{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 8.658008658008658,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.22,
      "learning_rate": 8.333333333333334e-06,
      "loss": 7866.0069,
      "step": 50
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 4137.7106,
      "step": 100
    },
    {
      "epoch": 0.43,
      "eval_loss": 3112.050537109375,
      "eval_runtime": 179.5895,
      "eval_samples_per_second": 9.355,
      "eval_steps_per_second": 0.468,
      "eval_wer": 1.0,
      "step": 100
    },
    {
      "epoch": 0.65,
      "learning_rate": 2.5e-05,
      "loss": 3114.5012,
      "step": 150
    },
    {
      "epoch": 0.87,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 2980.0003,
      "step": 200
    },
    {
      "epoch": 0.87,
      "eval_loss": 2958.919921875,
      "eval_runtime": 167.1595,
      "eval_samples_per_second": 10.05,
      "eval_steps_per_second": 0.503,
      "eval_wer": 1.0,
      "step": 200
    },
    {
      "epoch": 1.08,
      "learning_rate": 4.1666666666666665e-05,
      "loss": 3155.9772,
      "step": 250
    },
    {
      "epoch": 1.3,
      "learning_rate": 5e-05,
      "loss": 2779.3762,
      "step": 300
    },
    {
      "epoch": 1.3,
      "eval_loss": 2743.244140625,
      "eval_runtime": 177.6493,
      "eval_samples_per_second": 9.457,
      "eval_steps_per_second": 0.473,
      "eval_wer": 1.0,
      "step": 300
    },
    {
      "epoch": 1.52,
      "learning_rate": 5.833333333333333e-05,
      "loss": 2540.3578,
      "step": 350
    },
    {
      "epoch": 1.73,
      "learning_rate": 6.666666666666667e-05,
      "loss": 1841.6328,
      "step": 400
    },
    {
      "epoch": 1.73,
      "eval_loss": 1341.6024169921875,
      "eval_runtime": 181.145,
      "eval_samples_per_second": 9.274,
      "eval_steps_per_second": 0.464,
      "eval_wer": 0.9295629466739967,
      "step": 400
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.5e-05,
      "loss": 1248.4787,
      "step": 450
    },
    {
      "epoch": 2.16,
      "learning_rate": 8.333333333333333e-05,
      "loss": 1043.6277,
      "step": 500
    },
    {
      "epoch": 2.16,
      "eval_loss": 793.3690185546875,
      "eval_runtime": 175.8809,
      "eval_samples_per_second": 9.552,
      "eval_steps_per_second": 0.478,
      "eval_wer": 0.5867234744365035,
      "step": 500
    },
    {
      "epoch": 2.38,
      "learning_rate": 9.166666666666667e-05,
      "loss": 827.3966,
      "step": 550
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.0001,
      "loss": 834.6795,
      "step": 600
    },
    {
      "epoch": 2.6,
      "eval_loss": 582.5656127929688,
      "eval_runtime": 180.7161,
      "eval_samples_per_second": 9.296,
      "eval_steps_per_second": 0.465,
      "eval_wer": 0.44612424409015944,
      "step": 600
    },
    {
      "epoch": 2.81,
      "learning_rate": 0.00010833333333333334,
      "loss": 757.1235,
      "step": 650
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.00011666666666666667,
      "loss": 681.7599,
      "step": 700
    },
    {
      "epoch": 3.03,
      "eval_loss": 502.809814453125,
      "eval_runtime": 167.2696,
      "eval_samples_per_second": 10.044,
      "eval_steps_per_second": 0.502,
      "eval_wer": 0.40702308960967565,
      "step": 700
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.000125,
      "loss": 566.5348,
      "step": 750
    },
    {
      "epoch": 3.46,
      "learning_rate": 0.00013333333333333334,
      "loss": 499.7937,
      "step": 800
    },
    {
      "epoch": 3.46,
      "eval_loss": 472.62139892578125,
      "eval_runtime": 167.1792,
      "eval_samples_per_second": 10.049,
      "eval_steps_per_second": 0.502,
      "eval_wer": 0.3989142385926333,
      "step": 800
    },
    {
      "epoch": 3.68,
      "learning_rate": 0.00014166666666666668,
      "loss": 436.0044,
      "step": 850
    },
    {
      "epoch": 3.9,
      "learning_rate": 0.00015,
      "loss": 392.5165,
      "step": 900
    },
    {
      "epoch": 3.9,
      "eval_loss": 408.5908203125,
      "eval_runtime": 174.8526,
      "eval_samples_per_second": 9.608,
      "eval_steps_per_second": 0.48,
      "eval_wer": 0.34579439252336447,
      "step": 900
    },
    {
      "epoch": 4.11,
      "learning_rate": 0.00015833333333333332,
      "loss": 388.1201,
      "step": 950
    },
    {
      "epoch": 4.33,
      "learning_rate": 0.00016666666666666666,
      "loss": 351.2061,
      "step": 1000
    },
    {
      "epoch": 4.33,
      "eval_loss": 433.3687438964844,
      "eval_runtime": 170.1742,
      "eval_samples_per_second": 9.872,
      "eval_steps_per_second": 0.494,
      "eval_wer": 0.32449147883452445,
      "step": 1000
    },
    {
      "epoch": 4.55,
      "learning_rate": 0.000175,
      "loss": 371.0807,
      "step": 1050
    },
    {
      "epoch": 4.76,
      "learning_rate": 0.00018333333333333334,
      "loss": 332.344,
      "step": 1100
    },
    {
      "epoch": 4.76,
      "eval_loss": 412.85247802734375,
      "eval_runtime": 189.4552,
      "eval_samples_per_second": 8.868,
      "eval_steps_per_second": 0.443,
      "eval_wer": 0.32909565695437054,
      "step": 1100
    },
    {
      "epoch": 4.98,
      "learning_rate": 0.00019166666666666667,
      "loss": 332.4827,
      "step": 1150
    },
    {
      "epoch": 5.19,
      "learning_rate": 0.0002,
      "loss": 308.6846,
      "step": 1200
    },
    {
      "epoch": 5.19,
      "eval_loss": 440.1231994628906,
      "eval_runtime": 168.6145,
      "eval_samples_per_second": 9.964,
      "eval_steps_per_second": 0.498,
      "eval_wer": 0.3194062671797691,
      "step": 1200
    },
    {
      "epoch": 5.41,
      "learning_rate": 0.00020833333333333335,
      "loss": 298.0805,
      "step": 1250
    },
    {
      "epoch": 5.63,
      "learning_rate": 0.00021666666666666668,
      "loss": 318.8863,
      "step": 1300
    },
    {
      "epoch": 5.63,
      "eval_loss": 436.9075012207031,
      "eval_runtime": 166.5693,
      "eval_samples_per_second": 10.086,
      "eval_steps_per_second": 0.504,
      "eval_wer": 0.325109950522265,
      "step": 1300
    },
    {
      "epoch": 5.84,
      "learning_rate": 0.00022500000000000002,
      "loss": 317.8166,
      "step": 1350
    },
    {
      "epoch": 6.06,
      "learning_rate": 0.00023333333333333333,
      "loss": 303.7214,
      "step": 1400
    },
    {
      "epoch": 6.06,
      "eval_loss": 427.9447937011719,
      "eval_runtime": 201.1021,
      "eval_samples_per_second": 8.354,
      "eval_steps_per_second": 0.418,
      "eval_wer": 0.32387300714678396,
      "step": 1400
    },
    {
      "epoch": 6.28,
      "learning_rate": 0.00024166666666666667,
      "loss": 291.1951,
      "step": 1450
    },
    {
      "epoch": 6.49,
      "learning_rate": 0.00025,
      "loss": 282.5848,
      "step": 1500
    },
    {
      "epoch": 6.49,
      "eval_loss": 426.2596130371094,
      "eval_runtime": 183.3654,
      "eval_samples_per_second": 9.162,
      "eval_steps_per_second": 0.458,
      "eval_wer": 0.30648708081363385,
      "step": 1500
    },
    {
      "epoch": 6.71,
      "learning_rate": 0.00025833333333333334,
      "loss": 307.7221,
      "step": 1550
    },
    {
      "epoch": 6.93,
      "learning_rate": 0.0002666666666666667,
      "loss": 276.1861,
      "step": 1600
    },
    {
      "epoch": 6.93,
      "eval_loss": 495.3029479980469,
      "eval_runtime": 202.5189,
      "eval_samples_per_second": 8.296,
      "eval_steps_per_second": 0.415,
      "eval_wer": 0.3147333699835074,
      "step": 1600
    },
    {
      "epoch": 7.14,
      "learning_rate": 0.000275,
      "loss": 267.0663,
      "step": 1650
    },
    {
      "epoch": 7.36,
      "learning_rate": 0.00028333333333333335,
      "loss": 268.4355,
      "step": 1700
    },
    {
      "epoch": 7.36,
      "eval_loss": 458.63690185546875,
      "eval_runtime": 177.1854,
      "eval_samples_per_second": 9.482,
      "eval_steps_per_second": 0.474,
      "eval_wer": 0.32497251236943375,
      "step": 1700
    },
    {
      "epoch": 7.58,
      "learning_rate": 0.0002916666666666667,
      "loss": 280.6286,
      "step": 1750
    },
    {
      "epoch": 7.79,
      "learning_rate": 0.0003,
      "loss": 277.7815,
      "step": 1800
    },
    {
      "epoch": 7.79,
      "eval_loss": 540.147705078125,
      "eval_runtime": 176.8625,
      "eval_samples_per_second": 9.499,
      "eval_steps_per_second": 0.475,
      "eval_wer": 0.34819956019791093,
      "step": 1800
    },
    {
      "epoch": 8.01,
      "learning_rate": 0.00030833333333333337,
      "loss": 271.2937,
      "step": 1850
    },
    {
      "epoch": 8.23,
      "learning_rate": 0.00031666666666666665,
      "loss": 247.8558,
      "step": 1900
    },
    {
      "epoch": 8.23,
      "eval_loss": 457.4529113769531,
      "eval_runtime": 181.8308,
      "eval_samples_per_second": 9.239,
      "eval_steps_per_second": 0.462,
      "eval_wer": 0.33981583287520617,
      "step": 1900
    },
    {
      "epoch": 8.44,
      "learning_rate": 0.00032500000000000004,
      "loss": 271.9148,
      "step": 1950
    },
    {
      "epoch": 8.66,
      "learning_rate": 0.0003333333333333333,
      "loss": 325.8626,
      "step": 2000
    },
    {
      "epoch": 8.66,
      "eval_loss": 450.06524658203125,
      "eval_runtime": 162.9106,
      "eval_samples_per_second": 10.312,
      "eval_steps_per_second": 0.516,
      "eval_wer": 0.3224986256184717,
      "step": 2000
    }
  ],
  "max_steps": 2310,
  "num_train_epochs": 10,
  "total_flos": 1.1409490176804864e+18,
  "trial_name": null,
  "trial_params": null
}