{
  "best_metric": 11.37780883775938,
  "best_model_checkpoint": "./checkpoint-2000",
  "epoch": 7.1152,
  "global_step": 2500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 9.200000000000001e-07,
      "loss": 0.5204,
      "step": 25
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.9200000000000003e-06,
      "loss": 0.0582,
      "step": 50
    },
    {
      "epoch": 0.03,
      "learning_rate": 2.92e-06,
      "loss": 0.0527,
      "step": 75
    },
    {
      "epoch": 0.04,
      "learning_rate": 3.920000000000001e-06,
      "loss": 0.0587,
      "step": 100
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.92e-06,
      "loss": 0.0659,
      "step": 125
    },
    {
      "epoch": 0.06,
      "learning_rate": 5.92e-06,
      "loss": 0.0681,
      "step": 150
    },
    {
      "epoch": 0.07,
      "learning_rate": 6.92e-06,
      "loss": 0.07,
      "step": 175
    },
    {
      "epoch": 0.08,
      "learning_rate": 7.92e-06,
      "loss": 0.0856,
      "step": 200
    },
    {
      "epoch": 0.09,
      "learning_rate": 8.920000000000001e-06,
      "loss": 0.0894,
      "step": 225
    },
    {
      "epoch": 0.1,
      "learning_rate": 9.920000000000002e-06,
      "loss": 0.0751,
      "step": 250
    },
    {
      "epoch": 0.11,
      "learning_rate": 9.89777777777778e-06,
      "loss": 0.0772,
      "step": 275
    },
    {
      "epoch": 0.12,
      "learning_rate": 9.786666666666667e-06,
      "loss": 0.0698,
      "step": 300
    },
    {
      "epoch": 1.0,
      "learning_rate": 9.675555555555555e-06,
      "loss": 0.0893,
      "step": 325
    },
    {
      "epoch": 1.01,
      "learning_rate": 9.564444444444445e-06,
      "loss": 0.0782,
      "step": 350
    },
    {
      "epoch": 1.02,
      "learning_rate": 9.453333333333335e-06,
      "loss": 0.0595,
      "step": 375
    },
    {
      "epoch": 1.03,
      "learning_rate": 9.342222222222223e-06,
      "loss": 0.0537,
      "step": 400
    },
    {
      "epoch": 1.04,
      "learning_rate": 9.231111111111111e-06,
      "loss": 0.052,
      "step": 425
    },
    {
      "epoch": 1.05,
      "learning_rate": 9.12e-06,
      "loss": 0.05,
      "step": 450
    },
    {
      "epoch": 1.06,
      "learning_rate": 9.008888888888889e-06,
      "loss": 0.0522,
      "step": 475
    },
    {
      "epoch": 1.07,
      "learning_rate": 8.897777777777779e-06,
      "loss": 0.0467,
      "step": 500
    },
    {
      "epoch": 1.08,
      "learning_rate": 8.786666666666668e-06,
      "loss": 0.0586,
      "step": 525
    },
    {
      "epoch": 1.09,
      "learning_rate": 8.675555555555556e-06,
      "loss": 0.0437,
      "step": 550
    },
    {
      "epoch": 1.1,
      "learning_rate": 8.564444444444445e-06,
      "loss": 0.0395,
      "step": 575
    },
    {
      "epoch": 1.11,
      "learning_rate": 8.453333333333334e-06,
      "loss": 0.0371,
      "step": 600
    },
    {
      "epoch": 1.12,
      "learning_rate": 8.342222222222222e-06,
      "loss": 0.0316,
      "step": 625
    },
    {
      "epoch": 2.01,
      "learning_rate": 8.231111111111112e-06,
      "loss": 0.0412,
      "step": 650
    },
    {
      "epoch": 2.02,
      "learning_rate": 8.120000000000002e-06,
      "loss": 0.0335,
      "step": 675
    },
    {
      "epoch": 2.03,
      "learning_rate": 8.00888888888889e-06,
      "loss": 0.0314,
      "step": 700
    },
    {
      "epoch": 2.04,
      "learning_rate": 7.897777777777778e-06,
      "loss": 0.0329,
      "step": 725
    },
    {
      "epoch": 2.05,
      "learning_rate": 7.786666666666666e-06,
      "loss": 0.0275,
      "step": 750
    },
    {
      "epoch": 2.06,
      "learning_rate": 7.675555555555556e-06,
      "loss": 0.028,
      "step": 775
    },
    {
      "epoch": 2.07,
      "learning_rate": 7.564444444444446e-06,
      "loss": 0.0256,
      "step": 800
    },
    {
      "epoch": 2.08,
      "learning_rate": 7.453333333333334e-06,
      "loss": 0.026,
      "step": 825
    },
    {
      "epoch": 2.09,
      "learning_rate": 7.342222222222223e-06,
      "loss": 0.0295,
      "step": 850
    },
    {
      "epoch": 2.1,
      "learning_rate": 7.231111111111112e-06,
      "loss": 0.0259,
      "step": 875
    },
    {
      "epoch": 2.11,
      "learning_rate": 7.1200000000000004e-06,
      "loss": 0.0173,
      "step": 900
    },
    {
      "epoch": 2.12,
      "learning_rate": 7.008888888888889e-06,
      "loss": 0.014,
      "step": 925
    },
    {
      "epoch": 3.0,
      "learning_rate": 6.897777777777779e-06,
      "loss": 0.0158,
      "step": 950
    },
    {
      "epoch": 3.01,
      "learning_rate": 6.786666666666667e-06,
      "loss": 0.0193,
      "step": 975
    },
    {
      "epoch": 3.02,
      "learning_rate": 6.675555555555556e-06,
      "loss": 0.0146,
      "step": 1000
    },
    {
      "epoch": 3.02,
      "eval_loss": 0.2546452581882477,
      "eval_runtime": 2882.0567,
      "eval_samples_per_second": 1.759,
      "eval_steps_per_second": 0.11,
      "eval_wer": 11.942264272658853,
      "step": 1000
    },
    {
      "epoch": 3.03,
      "learning_rate": 6.564444444444446e-06,
      "loss": 0.0129,
      "step": 1025
    },
    {
      "epoch": 3.04,
      "learning_rate": 6.453333333333334e-06,
      "loss": 0.0111,
      "step": 1050
    },
    {
      "epoch": 3.05,
      "learning_rate": 6.342222222222223e-06,
      "loss": 0.0143,
      "step": 1075
    },
    {
      "epoch": 3.06,
      "learning_rate": 6.231111111111111e-06,
      "loss": 0.0128,
      "step": 1100
    },
    {
      "epoch": 3.07,
      "learning_rate": 6.120000000000001e-06,
      "loss": 0.0105,
      "step": 1125
    },
    {
      "epoch": 3.08,
      "learning_rate": 6.00888888888889e-06,
      "loss": 0.012,
      "step": 1150
    },
    {
      "epoch": 3.09,
      "learning_rate": 5.897777777777778e-06,
      "loss": 0.0133,
      "step": 1175
    },
    {
      "epoch": 3.1,
      "learning_rate": 5.7866666666666674e-06,
      "loss": 0.0093,
      "step": 1200
    },
    {
      "epoch": 3.11,
      "learning_rate": 5.675555555555556e-06,
      "loss": 0.0097,
      "step": 1225
    },
    {
      "epoch": 3.12,
      "learning_rate": 5.5644444444444444e-06,
      "loss": 0.007,
      "step": 1250
    },
    {
      "epoch": 4.0,
      "learning_rate": 5.453333333333334e-06,
      "loss": 0.0059,
      "step": 1275
    },
    {
      "epoch": 4.01,
      "learning_rate": 5.342222222222223e-06,
      "loss": 0.01,
      "step": 1300
    },
    {
      "epoch": 4.02,
      "learning_rate": 5.231111111111111e-06,
      "loss": 0.0086,
      "step": 1325
    },
    {
      "epoch": 4.03,
      "learning_rate": 5.12e-06,
      "loss": 0.0048,
      "step": 1350
    },
    {
      "epoch": 4.04,
      "learning_rate": 5.00888888888889e-06,
      "loss": 0.0058,
      "step": 1375
    },
    {
      "epoch": 4.05,
      "learning_rate": 4.897777777777778e-06,
      "loss": 0.0056,
      "step": 1400
    },
    {
      "epoch": 4.06,
      "learning_rate": 4.786666666666667e-06,
      "loss": 0.0059,
      "step": 1425
    },
    {
      "epoch": 4.07,
      "learning_rate": 4.675555555555556e-06,
      "loss": 0.0038,
      "step": 1450
    },
    {
      "epoch": 4.08,
      "learning_rate": 4.564444444444445e-06,
      "loss": 0.0049,
      "step": 1475
    },
    {
      "epoch": 4.09,
      "learning_rate": 4.453333333333334e-06,
      "loss": 0.0051,
      "step": 1500
    },
    {
      "epoch": 4.1,
      "learning_rate": 4.3422222222222225e-06,
      "loss": 0.0044,
      "step": 1525
    },
    {
      "epoch": 4.11,
      "learning_rate": 4.2311111111111114e-06,
      "loss": 0.0026,
      "step": 1550
    },
    {
      "epoch": 4.12,
      "learning_rate": 4.12e-06,
      "loss": 0.0036,
      "step": 1575
    },
    {
      "epoch": 5.01,
      "learning_rate": 4.008888888888889e-06,
      "loss": 0.0035,
      "step": 1600
    },
    {
      "epoch": 5.02,
      "learning_rate": 3.897777777777778e-06,
      "loss": 0.0024,
      "step": 1625
    },
    {
      "epoch": 5.03,
      "learning_rate": 3.7866666666666667e-06,
      "loss": 0.0026,
      "step": 1650
    },
    {
      "epoch": 5.04,
      "learning_rate": 3.675555555555556e-06,
      "loss": 0.0031,
      "step": 1675
    },
    {
      "epoch": 5.05,
      "learning_rate": 3.564444444444445e-06,
      "loss": 0.0045,
      "step": 1700
    },
    {
      "epoch": 5.06,
      "learning_rate": 3.4533333333333334e-06,
      "loss": 0.0019,
      "step": 1725
    },
    {
      "epoch": 5.07,
      "learning_rate": 3.3422222222222224e-06,
      "loss": 0.0018,
      "step": 1750
    },
    {
      "epoch": 5.08,
      "learning_rate": 3.2311111111111117e-06,
      "loss": 0.0033,
      "step": 1775
    },
    {
      "epoch": 5.09,
      "learning_rate": 3.12e-06,
      "loss": 0.0048,
      "step": 1800
    },
    {
      "epoch": 5.1,
      "learning_rate": 3.008888888888889e-06,
      "loss": 0.0019,
      "step": 1825
    },
    {
      "epoch": 5.11,
      "learning_rate": 2.8977777777777785e-06,
      "loss": 0.0015,
      "step": 1850
    },
    {
      "epoch": 5.12,
      "learning_rate": 2.786666666666667e-06,
      "loss": 0.0018,
      "step": 1875
    },
    {
      "epoch": 6.0,
      "learning_rate": 2.675555555555556e-06,
      "loss": 0.0017,
      "step": 1900
    },
    {
      "epoch": 6.01,
      "learning_rate": 2.5644444444444444e-06,
      "loss": 0.0017,
      "step": 1925
    },
    {
      "epoch": 6.02,
      "learning_rate": 2.4533333333333333e-06,
      "loss": 0.0009,
      "step": 1950
    },
    {
      "epoch": 6.03,
      "learning_rate": 2.342222222222222e-06,
      "loss": 0.0011,
      "step": 1975
    },
    {
      "epoch": 6.04,
      "learning_rate": 2.2311111111111115e-06,
      "loss": 0.0017,
      "step": 2000
    },
    {
      "epoch": 6.04,
      "eval_loss": 0.29696929454803467,
      "eval_runtime": 2899.6818,
      "eval_samples_per_second": 1.748,
      "eval_steps_per_second": 0.109,
      "eval_wer": 11.37780883775938,
      "step": 2000
    },
    {
      "epoch": 6.05,
      "learning_rate": 2.12e-06,
      "loss": 0.0013,
      "step": 2025
    },
    {
      "epoch": 6.06,
      "learning_rate": 2.008888888888889e-06,
      "loss": 0.0014,
      "step": 2050
    },
    {
      "epoch": 6.07,
      "learning_rate": 1.8977777777777779e-06,
      "loss": 0.0008,
      "step": 2075
    },
    {
      "epoch": 6.08,
      "learning_rate": 1.7866666666666668e-06,
      "loss": 0.001,
      "step": 2100
    },
    {
      "epoch": 6.09,
      "learning_rate": 1.675555555555556e-06,
      "loss": 0.0009,
      "step": 2125
    },
    {
      "epoch": 6.1,
      "learning_rate": 1.5644444444444446e-06,
      "loss": 0.0006,
      "step": 2150
    },
    {
      "epoch": 6.11,
      "learning_rate": 1.4533333333333335e-06,
      "loss": 0.0011,
      "step": 2175
    },
    {
      "epoch": 6.12,
      "learning_rate": 1.3422222222222222e-06,
      "loss": 0.0006,
      "step": 2200
    },
    {
      "epoch": 7.01,
      "learning_rate": 1.2311111111111112e-06,
      "loss": 0.0006,
      "step": 2225
    },
    {
      "epoch": 7.02,
      "learning_rate": 1.12e-06,
      "loss": 0.0006,
      "step": 2250
    },
    {
      "epoch": 7.03,
      "learning_rate": 1.008888888888889e-06,
      "loss": 0.0006,
      "step": 2275
    },
    {
      "epoch": 7.04,
      "learning_rate": 8.977777777777778e-07,
      "loss": 0.0007,
      "step": 2300
    },
    {
      "epoch": 7.05,
      "learning_rate": 7.866666666666667e-07,
      "loss": 0.0009,
      "step": 2325
    },
    {
      "epoch": 7.06,
      "learning_rate": 6.755555555555555e-07,
      "loss": 0.0007,
      "step": 2350
    },
    {
      "epoch": 7.07,
      "learning_rate": 5.644444444444445e-07,
      "loss": 0.0007,
      "step": 2375
    },
    {
      "epoch": 7.08,
      "learning_rate": 4.533333333333334e-07,
      "loss": 0.0006,
      "step": 2400
    },
    {
      "epoch": 7.09,
      "learning_rate": 3.422222222222223e-07,
      "loss": 0.0007,
      "step": 2425
    },
    {
      "epoch": 7.1,
      "learning_rate": 2.3111111111111112e-07,
      "loss": 0.0006,
      "step": 2450
    },
    {
      "epoch": 7.11,
      "learning_rate": 1.2000000000000002e-07,
      "loss": 0.0006,
      "step": 2475
    },
    {
      "epoch": 7.12,
      "learning_rate": 8.88888888888889e-09,
      "loss": 0.0006,
      "step": 2500
    },
    {
      "epoch": 7.12,
      "step": 2500,
      "total_flos": 8.159123872677888e+19,
      "train_loss": 0.026056346493959427,
      "train_runtime": 20075.1792,
      "train_samples_per_second": 3.985,
      "train_steps_per_second": 0.125
    }
  ],
  "max_steps": 2500,
  "num_train_epochs": 9223372036854775807,
  "total_flos": 8.159123872677888e+19,
  "trial_name": null,
  "trial_params": null
}