{
  "best_metric": 102.0,
  "best_model_checkpoint": "./model_005/checkpoint-2000",
  "epoch": 100.0,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.25,
      "learning_rate": 4.4e-08,
      "loss": 0.8082,
      "step": 25
    },
    {
      "epoch": 2.5,
      "learning_rate": 9.4e-08,
      "loss": 0.8811,
      "step": 50
    },
    {
      "epoch": 3.75,
      "learning_rate": 1.44e-07,
      "loss": 0.7267,
      "step": 75
    },
    {
      "epoch": 5.0,
      "learning_rate": 1.94e-07,
      "loss": 0.7555,
      "step": 100
    },
    {
      "epoch": 6.25,
      "learning_rate": 2.4399999999999996e-07,
      "loss": 0.6289,
      "step": 125
    },
    {
      "epoch": 7.5,
      "learning_rate": 2.9399999999999996e-07,
      "loss": 0.6541,
      "step": 150
    },
    {
      "epoch": 8.75,
      "learning_rate": 3.4399999999999996e-07,
      "loss": 0.6237,
      "step": 175
    },
    {
      "epoch": 10.0,
      "learning_rate": 3.94e-07,
      "loss": 0.6355,
      "step": 200
    },
    {
      "epoch": 11.25,
      "learning_rate": 4.44e-07,
      "loss": 0.5445,
      "step": 225
    },
    {
      "epoch": 12.5,
      "learning_rate": 4.94e-07,
      "loss": 0.5228,
      "step": 250
    },
    {
      "epoch": 13.75,
      "learning_rate": 5.44e-07,
      "loss": 0.6126,
      "step": 275
    },
    {
      "epoch": 15.0,
      "learning_rate": 5.939999999999999e-07,
      "loss": 0.5067,
      "step": 300
    },
    {
      "epoch": 16.25,
      "learning_rate": 6.44e-07,
      "loss": 0.4905,
      "step": 325
    },
    {
      "epoch": 17.5,
      "learning_rate": 6.939999999999999e-07,
      "loss": 0.46,
      "step": 350
    },
    {
      "epoch": 18.75,
      "learning_rate": 7.44e-07,
      "loss": 0.4899,
      "step": 375
    },
    {
      "epoch": 20.0,
      "learning_rate": 7.94e-07,
      "loss": 0.4079,
      "step": 400
    },
    {
      "epoch": 21.25,
      "learning_rate": 8.439999999999999e-07,
      "loss": 0.4024,
      "step": 425
    },
    {
      "epoch": 22.5,
      "learning_rate": 8.939999999999999e-07,
      "loss": 0.3725,
      "step": 450
    },
    {
      "epoch": 23.75,
      "learning_rate": 9.439999999999999e-07,
      "loss": 0.3226,
      "step": 475
    },
    {
      "epoch": 25.0,
      "learning_rate": 9.94e-07,
      "loss": 0.3195,
      "step": 500
    },
    {
      "epoch": 26.25,
      "learning_rate": 9.959999999999999e-07,
      "loss": 0.3305,
      "step": 525
    },
    {
      "epoch": 27.5,
      "learning_rate": 9.914545454545455e-07,
      "loss": 0.2994,
      "step": 550
    },
    {
      "epoch": 28.75,
      "learning_rate": 9.86909090909091e-07,
      "loss": 0.2754,
      "step": 575
    },
    {
      "epoch": 30.0,
      "learning_rate": 9.823636363636362e-07,
      "loss": 0.2541,
      "step": 600
    },
    {
      "epoch": 31.25,
      "learning_rate": 9.778181818181817e-07,
      "loss": 0.2476,
      "step": 625
    },
    {
      "epoch": 32.5,
      "learning_rate": 9.732727272727272e-07,
      "loss": 0.1813,
      "step": 650
    },
    {
      "epoch": 33.75,
      "learning_rate": 9.687272727272726e-07,
      "loss": 0.1728,
      "step": 675
    },
    {
      "epoch": 35.0,
      "learning_rate": 9.64181818181818e-07,
      "loss": 0.2009,
      "step": 700
    },
    {
      "epoch": 36.25,
      "learning_rate": 9.596363636363635e-07,
      "loss": 0.1723,
      "step": 725
    },
    {
      "epoch": 37.5,
      "learning_rate": 9.55090909090909e-07,
      "loss": 0.1857,
      "step": 750
    },
    {
      "epoch": 38.75,
      "learning_rate": 9.505454545454546e-07,
      "loss": 0.1074,
      "step": 775
    },
    {
      "epoch": 40.0,
      "learning_rate": 9.459999999999999e-07,
      "loss": 0.1422,
      "step": 800
    },
    {
      "epoch": 41.25,
      "learning_rate": 9.414545454545454e-07,
      "loss": 0.1562,
      "step": 825
    },
    {
      "epoch": 42.5,
      "learning_rate": 9.369090909090908e-07,
      "loss": 0.0662,
      "step": 850
    },
    {
      "epoch": 43.75,
      "learning_rate": 9.323636363636363e-07,
      "loss": 0.0854,
      "step": 875
    },
    {
      "epoch": 45.0,
      "learning_rate": 9.278181818181818e-07,
      "loss": 0.1199,
      "step": 900
    },
    {
      "epoch": 46.25,
      "learning_rate": 9.232727272727272e-07,
      "loss": 0.0943,
      "step": 925
    },
    {
      "epoch": 47.5,
      "learning_rate": 9.187272727272727e-07,
      "loss": 0.0702,
      "step": 950
    },
    {
      "epoch": 48.75,
      "learning_rate": 9.141818181818181e-07,
      "loss": 0.0529,
      "step": 975
    },
    {
      "epoch": 50.0,
      "learning_rate": 9.096363636363636e-07,
      "loss": 0.1109,
      "step": 1000
    },
    {
      "epoch": 50.0,
      "eval_loss": 0.2612619996070862,
      "eval_runtime": 10.5374,
      "eval_samples_per_second": 2.183,
      "eval_steps_per_second": 0.285,
      "eval_wer": 103.49999999999999,
      "step": 1000
    },
    {
      "epoch": 51.25,
      "learning_rate": 9.050909090909091e-07,
      "loss": 0.1209,
      "step": 1025
    },
    {
      "epoch": 52.5,
      "learning_rate": 9.005454545454545e-07,
      "loss": 0.0639,
      "step": 1050
    },
    {
      "epoch": 53.75,
      "learning_rate": 8.96e-07,
      "loss": 0.0627,
      "step": 1075
    },
    {
      "epoch": 55.0,
      "learning_rate": 8.914545454545454e-07,
      "loss": 0.0265,
      "step": 1100
    },
    {
      "epoch": 56.25,
      "learning_rate": 8.869090909090908e-07,
      "loss": 0.0695,
      "step": 1125
    },
    {
      "epoch": 57.5,
      "learning_rate": 8.823636363636364e-07,
      "loss": 0.0396,
      "step": 1150
    },
    {
      "epoch": 58.75,
      "learning_rate": 8.778181818181818e-07,
      "loss": 0.0206,
      "step": 1175
    },
    {
      "epoch": 60.0,
      "learning_rate": 8.732727272727273e-07,
      "loss": 0.0479,
      "step": 1200
    },
    {
      "epoch": 61.25,
      "learning_rate": 8.687272727272726e-07,
      "loss": 0.0663,
      "step": 1225
    },
    {
      "epoch": 62.5,
      "learning_rate": 8.641818181818181e-07,
      "loss": 0.0417,
      "step": 1250
    },
    {
      "epoch": 63.75,
      "learning_rate": 8.596363636363637e-07,
      "loss": 0.0286,
      "step": 1275
    },
    {
      "epoch": 65.0,
      "learning_rate": 8.550909090909091e-07,
      "loss": 0.0902,
      "step": 1300
    },
    {
      "epoch": 66.25,
      "learning_rate": 8.505454545454545e-07,
      "loss": 0.1234,
      "step": 1325
    },
    {
      "epoch": 67.5,
      "learning_rate": 8.459999999999999e-07,
      "loss": 0.0291,
      "step": 1350
    },
    {
      "epoch": 68.75,
      "learning_rate": 8.414545454545454e-07,
      "loss": 0.042,
      "step": 1375
    },
    {
      "epoch": 70.0,
      "learning_rate": 8.36909090909091e-07,
      "loss": 0.0828,
      "step": 1400
    },
    {
      "epoch": 71.25,
      "learning_rate": 8.323636363636364e-07,
      "loss": 0.0753,
      "step": 1425
    },
    {
      "epoch": 72.5,
      "learning_rate": 8.278181818181818e-07,
      "loss": 0.0356,
      "step": 1450
    },
    {
      "epoch": 73.75,
      "learning_rate": 8.232727272727272e-07,
      "loss": 0.0671,
      "step": 1475
    },
    {
      "epoch": 75.0,
      "learning_rate": 8.187272727272727e-07,
      "loss": 0.0507,
      "step": 1500
    },
    {
      "epoch": 76.25,
      "learning_rate": 8.141818181818181e-07,
      "loss": 0.0764,
      "step": 1525
    },
    {
      "epoch": 77.5,
      "learning_rate": 8.096363636363636e-07,
      "loss": 0.0204,
      "step": 1550
    },
    {
      "epoch": 78.75,
      "learning_rate": 8.050909090909091e-07,
      "loss": 0.053,
      "step": 1575
    },
    {
      "epoch": 80.0,
      "learning_rate": 8.005454545454545e-07,
      "loss": 0.0317,
      "step": 1600
    },
    {
      "epoch": 81.25,
      "learning_rate": 7.96e-07,
      "loss": 0.1174,
      "step": 1625
    },
    {
      "epoch": 82.5,
      "learning_rate": 7.914545454545453e-07,
      "loss": 0.0212,
      "step": 1650
    },
    {
      "epoch": 83.75,
      "learning_rate": 7.869090909090909e-07,
      "loss": 0.0286,
      "step": 1675
    },
    {
      "epoch": 85.0,
      "learning_rate": 7.823636363636364e-07,
      "loss": 0.0235,
      "step": 1700
    },
    {
      "epoch": 86.25,
      "learning_rate": 7.778181818181818e-07,
      "loss": 0.0648,
      "step": 1725
    },
    {
      "epoch": 87.5,
      "learning_rate": 7.732727272727272e-07,
      "loss": 0.0332,
      "step": 1750
    },
    {
      "epoch": 88.75,
      "learning_rate": 7.687272727272726e-07,
      "loss": 0.045,
      "step": 1775
    },
    {
      "epoch": 90.0,
      "learning_rate": 7.641818181818182e-07,
      "loss": 0.0575,
      "step": 1800
    },
    {
      "epoch": 91.25,
      "learning_rate": 7.596363636363637e-07,
      "loss": 0.0383,
      "step": 1825
    },
    {
      "epoch": 92.5,
      "learning_rate": 7.550909090909091e-07,
      "loss": 0.0255,
      "step": 1850
    },
    {
      "epoch": 93.75,
      "learning_rate": 7.505454545454545e-07,
      "loss": 0.0287,
      "step": 1875
    },
    {
      "epoch": 95.0,
      "learning_rate": 7.459999999999999e-07,
      "loss": 0.0351,
      "step": 1900
    },
    {
      "epoch": 96.25,
      "learning_rate": 7.414545454545455e-07,
      "loss": 0.0593,
      "step": 1925
    },
    {
      "epoch": 97.5,
      "learning_rate": 7.36909090909091e-07,
      "loss": 0.0392,
      "step": 1950
    },
    {
      "epoch": 98.75,
      "learning_rate": 7.323636363636363e-07,
      "loss": 0.0206,
      "step": 1975
    },
    {
      "epoch": 100.0,
      "learning_rate": 7.278181818181818e-07,
      "loss": 0.0222,
      "step": 2000
    },
    {
      "epoch": 100.0,
      "eval_loss": 0.29138991236686707,
      "eval_runtime": 10.4785,
      "eval_samples_per_second": 2.195,
      "eval_steps_per_second": 0.286,
      "eval_wer": 102.0,
      "step": 2000
    }
  ],
  "max_steps": 6000,
  "num_train_epochs": 300,
  "total_flos": 2.04120981504e+19,
  "trial_name": null,
  "trial_params": null
}