{
  "best_metric": 105.08474576271188,
  "best_model_checkpoint": "./medium_TH/checkpoint-1000",
  "epoch": 55.55555555555556,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.39,
      "learning_rate": 4.2000000000000006e-07,
      "loss": 1.7782,
      "step": 25
    },
    {
      "epoch": 2.78,
      "learning_rate": 9.200000000000001e-07,
      "loss": 1.259,
      "step": 50
    },
    {
      "epoch": 4.17,
      "learning_rate": 1.42e-06,
      "loss": 0.8323,
      "step": 75
    },
    {
      "epoch": 5.56,
      "learning_rate": 1.9200000000000003e-06,
      "loss": 0.4965,
      "step": 100
    },
    {
      "epoch": 6.94,
      "learning_rate": 2.42e-06,
      "loss": 0.3197,
      "step": 125
    },
    {
      "epoch": 8.33,
      "learning_rate": 2.92e-06,
      "loss": 0.2263,
      "step": 150
    },
    {
      "epoch": 9.72,
      "learning_rate": 3.4200000000000007e-06,
      "loss": 0.1433,
      "step": 175
    },
    {
      "epoch": 11.11,
      "learning_rate": 3.920000000000001e-06,
      "loss": 0.0249,
      "step": 200
    },
    {
      "epoch": 12.5,
      "learning_rate": 4.42e-06,
      "loss": 0.0112,
      "step": 225
    },
    {
      "epoch": 13.89,
      "learning_rate": 4.92e-06,
      "loss": 0.0065,
      "step": 250
    },
    {
      "epoch": 15.28,
      "learning_rate": 5.420000000000001e-06,
      "loss": 0.0129,
      "step": 275
    },
    {
      "epoch": 16.67,
      "learning_rate": 5.92e-06,
      "loss": 0.0097,
      "step": 300
    },
    {
      "epoch": 18.06,
      "learning_rate": 6.42e-06,
      "loss": 0.0102,
      "step": 325
    },
    {
      "epoch": 19.44,
      "learning_rate": 6.92e-06,
      "loss": 0.0143,
      "step": 350
    },
    {
      "epoch": 20.83,
      "learning_rate": 7.420000000000001e-06,
      "loss": 0.0108,
      "step": 375
    },
    {
      "epoch": 22.22,
      "learning_rate": 7.92e-06,
      "loss": 0.0131,
      "step": 400
    },
    {
      "epoch": 23.61,
      "learning_rate": 8.42e-06,
      "loss": 0.0193,
      "step": 425
    },
    {
      "epoch": 25.0,
      "learning_rate": 8.920000000000001e-06,
      "loss": 0.012,
      "step": 450
    },
    {
      "epoch": 26.39,
      "learning_rate": 9.42e-06,
      "loss": 0.0084,
      "step": 475
    },
    {
      "epoch": 27.78,
      "learning_rate": 9.920000000000002e-06,
      "loss": 0.028,
      "step": 500
    },
    {
      "epoch": 29.17,
      "learning_rate": 9.940000000000001e-06,
      "loss": 0.0349,
      "step": 525
    },
    {
      "epoch": 30.56,
      "learning_rate": 9.86857142857143e-06,
      "loss": 0.0183,
      "step": 550
    },
    {
      "epoch": 31.94,
      "learning_rate": 9.797142857142858e-06,
      "loss": 0.0146,
      "step": 575
    },
    {
      "epoch": 33.33,
      "learning_rate": 9.725714285714287e-06,
      "loss": 0.0111,
      "step": 600
    },
    {
      "epoch": 34.72,
      "learning_rate": 9.654285714285716e-06,
      "loss": 0.0045,
      "step": 625
    },
    {
      "epoch": 36.11,
      "learning_rate": 9.582857142857143e-06,
      "loss": 0.0041,
      "step": 650
    },
    {
      "epoch": 37.5,
      "learning_rate": 9.511428571428572e-06,
      "loss": 0.01,
      "step": 675
    },
    {
      "epoch": 38.89,
      "learning_rate": 9.440000000000001e-06,
      "loss": 0.0091,
      "step": 700
    },
    {
      "epoch": 40.28,
      "learning_rate": 9.368571428571428e-06,
      "loss": 0.0102,
      "step": 725
    },
    {
      "epoch": 41.67,
      "learning_rate": 9.297142857142857e-06,
      "loss": 0.0064,
      "step": 750
    },
    {
      "epoch": 43.06,
      "learning_rate": 9.225714285714286e-06,
      "loss": 0.0069,
      "step": 775
    },
    {
      "epoch": 44.44,
      "learning_rate": 9.154285714285715e-06,
      "loss": 0.0053,
      "step": 800
    },
    {
      "epoch": 45.83,
      "learning_rate": 9.082857142857143e-06,
      "loss": 0.0053,
      "step": 825
    },
    {
      "epoch": 47.22,
      "learning_rate": 9.011428571428572e-06,
      "loss": 0.0032,
      "step": 850
    },
    {
      "epoch": 48.61,
      "learning_rate": 8.94e-06,
      "loss": 0.004,
      "step": 875
    },
    {
      "epoch": 50.0,
      "learning_rate": 8.86857142857143e-06,
      "loss": 0.0021,
      "step": 900
    },
    {
      "epoch": 51.39,
      "learning_rate": 8.797142857142857e-06,
      "loss": 0.0016,
      "step": 925
    },
    {
      "epoch": 52.78,
      "learning_rate": 8.725714285714286e-06,
      "loss": 0.0001,
      "step": 950
    },
    {
      "epoch": 54.17,
      "learning_rate": 8.654285714285715e-06,
      "loss": 0.0,
      "step": 975
    },
    {
      "epoch": 55.56,
      "learning_rate": 8.582857142857144e-06,
      "loss": 0.0,
      "step": 1000
    },
    {
      "epoch": 55.56,
      "eval_loss": 5.337996482849121,
      "eval_runtime": 8.4203,
      "eval_samples_per_second": 3.563,
      "eval_steps_per_second": 0.475,
      "eval_wer": 105.08474576271188,
      "step": 1000
    }
  ],
  "max_steps": 4000,
  "num_train_epochs": 223,
  "total_flos": 3.9701530902528e+18,
  "trial_name": null,
  "trial_params": null
}