 250
    },
    {
      "epoch": 0.99,
      "learning_rate": 8.990508443836999e-06,
      "loss": 0.1541,
      "step": 275
    },
    {
      "epoch": 1.08,
      "learning_rate": 9.13453223817852e-06,
      "loss": 0.1217,
      "step": 300
    },
    {
      "epoch": 1.16,
      "learning_rate": 9.266717518904195e-06,
      "loss": 0.1027,
      "step": 325
    },
    {
      "epoch": 1.25,
      "learning_rate": 9.388863536775534e-06,
      "loss": 0.1145,
      "step": 350
    },
    {
      "epoch": 1.34,
      "learning_rate": 9.502388170790473e-06,
      "loss": 0.0976,
      "step": 375
    },
    {
      "epoch": 1.43,
      "learning_rate": 9.608428633346146e-06,
      "loss": 0.1059,
      "step": 400
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.707910983687811e-06,
      "loss": 0.1032,
      "step": 425
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.801599369756241e-06,
      "loss": 0.1071,
      "step": 450
    },
    {
      "epoch": 1.7,
      "learning_rate": 9.890131703122795e-06,
      "loss": 0.0938,
      "step": 475
    },
    {
      "epoch": 1.79,
      "learning_rate": 9.974046019194712e-06,
      "loss": 0.1066,
      "step": 500
    },
    {
      "epoch": 1.88,
      "learning_rate": 9.954285714285715e-06,
      "loss": 0.0994,
      "step": 525
    },
    {
      "epoch": 1.97,
      "learning_rate": 9.882857142857144e-06,
      "loss": 0.0884,
      "step": 550
    },
    {
      "epoch": 2.06,
      "learning_rate": 9.811428571428571e-06,
      "loss": 0.0736,
      "step": 575
    },
    {
      "epoch": 2.15,
      "learning_rate": 9.74e-06,
      "loss": 0.0493,
      "step": 600
    },
    {
      "epoch": 2.24,
      "learning_rate": 9.668571428571429e-06,
      "loss": 0.0636,
      "step": 625
    },
    {
      "epoch": 2.33,
      "learning_rate": 9.597142857142858e-06,
      "loss": 0.0534,
      "step": 650
    },
    {
      "epoch": 2.42,
      "learning_rate": 9.525714285714287e-06,
      "loss": 0.0651,
      "step": 675
    },
    {
      "epoch": 2.51,
      "learning_rate": 9.454285714285714e-06,
      "loss": 0.052,
      "step": 700
    },
    {
      "epoch": 2.6,
      "learning_rate": 9.382857142857143e-06,
      "loss": 0.0632,
      "step": 725
    },
    {
      "epoch": 2.69,
      "learning_rate": 9.311428571428572e-06,
      "loss": 0.058,
      "step": 750
    },
    {
      "epoch": 2.78,
      "learning_rate": 9.240000000000001e-06,
      "loss": 0.0638,
      "step": 775
    },
    {
      "epoch": 2.87,
      "learning_rate": 9.168571428571429e-06,
      "loss": 0.0553,
      "step": 800
    },
    {
      "epoch": 2.96,
      "learning_rate": 9.097142857142858e-06,
      "loss": 0.0561,
      "step": 825
    },
    {
      "epoch": 3.05,
      "learning_rate": 9.025714285714287e-06,
      "loss": 0.0483,
      "step": 850
    },
    {
      "epoch": 3.14,
      "learning_rate": 8.954285714285716e-06,
      "loss": 0.0293,
      "step": 875
    },
    {
      "epoch": 3.23,
      "learning_rate": 8.882857142857143e-06,
      "loss": 0.0385,
      "step": 900
    },
    {
      "epoch": 3.32,
      "learning_rate": 8.811428571428572e-06,
      "loss": 0.0348,
      "step": 925
    },
    {
      "epoch": 3.41,
      "learning_rate": 8.740000000000001e-06,
      "loss": 0.0406,
      "step": 950
    },
    {
      "epoch": 3.49,
      "learning_rate": 8.66857142857143e-06,
      "loss": 0.0329,
      "step": 975
    },
    {
      "epoch": 3.58,
      "learning_rate": 8.597142857142859e-06,
      "loss": 0.0349,
      "step": 1000
    },
    {
      "epoch": 3.58,
      "eval_loss": 0.1622314453125,
      "eval_runtime": 624.6921,
      "eval_samples_per_second": 3.541,
      "eval_steps_per_second": 0.223,
      "eval_wer": 9.943665890044164,
      "step": 1000
    },
    {
      "epoch": 3.67,
      "learning_rate": 8.525714285714286e-06,
      "loss": 0.036,
      "step": 1025
    },
    {
      "epoch": 3.76,
      "learning_rate": 8.454285714285715e-06,
      "loss": 0.0446,
      "step": 1050
    },
    {
      "epoch": 3.85,
      "learning_rate": 8.382857142857144e-06,
      "loss": 0.0332,
      "step": 1075
    },
    {
      "epoch": 3.94,
      "learning_rate": 8.311428571428572e-06,
      "loss": 0.0336,
      "step": 1100
    },
    {
      "epoch": 4.03,
      "learning_rate": 8.24e-06,
      "loss": 0.0308,
      "step": 1125
    },
    {
      "epoch": 4.12,
      "learning_rate": 8.16857142857143e-06,
      "loss": 0.0185,
      "step": 1150
    },
    {
      "epoch": 4.21,
      "learning_rate": 8.097142857142857e-06,
      "loss": 0.0243,
      "step": 1175
    },
    {
      "epoch": 4.3,
      "learning_rate": 8.025714285714286e-06,
      "loss": 0.0195,
      "step": 1200
    },
    {
      "epoch": 4.39,
      "learning_rate": 7.954285714285715e-06,
      "loss": 0.0253,
      "step": 1225
    },
    {
      "epoch": 4.48,
      "learning_rate": 7.882857142857142e-06,
      "loss": 0.019,
      "step": 1250
    },
    {
      "epoch": 4.57,
      "learning_rate": 7.811428571428571e-06,
      "loss": 0.0248,
      "step": 1275
    },
    {
      "epoch": 4.66,
      "learning_rate": 7.74e-06,
      "loss": 0.0216,
      "step": 1300
    },
    {
      "epoch": 4.75,
      "learning_rate": 7.66857142857143e-06,
      "loss": 0.0236,
      "step": 1325
    },
    {
      "epoch": 4.84,
      "learning_rate": 7.597142857142858e-06,
      "loss": 0.0211,
      "step": 1350
    },
    {
      "epoch": 4.93,
      "learning_rate": 7.525714285714286e-06,
      "loss": 0.0212,
      "step": 1375
    },
    {
      "epoch": 5.02,
      "learning_rate": 7.454285714285715e-06,
      "loss": 0.0188,
      "step": 1400
    },
    {
      "epoch": 5.11,
      "learning_rate": 7.382857142857144e-06,
      "loss": 0.0116,
      "step": 1425
    },
    {
      "epoch": 5.2,
      "learning_rate": 7.311428571428571e-06,
      "loss": 0.0143,
      "step": 1450
    },
    {
      "epoch": 5.29,
      "learning_rate": 7.24e-06,
      "loss": 0.0137,
      "step": 1475
    },
    {
      "epoch": 5.38,
      "learning_rate": 7.168571428571429e-06,
      "loss": 0.013,
      "step": 1500
    },
    {
      "epoch": 5.47,
      "learning_rate": 7.097142857142858e-06,
      "loss": 0.0121,
      "step": 1525
    },
    {
      "epoch": 5.56,
      "learning_rate": 7.025714285714287e-06,
      "loss": 0.0123,
      "step": 1550
    },
    {
      "epoch": 5.65,
      "learning_rate": 6.9542857142857145e-06,
      "loss": 0.0122,
      "step": 1575
    },
    {
      "epoch": 5.73,
      "learning_rate": 6.8828571428571435e-06,
      "loss": 0.0123,
      "step": 1600
    },
    {
      "epoch": 5.82,
      "learning_rate": 6.8114285714285725e-06,
      "loss": 0.0125,
      "step": 1625
    },
    {
      "epoch": 5.91,
      "learning_rate": 6.740000000000001e-06,
      "loss": 0.0146,
      "step": 1650
    },
    {
      "epoch": 6.0,
      "learning_rate": 6.668571428571429e-06,
      "loss": 0.0121,
      "step": 1675
    },
    {
      "epoch": 6.09,
      "learning_rate": 6.597142857142858e-06,
      "loss": 0.0075,
      "step": 1700
    },
    {
      "epoch": 6.18,
      "learning_rate": 6.525714285714286e-06,
      "loss": 0.0086,
      "step": 1725
    },
    {
      "epoch": 6.27,
      "learning_rate": 6.454285714285715e-06,
      "loss": 0.0099,
      "step": 1750
    },
    {
      "epoch": 6.36,
      "learning_rate": 6.382857142857143e-06,
      "loss": 0.0058,
      "step": 1775
    },
    {
      "epoch": 6.45,
      "learning_rate": 6.311428571428571e-06,
      "loss": 0.0093,
      "step": 1800
    },
    {
      "epoch": 6.54,
      "learning_rate": 6.24e-06,
      "loss": 0.0063,
      "step": 1825
    },
    {
      "epoch": 6.63,
      "learning_rate": 6.168571428571429e-06,
      "loss": 0.009,
      "step": 1850
    },
    {
      "epoch": 6.72,
      "learning_rate": 6.097142857142858e-06,
      "loss": 0.0051,
      "step": 1875
    },
    {
      "epoch": 6.81,
      "learning_rate": 6.025714285714286e-06,
      "loss": 0.0079,
      "step": 1900
    },
    {
      "epoch": 6.9,
      "learning_rate": 5.954285714285715e-06,
      "loss": 0.0061,
      "step": 1925
    },
    {
      "epoch": 6.99,
      "learning_rate": 5.882857142857144e-06,
      "loss": 0.0087,
      "step": 1950
    },
    {
      "epoch": 7.08,
      "learning_rate": 5.811428571428572e-06,
      "loss": 0.0076,
      "step": 1975
    },
    {
      "epoch": 7.17,
      "learning_rate": 5.74e-06,
      "loss": 0.0046,
      "step": 2000
    },
    {
      "epoch": 7.17,
      "eval_loss": 0.228759765625,
      "eval_runtime": 624.7011,
      "eval_samples_per_second": 3.541,
      "eval_steps_per_second": 0.223,
      "eval_wer": 9.50898911569357,
      "step": 2000
    },
    {
      "epoch": 7.26,
      "learning_rate": 5.668571428571429e-06,
      "loss": 0.0061,
      "step": 2025
    },
    {
      "epoch": 7.35,
      "learning_rate": 5.597142857142857e-06,
      "loss": 0.0018,
      "step": 2050
    },
    {
      "epoch": 7.44,
      "learning_rate": 5.525714285714286e-06,
      "loss": 0.0049,
      "step": 2075
    },
    {
      "epoch": 7.53,
      "learning_rate": 5.4542857142857145e-06,
      "loss": 0.0032,
      "step": 2100
    },
    {
      "epoch": 7.62,
      "learning_rate": 5.382857142857143e-06,
      "loss": 0.0064,
      "step": 2125
    },
    {
      "epoch": 7.71,
      "learning_rate": 5.311428571428572e-06,
      "loss": 0.0041,
      "step": 2150
    },
    {
      "epoch": 7.8,
      "learning_rate": 5.240000000000001e-06,
      "loss": 0.005,
      "step": 2175
    },
    {
      "epoch": 7.89,
      "learning_rate": 5.16857142857143e-06,
      "loss": 0.0027,
      "step": 2200
    },
    {
      "epoch": 7.97,
      "learning_rate": 5.097142857142857e-06,
      "loss": 0.0043,
      "step": 2225
    },
    {
      "epoch": 8.06,
      "learning_rate": 5.025714285714286e-06,
      "loss": 0.0026,
      "step": 2250
    },
    {
      "epoch": 8.15,
      "learning_rate": 4.954285714285715e-06,
      "loss": 0.0019,
      "step": 2275
    },
    {
      "epoch": 8.24,
      "learning_rate": 4.882857142857143e-06,
      "loss": 0.0035,
      "step": 2300
    },
    {
      "epoch": 8.33,
      "learning_rate": 4.811428571428572e-06,
      "loss": 0.0017,
      "step": 2325
    },
    {
      "epoch": 8.42,
      "learning_rate": 4.74e-06,
      "loss": 0.003,
      "step": 2350
    },
    {
      "epoch": 8.51,
      "learning_rate": 4.668571428571429e-06,
      "loss": 0.0015,
      "step": 2375
    },
    {
      "epoch": 8.6,
      "learning_rate": 4.5971428571428575e-06,
      "loss": 0.0032,
      "step": 2400
    },
    {
      "epoch": 8.69,
      "learning_rate": 4.525714285714286e-06,
      "loss": 0.0017,
      "step": 2425
    },
    {
      "epoch": 8.78,
      "learning_rate": 4.454285714285715e-06,
      "loss": 0.0021,
      "step": 2450
    },
    {
      "epoch": 8.87,
      "learning_rate": 4.382857142857143e-06,
      "loss": 0.0025,
      "step": 2475
    },
    {
      "epoch": 8.96,
      "learning_rate": 4.311428571428572e-06,
      "loss": 0.0035,
      "step": 2500
    },
    {
      "epoch": 9.05,
      "learning_rate": 4.24e-06,
      "loss": 0.0036,
      "step": 2525
    },
    {
      "epoch": 9.14,
      "learning_rate": 4.168571428571429e-06,
      "loss": 0.0009,
      "step": 2550
    },
    {
      "epoch": 9.23,
      "learning_rate": 4.097142857142857e-06,
      "loss": 0.0009,
      "step": 2575
    },
    {
      "epoch": 9.32,
      "learning_rate": 4.025714285714286e-06,
      "loss": 0.0014,
      "step": 2600
    },
    {
      "epoch": 9.41,
      "learning_rate": 3.954285714285714e-06,
      "loss": 0.0017,
      "step": 2625
    },
    {
      "epoch": 9.5,
      "learning_rate": 3.8828571428571434e-06,
      "loss": 0.0006,
      "step": 2650
    },
    {
      "epoch": 9.59,
      "learning_rate": 3.8114285714285716e-06,
      "loss": 0.0011,
      "step": 2675
    },
    {
      "epoch": 9.68,
      "learning_rate": 3.74e-06,
      "loss": 0.0012,
      "step": 2700
    },
    {
      "epoch": 9.77,
      "learning_rate": 3.668571428571429e-06,
      "loss": 0.0026,
      "step": 2725
    },
    {
      "epoch": 9.86,
      "learning_rate": 3.5971428571428574e-06,
      "loss": 0.0011,
      "step": 2750
    },
    {
      "epoch": 9.95,
      "learning_rate": 3.5257142857142864e-06,
      "loss": 0.0006,
      "step": 2775
    },
    {
      "epoch": 10.04,
      "learning_rate": 3.4542857142857145e-06,
      "loss": 0.0012,
      "step": 2800
    },
    {
      "epoch": 10.13,
      "learning_rate": 3.382857142857143e-06,
      "loss": 0.0004,
      "step": 2825
    },
    {
      "epoch": 10.22,
      "learning_rate": 3.3114285714285717e-06,
      "loss": 0.0003,
      "step": 2850
    },
    {
      "epoch": 10.3,
      "learning_rate": 3.2400000000000003e-06,
      "loss": 0.0003,
      "step": 2875
    },
    {
      "epoch": 10.39,
      "learning_rate": 3.1685714285714285e-06,
      "loss": 0.0004,
      "step": 2900
    },
    {
      "epoch": 10.48,
      "learning_rate": 3.0971428571428575e-06,
      "loss": 0.0004,
      "step": 2925
    },
    {
      "epoch": 10.57,
      "learning_rate": 3.0257142857142857e-06,
      "loss": 0.0008,
      "step": 2950
    },
    {
      "epoch": 10.66,
      "learning_rate": 2.9542857142857147e-06,
      "loss": 0.0003,
      "step": 2975
    },
    {
      "epoch": 10.75,
      "learning_rate": 2.8828571428571433e-06,
      "loss": 0.0007,
      "step": 3000
    },
    {
      "epoch": 10.75,
      "eval_loss": 0.281982421875,
      "eval_runtime": 624.9016,
      "eval_samples_per_second": 3.54,
      "eval_steps_per_second": 0.222,
      "eval_wer": 9.095176826511805,
      "step": 3000
    },
    {
      "epoch": 10.84,
      "learning_rate": 2.8114285714285714e-06,
      "loss": 0.0003,
      "step": 3025
    },
    {
      "epoch": 10.93,
      "learning_rate": 2.7400000000000004e-06,
      "loss": 0.0005,
      "step": 3050
    },
    {
      "epoch": 11.02,
      "learning_rate": 2.6685714285714286e-06,
      "loss": 0.0001,
      "step": 3075
    },
    {
      "epoch": 11.11,
      "learning_rate": 2.5971428571428576e-06,
      "loss": 0.0002,
      "step": 3100
    },
    {
      "epoch": 11.2,
      "learning_rate": 2.525714285714286e-06,
      "loss": 0.0003,
      "step": 3125
    },
    {
      "epoch": 11.29,
      "learning_rate": 2.454285714285715e-06,
      "loss": 0.0001,
      "step": 3150
    },
    {
      "epoch": 11.38,
      "learning_rate": 2.382857142857143e-06,
      "loss": 0.0003,
      "step": 3175
    },
    {
      "epoch": 11.47,
      "learning_rate": 2.3114285714285716e-06,
      "loss": 0.0001,
      "step": 3200
    },
    {
      "epoch": 11.56,
      "learning_rate": 2.24e-06,
      "loss": 0.0007,
      "step": 3225
    },
    {
      "epoch": 11.65,
      "learning_rate": 2.1685714285714287e-06,
      "loss": 0.0002,
      "step": 3250
    },
    {
      "epoch": 11.74,
      "learning_rate": 2.0971428571428573e-06,
      "loss": 0.0001,
      "step": 3275
    },
    {
      "epoch": 11.83,
      "learning_rate": 2.025714285714286e-06,
      "loss": 0.0001,
      "step": 3300
    },
    {
      "epoch": 11.92,
      "learning_rate": 1.9542857142857145e-06,
      "loss": 0.0001,
      "step": 3325
    },
    {
      "epoch": 12.01,
      "learning_rate": 1.8828571428571429e-06,
      "loss": 0.0006,
      "step": 3350
    },
    {
      "epoch": 12.1,
      "learning_rate": 1.8114285714285715e-06,
      "loss": 0.0002,
      "step": 3375
    },
    {
      "epoch": 12.19,
      "learning_rate": 1.74e-06,
      "loss": 0.0,
      "step": 3400
    },
    {
      "epoch": 12.28,
      "learning_rate": 1.6685714285714289e-06,
      "loss": 0.0001,
      "step": 3425
    },
    {
      "epoch": 12.37,
      "learning_rate": 1.5971428571428575e-06,
      "loss": 0.0001,
      "step": 3450
    },
    {
      "epoch": 12.46,
      "learning_rate": 1.5257142857142858e-06,
      "loss": 0.0001,
      "step": 3475
    },
    {
      "epoch": 12.54,
      "learning_rate": 1.4542857142857144e-06,
      "loss": 0.0,
      "step": 3500
    },
    {
      "epoch": 12.63,
      "learning_rate": 1.382857142857143e-06,
      "loss": 0.0001,
      "step": 3525
    },
    {
      "epoch": 12.72,
      "learning_rate": 1.3114285714285716e-06,
      "loss": 0.0,
      "step": 3550
    },
    {
      "epoch": 12.81,
      "learning_rate": 1.2400000000000002e-06,
      "loss": 0.0001,
      "step": 3575
    },
    {
      "epoch": 12.9,
      "learning_rate": 1.1685714285714286e-06,
      "loss": 0.0,
      "step": 3600
    },
    {
      "epoch": 12.99,
      "learning_rate": 1.0971428571428572e-06,
      "loss": 0.0,
      "step": 3625
    },
    {
      "epoch": 13.08,
      "learning_rate": 1.0257142857142858e-06,
      "loss": 0.0001,
      "step": 3650
    },
    {
      "epoch": 13.17,
      "learning_rate": 9.542857142857143e-07,
      "loss": 0.0,
      "step": 3675
    },
    {
      "epoch": 13.26,
      "learning_rate": 8.828571428571429e-07,
      "loss": 0.0001,
      "step": 3700
    },
    {
      "epoch": 13.35,
      "learning_rate": 8.114285714285715e-07,
      "loss": 0.0,
      "step": 3725
    },
    {
      "epoch": 13.44,
      "learning_rate": 7.4e-07,
      "loss": 0.0,
      "step": 3750
    },
    {
      "epoch": 13.53,
      "learning_rate": 6.685714285714286e-07,
      "loss": 0.0,
      "step": 3775
    },
    {
      "epoch": 13.62,
      "learning_rate": 5.971428571428572e-07,
      "loss": 0.0007,
      "step": 3800
    },
    {
      "epoch": 13.71,
      "learning_rate": 5.257142857142858e-07,
      "loss": 0.0,
      "step": 3825
    },
    {
      "epoch": 13.8,
      "learning_rate": 4.542857142857143e-07,
      "loss": 0.0001,
      "step": 3850
    },
    {
      "epoch": 13.89,
      "learning_rate": 3.8285714285714285e-07,
      "loss": 0.0,
      "step": 3875
    },
    {
      "epoch": 13.98,
      "learning_rate": 3.1142857142857144e-07,
      "loss": 0.0,
      "step": 3900
    },
    {
      "epoch": 14.07,
      "learning_rate": 2.4000000000000003e-07,
      "loss": 0.0,
      "step": 3925
    },
    {
      "epoch": 14.16,
      "learning_rate": 1.685714285714286e-07,
      "loss": 0.0,
      "step": 3950
    },
    {
      "epoch": 14.25,
      "learning_rate": 9.714285714285714e-08,
      "loss": 0.0,
      "step": 3975
    },
    {
      "epoch": 14.34,
      "learning_rate": 2.5714285714285715e-08,
      "loss": 0.0,
      "step": 4000
    },
    {
      "epoch": 14.34,
      "eval_loss": 0.2998046875,
      "eval_runtime": 641.2744,
      "eval_samples_per_second": 3.449,
      "eval_steps_per_second": 0.217,
      "eval_wer": 8.957818965817019,
      "step": 4000
    },
    {
      "epoch": 14.34,
      "step": 4000,
      "total_flos": 6.517583013067358e+19,
      "train_loss": 0.08977190920710564,
      "train_runtime": 17594.9662,
      "train_samples_per_second": 3.637,
      "train_steps_per_second": 0.227
    }
  ],
  "max_steps": 4000,
  "num_train_epochs": 15,
  "total_flos": 6.517583013067358e+19,
  "trial_name": null,
  "trial_params": null
}