{
  "best_metric": 0.3645675480365753,
  "best_model_checkpoint": "./speecht5_tts_lg/checkpoint-4000",
  "epoch": 59.25925925925926,
  "eval_steps": 1000,
  "global_step": 4000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.37,
      "learning_rate": 4.800000000000001e-07,
      "loss": 0.8112,
      "step": 25
    },
    {
      "epoch": 0.74,
      "learning_rate": 9.800000000000001e-07,
      "loss": 0.8001,
      "step": 50
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.48e-06,
      "loss": 0.7423,
      "step": 75
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.98e-06,
      "loss": 0.7274,
      "step": 100
    },
    {
      "epoch": 1.85,
      "learning_rate": 2.4800000000000004e-06,
      "loss": 0.687,
      "step": 125
    },
    {
      "epoch": 2.22,
      "learning_rate": 2.9800000000000003e-06,
      "loss": 0.6456,
      "step": 150
    },
    {
      "epoch": 2.59,
      "learning_rate": 3.48e-06,
      "loss": 0.656,
      "step": 175
    },
    {
      "epoch": 2.96,
      "learning_rate": 3.980000000000001e-06,
      "loss": 0.6285,
      "step": 200
    },
    {
      "epoch": 3.33,
      "learning_rate": 4.48e-06,
      "loss": 0.6037,
      "step": 225
    },
    {
      "epoch": 3.7,
      "learning_rate": 4.980000000000001e-06,
      "loss": 0.5912,
      "step": 250
    },
    {
      "epoch": 4.07,
      "learning_rate": 5.480000000000001e-06,
      "loss": 0.5756,
      "step": 275
    },
    {
      "epoch": 4.44,
      "learning_rate": 5.98e-06,
      "loss": 0.535,
      "step": 300
    },
    {
      "epoch": 4.81,
      "learning_rate": 6.480000000000001e-06,
      "loss": 0.5014,
      "step": 325
    },
    {
      "epoch": 5.19,
      "learning_rate": 6.98e-06,
      "loss": 0.4977,
      "step": 350
    },
    {
      "epoch": 5.56,
      "learning_rate": 7.48e-06,
      "loss": 0.4805,
      "step": 375
    },
    {
      "epoch": 5.93,
      "learning_rate": 7.980000000000002e-06,
      "loss": 0.4732,
      "step": 400
    },
    {
      "epoch": 6.3,
      "learning_rate": 8.48e-06,
      "loss": 0.4809,
      "step": 425
    },
    {
      "epoch": 6.67,
      "learning_rate": 8.98e-06,
      "loss": 0.4678,
      "step": 450
    },
    {
      "epoch": 7.04,
      "learning_rate": 9.48e-06,
      "loss": 0.4584,
      "step": 475
    },
    {
      "epoch": 7.41,
      "learning_rate": 9.980000000000001e-06,
      "loss": 0.458,
      "step": 500
    },
    {
      "epoch": 7.78,
      "learning_rate": 9.931428571428571e-06,
      "loss": 0.4526,
      "step": 525
    },
    {
      "epoch": 8.15,
      "learning_rate": 9.86e-06,
      "loss": 0.4509,
      "step": 550
    },
    {
      "epoch": 8.52,
      "learning_rate": 9.78857142857143e-06,
      "loss": 0.4437,
      "step": 575
    },
    {
      "epoch": 8.89,
      "learning_rate": 9.717142857142858e-06,
      "loss": 0.4523,
      "step": 600
    },
    {
      "epoch": 9.26,
      "learning_rate": 9.645714285714286e-06,
      "loss": 0.4467,
      "step": 625
    },
    {
      "epoch": 9.63,
      "learning_rate": 9.574285714285715e-06,
      "loss": 0.4478,
      "step": 650
    },
    {
      "epoch": 10.0,
      "learning_rate": 9.502857142857144e-06,
      "loss": 0.4455,
      "step": 675
    },
    {
      "epoch": 10.37,
      "learning_rate": 9.431428571428573e-06,
      "loss": 0.4349,
      "step": 700
    },
    {
      "epoch": 10.74,
      "learning_rate": 9.360000000000002e-06,
      "loss": 0.4336,
      "step": 725
    },
    {
      "epoch": 11.11,
      "learning_rate": 9.28857142857143e-06,
      "loss": 0.4311,
      "step": 750
    },
    {
      "epoch": 11.48,
      "learning_rate": 9.217142857142858e-06,
      "loss": 0.4314,
      "step": 775
    },
    {
      "epoch": 11.85,
      "learning_rate": 9.145714285714287e-06,
      "loss": 0.4291,
      "step": 800
    },
    {
      "epoch": 12.22,
      "learning_rate": 9.074285714285716e-06,
      "loss": 0.4344,
      "step": 825
    },
    {
      "epoch": 12.59,
      "learning_rate": 9.002857142857144e-06,
      "loss": 0.4276,
      "step": 850
    },
    {
      "epoch": 12.96,
      "learning_rate": 8.931428571428573e-06,
      "loss": 0.4191,
      "step": 875
    },
    {
      "epoch": 13.33,
      "learning_rate": 8.860000000000002e-06,
      "loss": 0.4271,
      "step": 900
    },
    {
      "epoch": 13.7,
      "learning_rate": 8.788571428571429e-06,
      "loss": 0.4217,
      "step": 925
    },
    {
      "epoch": 14.07,
      "learning_rate": 8.717142857142858e-06,
      "loss": 0.4205,
      "step": 950
    },
    {
      "epoch": 14.44,
      "learning_rate": 8.645714285714287e-06,
      "loss": 0.4191,
      "step": 975
    },
    {
      "epoch": 14.81,
      "learning_rate": 8.574285714285714e-06,
      "loss": 0.4213,
      "step": 1000
    },
    {
      "epoch": 14.81,
      "eval_loss": 0.38490030169487,
      "eval_runtime": 5.2563,
      "eval_samples_per_second": 45.66,
      "eval_steps_per_second": 1.902,
      "step": 1000
    },
    {
      "epoch": 15.19,
      "learning_rate": 8.502857142857143e-06,
      "loss": 0.415,
      "step": 1025
    },
    {
      "epoch": 15.56,
      "learning_rate": 8.431428571428572e-06,
      "loss": 0.4259,
      "step": 1050
    },
    {
      "epoch": 15.93,
      "learning_rate": 8.36e-06,
      "loss": 0.4189,
      "step": 1075
    },
    {
      "epoch": 16.3,
      "learning_rate": 8.288571428571429e-06,
      "loss": 0.4171,
      "step": 1100
    },
    {
      "epoch": 16.67,
      "learning_rate": 8.217142857142858e-06,
      "loss": 0.4111,
      "step": 1125
    },
    {
      "epoch": 17.04,
      "learning_rate": 8.145714285714287e-06,
      "loss": 0.415,
      "step": 1150
    },
    {
      "epoch": 17.41,
      "learning_rate": 8.074285714285714e-06,
      "loss": 0.4078,
      "step": 1175
    },
    {
      "epoch": 17.78,
      "learning_rate": 8.002857142857143e-06,
      "loss": 0.4099,
      "step": 1200
    },
    {
      "epoch": 18.15,
      "learning_rate": 7.931428571428572e-06,
      "loss": 0.4137,
      "step": 1225
    },
    {
      "epoch": 18.52,
      "learning_rate": 7.860000000000001e-06,
      "loss": 0.4137,
      "step": 1250
    },
    {
      "epoch": 18.89,
      "learning_rate": 7.788571428571428e-06,
      "loss": 0.4147,
      "step": 1275
    },
    {
      "epoch": 19.26,
      "learning_rate": 7.717142857142857e-06,
      "loss": 0.4111,
      "step": 1300
    },
    {
      "epoch": 19.63,
      "learning_rate": 7.645714285714286e-06,
      "loss": 0.4139,
      "step": 1325
    },
    {
      "epoch": 20.0,
      "learning_rate": 7.574285714285715e-06,
      "loss": 0.4104,
      "step": 1350
    },
    {
      "epoch": 20.37,
      "learning_rate": 7.502857142857144e-06,
      "loss": 0.4131,
      "step": 1375
    },
    {
      "epoch": 20.74,
      "learning_rate": 7.431428571428572e-06,
      "loss": 0.4001,
      "step": 1400
    },
    {
      "epoch": 21.11,
      "learning_rate": 7.360000000000001e-06,
      "loss": 0.4021,
      "step": 1425
    },
    {
      "epoch": 21.48,
      "learning_rate": 7.28857142857143e-06,
      "loss": 0.4009,
      "step": 1450
    },
    {
      "epoch": 21.85,
      "learning_rate": 7.217142857142858e-06,
      "loss": 0.4036,
      "step": 1475
    },
    {
      "epoch": 22.22,
      "learning_rate": 7.145714285714286e-06,
      "loss": 0.4068,
      "step": 1500
    },
    {
      "epoch": 22.59,
      "learning_rate": 7.074285714285715e-06,
      "loss": 0.4,
      "step": 1525
    },
    {
      "epoch": 22.96,
      "learning_rate": 7.002857142857143e-06,
      "loss": 0.3998,
      "step": 1550
    },
    {
      "epoch": 23.33,
      "learning_rate": 6.931428571428572e-06,
      "loss": 0.4017,
      "step": 1575
    },
    {
      "epoch": 23.7,
      "learning_rate": 6.860000000000001e-06,
      "loss": 0.3986,
      "step": 1600
    },
    {
      "epoch": 24.07,
      "learning_rate": 6.7885714285714286e-06,
      "loss": 0.404,
      "step": 1625
    },
    {
      "epoch": 24.44,
      "learning_rate": 6.7171428571428576e-06,
      "loss": 0.3927,
      "step": 1650
    },
    {
      "epoch": 24.81,
      "learning_rate": 6.645714285714287e-06,
      "loss": 0.4004,
      "step": 1675
    },
    {
      "epoch": 25.19,
      "learning_rate": 6.574285714285716e-06,
      "loss": 0.398,
      "step": 1700
    },
    {
      "epoch": 25.56,
      "learning_rate": 6.502857142857143e-06,
      "loss": 0.3982,
      "step": 1725
    },
    {
      "epoch": 25.93,
      "learning_rate": 6.431428571428572e-06,
      "loss": 0.3975,
      "step": 1750
    },
    {
      "epoch": 26.3,
      "learning_rate": 6.360000000000001e-06,
      "loss": 0.3952,
      "step": 1775
    },
    {
      "epoch": 26.67,
      "learning_rate": 6.288571428571429e-06,
      "loss": 0.3964,
      "step": 1800
    },
    {
      "epoch": 27.04,
      "learning_rate": 6.217142857142857e-06,
      "loss": 0.3983,
      "step": 1825
    },
    {
      "epoch": 27.41,
      "learning_rate": 6.145714285714286e-06,
      "loss": 0.3937,
      "step": 1850
    },
    {
      "epoch": 27.78,
      "learning_rate": 6.0742857142857145e-06,
      "loss": 0.398,
      "step": 1875
    },
    {
      "epoch": 28.15,
      "learning_rate": 6.0028571428571435e-06,
      "loss": 0.398,
      "step": 1900
    },
    {
      "epoch": 28.52,
      "learning_rate": 5.9314285714285725e-06,
      "loss": 0.3976,
      "step": 1925
    },
    {
      "epoch": 28.89,
      "learning_rate": 5.86e-06,
      "loss": 0.3945,
      "step": 1950
    },
    {
      "epoch": 29.26,
      "learning_rate": 5.791428571428572e-06,
      "loss": 0.3909,
      "step": 1975
    },
    {
      "epoch": 29.63,
      "learning_rate": 5.72e-06,
      "loss": 0.392,
      "step": 2000
    },
    {
      "epoch": 29.63,
      "eval_loss": 0.3671075701713562,
      "eval_runtime": 5.4439,
      "eval_samples_per_second": 44.086,
      "eval_steps_per_second": 1.837,
      "step": 2000
    },
    {
      "epoch": 30.0,
      "learning_rate": 5.6485714285714285e-06,
      "loss": 0.3933,
      "step": 2025
    },
    {
      "epoch": 30.37,
      "learning_rate": 5.5771428571428575e-06,
      "loss": 0.394,
      "step": 2050
    },
    {
      "epoch": 30.74,
      "learning_rate": 5.5057142857142865e-06,
      "loss": 0.3975,
      "step": 2075
    },
    {
      "epoch": 31.11,
      "learning_rate": 5.4342857142857155e-06,
      "loss": 0.3916,
      "step": 2100
    },
    {
      "epoch": 31.48,
      "learning_rate": 5.362857142857143e-06,
      "loss": 0.3991,
      "step": 2125
    },
    {
      "epoch": 31.85,
      "learning_rate": 5.291428571428572e-06,
      "loss": 0.3934,
      "step": 2150
    },
    {
      "epoch": 32.22,
      "learning_rate": 5.220000000000001e-06,
      "loss": 0.3944,
      "step": 2175
    },
    {
      "epoch": 32.59,
      "learning_rate": 5.14857142857143e-06,
      "loss": 0.3946,
      "step": 2200
    },
    {
      "epoch": 32.96,
      "learning_rate": 5.077142857142857e-06,
      "loss": 0.3907,
      "step": 2225
    },
    {
      "epoch": 33.33,
      "learning_rate": 5.005714285714286e-06,
      "loss": 0.3926,
      "step": 2250
    },
    {
      "epoch": 33.7,
      "learning_rate": 4.934285714285715e-06,
      "loss": 0.3954,
      "step": 2275
    },
    {
      "epoch": 34.07,
      "learning_rate": 4.862857142857143e-06,
      "loss": 0.3994,
      "step": 2300
    },
    {
      "epoch": 34.44,
      "learning_rate": 4.7914285714285715e-06,
      "loss": 0.3954,
      "step": 2325
    },
    {
      "epoch": 34.81,
      "learning_rate": 4.7200000000000005e-06,
      "loss": 0.3916,
      "step": 2350
    },
    {
      "epoch": 35.19,
      "learning_rate": 4.648571428571429e-06,
      "loss": 0.3904,
      "step": 2375
    },
    {
      "epoch": 35.56,
      "learning_rate": 4.577142857142858e-06,
      "loss": 0.3939,
      "step": 2400
    },
    {
      "epoch": 35.93,
      "learning_rate": 4.505714285714286e-06,
      "loss": 0.3867,
      "step": 2425
    },
    {
      "epoch": 36.3,
      "learning_rate": 4.434285714285715e-06,
      "loss": 0.3843,
      "step": 2450
    },
    {
      "epoch": 36.67,
      "learning_rate": 4.362857142857143e-06,
      "loss": 0.389,
      "step": 2475
    },
    {
      "epoch": 37.04,
      "learning_rate": 4.291428571428572e-06,
      "loss": 0.3915,
      "step": 2500
    },
    {
      "epoch": 37.41,
      "learning_rate": 4.22e-06,
      "loss": 0.3883,
      "step": 2525
    },
    {
      "epoch": 37.78,
      "learning_rate": 4.148571428571429e-06,
      "loss": 0.3886,
      "step": 2550
    },
    {
      "epoch": 38.15,
      "learning_rate": 4.0771428571428574e-06,
      "loss": 0.3916,
      "step": 2575
    },
    {
      "epoch": 38.52,
      "learning_rate": 4.0057142857142864e-06,
      "loss": 0.386,
      "step": 2600
    },
    {
      "epoch": 38.89,
      "learning_rate": 3.934285714285715e-06,
      "loss": 0.3877,
      "step": 2625
    },
    {
      "epoch": 39.26,
      "learning_rate": 3.862857142857143e-06,
      "loss": 0.388,
      "step": 2650
    },
    {
      "epoch": 39.63,
      "learning_rate": 3.7914285714285722e-06,
      "loss": 0.3855,
      "step": 2675
    },
    {
      "epoch": 40.0,
      "learning_rate": 3.7200000000000004e-06,
      "loss": 0.3846,
      "step": 2700
    },
    {
      "epoch": 40.37,
      "learning_rate": 3.648571428571429e-06,
      "loss": 0.3861,
      "step": 2725
    },
    {
      "epoch": 40.74,
      "learning_rate": 3.5771428571428576e-06,
      "loss": 0.3914,
      "step": 2750
    },
    {
      "epoch": 41.11,
      "learning_rate": 3.505714285714286e-06,
      "loss": 0.3856,
      "step": 2775
    },
    {
      "epoch": 41.48,
      "learning_rate": 3.4342857142857143e-06,
      "loss": 0.3893,
      "step": 2800
    },
    {
      "epoch": 41.85,
      "learning_rate": 3.3628571428571433e-06,
      "loss": 0.3883,
      "step": 2825
    },
    {
      "epoch": 42.22,
      "learning_rate": 3.2914285714285715e-06,
      "loss": 0.3854,
      "step": 2850
    },
    {
      "epoch": 42.59,
      "learning_rate": 3.2200000000000005e-06,
      "loss": 0.3873,
      "step": 2875
    },
    {
      "epoch": 42.96,
      "learning_rate": 3.1485714285714287e-06,
      "loss": 0.3871,
      "step": 2900
    },
    {
      "epoch": 43.33,
      "learning_rate": 3.0771428571428573e-06,
      "loss": 0.3859,
      "step": 2925
    },
    {
      "epoch": 43.7,
      "learning_rate": 3.005714285714286e-06,
      "loss": 0.3857,
      "step": 2950
    },
    {
      "epoch": 44.07,
      "learning_rate": 2.9342857142857144e-06,
      "loss": 0.3868,
      "step": 2975
    },
    {
      "epoch": 44.44,
      "learning_rate": 2.8628571428571435e-06,
      "loss": 0.3884,
      "step": 3000
    },
    {
      "epoch": 44.44,
      "eval_loss": 0.3664553463459015,
      "eval_runtime": 5.1439,
      "eval_samples_per_second": 46.657,
      "eval_steps_per_second": 1.944,
      "step": 3000
    },
    {
      "epoch": 44.81,
      "learning_rate": 2.7914285714285716e-06,
      "loss": 0.3842,
      "step": 3025
    },
    {
      "epoch": 45.19,
      "learning_rate": 2.7200000000000002e-06,
      "loss": 0.3859,
      "step": 3050
    },
    {
      "epoch": 45.56,
      "learning_rate": 2.648571428571429e-06,
      "loss": 0.3932,
      "step": 3075
    },
    {
      "epoch": 45.93,
      "learning_rate": 2.5771428571428574e-06,
      "loss": 0.3852,
      "step": 3100
    },
    {
      "epoch": 46.3,
      "learning_rate": 2.5057142857142856e-06,
      "loss": 0.3872,
      "step": 3125
    },
    {
      "epoch": 46.67,
      "learning_rate": 2.4342857142857146e-06,
      "loss": 0.3843,
      "step": 3150
    },
    {
      "epoch": 47.04,
      "learning_rate": 2.362857142857143e-06,
      "loss": 0.3927,
      "step": 3175
    },
    {
      "epoch": 47.41,
      "learning_rate": 2.2914285714285718e-06,
      "loss": 0.3871,
      "step": 3200
    },
    {
      "epoch": 47.78,
      "learning_rate": 2.2200000000000003e-06,
      "loss": 0.3819,
      "step": 3225
    },
    {
      "epoch": 48.15,
      "learning_rate": 2.148571428571429e-06,
      "loss": 0.3871,
      "step": 3250
    },
    {
      "epoch": 48.52,
      "learning_rate": 2.077142857142857e-06,
      "loss": 0.3851,
      "step": 3275
    },
    {
      "epoch": 48.89,
      "learning_rate": 2.0057142857142857e-06,
      "loss": 0.3841,
      "step": 3300
    },
    {
      "epoch": 49.26,
      "learning_rate": 1.9342857142857143e-06,
      "loss": 0.3814,
      "step": 3325
    },
    {
      "epoch": 49.63,
      "learning_rate": 1.8628571428571429e-06,
      "loss": 0.3846,
      "step": 3350
    },
    {
      "epoch": 50.0,
      "learning_rate": 1.7914285714285715e-06,
      "loss": 0.3811,
      "step": 3375
    },
    {
      "epoch": 50.37,
      "learning_rate": 1.72e-06,
      "loss": 0.3854,
      "step": 3400
    },
    {
      "epoch": 50.74,
      "learning_rate": 1.6485714285714289e-06,
      "loss": 0.3821,
      "step": 3425
    },
    {
      "epoch": 51.11,
      "learning_rate": 1.5771428571428574e-06,
      "loss": 0.3818,
      "step": 3450
    },
    {
      "epoch": 51.48,
      "learning_rate": 1.5057142857142858e-06,
      "loss": 0.3895,
      "step": 3475
    },
    {
      "epoch": 51.85,
      "learning_rate": 1.4342857142857144e-06,
      "loss": 0.3829,
      "step": 3500
    },
    {
      "epoch": 52.22,
      "learning_rate": 1.362857142857143e-06,
      "loss": 0.385,
      "step": 3525
    },
    {
      "epoch": 52.59,
      "learning_rate": 1.2914285714285716e-06,
      "loss": 0.3834,
      "step": 3550
    },
    {
      "epoch": 52.96,
      "learning_rate": 1.2200000000000002e-06,
      "loss": 0.3833,
      "step": 3575
    },
    {
      "epoch": 53.33,
      "learning_rate": 1.1485714285714286e-06,
      "loss": 0.3797,
      "step": 3600
    },
    {
      "epoch": 53.7,
      "learning_rate": 1.0771428571428574e-06,
      "loss": 0.3874,
      "step": 3625
    },
    {
      "epoch": 54.07,
      "learning_rate": 1.0057142857142857e-06,
      "loss": 0.3785,
      "step": 3650
    },
    {
      "epoch": 54.44,
      "learning_rate": 9.342857142857144e-07,
      "loss": 0.3792,
      "step": 3675
    },
    {
      "epoch": 54.81,
      "learning_rate": 8.628571428571429e-07,
      "loss": 0.3814,
      "step": 3700
    },
    {
      "epoch": 55.19,
      "learning_rate": 7.914285714285715e-07,
      "loss": 0.3815,
      "step": 3725
    },
    {
      "epoch": 55.56,
      "learning_rate": 7.2e-07,
      "loss": 0.3828,
      "step": 3750
    },
    {
      "epoch": 55.93,
      "learning_rate": 6.485714285714287e-07,
      "loss": 0.386,
      "step": 3775
    },
    {
      "epoch": 56.3,
      "learning_rate": 5.771428571428572e-07,
      "loss": 0.3846,
      "step": 3800
    },
    {
      "epoch": 56.67,
      "learning_rate": 5.057142857142858e-07,
      "loss": 0.3793,
      "step": 3825
    },
    {
      "epoch": 57.04,
      "learning_rate": 4.342857142857143e-07,
      "loss": 0.3814,
      "step": 3850
    },
    {
      "epoch": 57.41,
      "learning_rate": 3.6285714285714283e-07,
      "loss": 0.3808,
      "step": 3875
    },
    {
      "epoch": 57.78,
      "learning_rate": 2.914285714285715e-07,
      "loss": 0.3798,
      "step": 3900
    },
    {
      "epoch": 58.15,
      "learning_rate": 2.2e-07,
      "loss": 0.3852,
      "step": 3925
    },
    {
      "epoch": 58.52,
      "learning_rate": 1.4857142857142857e-07,
      "loss": 0.3812,
      "step": 3950
    },
    {
      "epoch": 58.89,
      "learning_rate": 7.714285714285715e-08,
      "loss": 0.3793,
      "step": 3975
    },
    {
      "epoch": 59.26,
      "learning_rate": 5.714285714285715e-09,
      "loss": 0.3879,
      "step": 4000
    },
    {
      "epoch": 59.26,
      "eval_loss": 0.3645675480365753,
      "eval_runtime": 6.1965,
      "eval_samples_per_second": 38.732,
      "eval_steps_per_second": 1.614,
      "step": 4000
    }
  ],
  "logging_steps": 25,
  "max_steps": 4000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 60,
  "save_steps": 1000,
  "total_flos": 1.0116465250588488e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}