{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "eval_steps": 1,
  "global_step": 80,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.125,
      "grad_norm": 117.5,
      "learning_rate": 2.5e-05,
      "loss": 3.4115,
      "step": 1
    },
    {
      "epoch": 0.125,
      "eval_accuracy": 0.08064516129032258,
      "eval_f1": 0.04975124378109453,
      "eval_loss": 2.966907024383545,
      "eval_runtime": 1.5236,
      "eval_samples_per_second": 40.694,
      "eval_steps_per_second": 2.625,
      "step": 1
    },
    {
      "epoch": 0.25,
      "grad_norm": 107.5,
      "learning_rate": 5e-05,
      "loss": 3.0649,
      "step": 2
    },
    {
      "epoch": 0.25,
      "eval_accuracy": 0.08064516129032258,
      "eval_f1": 0.04975124378109453,
      "eval_loss": 2.770594358444214,
      "eval_runtime": 1.5518,
      "eval_samples_per_second": 39.953,
      "eval_steps_per_second": 2.578,
      "step": 2
    },
    {
      "epoch": 0.375,
      "grad_norm": 114.0,
      "learning_rate": 4.935897435897436e-05,
      "loss": 2.8262,
      "step": 3
    },
    {
      "epoch": 0.375,
      "eval_accuracy": 0.1774193548387097,
      "eval_f1": 0.16345029239766082,
      "eval_loss": 1.9383505582809448,
      "eval_runtime": 1.5505,
      "eval_samples_per_second": 39.988,
      "eval_steps_per_second": 2.58,
      "step": 3
    },
    {
      "epoch": 0.5,
      "grad_norm": 77.5,
      "learning_rate": 4.871794871794872e-05,
      "loss": 1.9446,
      "step": 4
    },
    {
      "epoch": 0.5,
      "eval_accuracy": 0.46774193548387094,
      "eval_f1": 0.3246753246753247,
      "eval_loss": 1.1027910709381104,
      "eval_runtime": 1.5511,
      "eval_samples_per_second": 39.973,
      "eval_steps_per_second": 2.579,
      "step": 4
    },
    {
      "epoch": 0.625,
      "grad_norm": 25.625,
      "learning_rate": 4.8076923076923084e-05,
      "loss": 1.0798,
      "step": 5
    },
    {
      "epoch": 0.625,
      "eval_accuracy": 0.5806451612903226,
      "eval_f1": 0.3848238482384823,
      "eval_loss": 0.9913526773452759,
      "eval_runtime": 1.5011,
      "eval_samples_per_second": 41.303,
      "eval_steps_per_second": 2.665,
      "step": 5
    },
    {
      "epoch": 0.75,
      "grad_norm": 38.5,
      "learning_rate": 4.7435897435897435e-05,
      "loss": 1.1149,
      "step": 6
    },
    {
      "epoch": 0.75,
      "eval_accuracy": 0.6129032258064516,
      "eval_f1": 0.4222222222222222,
      "eval_loss": 1.0127110481262207,
      "eval_runtime": 1.5505,
      "eval_samples_per_second": 39.986,
      "eval_steps_per_second": 2.58,
      "step": 6
    },
    {
      "epoch": 0.875,
      "grad_norm": 14.0625,
      "learning_rate": 4.67948717948718e-05,
      "loss": 1.0156,
      "step": 7
    },
    {
      "epoch": 0.875,
      "eval_accuracy": 0.6612903225806451,
      "eval_f1": 0.45972458026787394,
      "eval_loss": 0.9800750017166138,
      "eval_runtime": 1.5499,
      "eval_samples_per_second": 40.001,
      "eval_steps_per_second": 2.581,
      "step": 7
    },
    {
      "epoch": 1.0,
      "grad_norm": 16.125,
      "learning_rate": 4.615384615384616e-05,
      "loss": 1.1665,
      "step": 8
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6129032258064516,
      "eval_f1": 0.42463686096962844,
      "eval_loss": 0.9667417407035828,
      "eval_runtime": 1.5498,
      "eval_samples_per_second": 40.006,
      "eval_steps_per_second": 2.581,
      "step": 8
    },
    {
      "epoch": 1.125,
      "grad_norm": 16.625,
      "learning_rate": 4.5512820512820516e-05,
      "loss": 0.9373,
      "step": 9
    },
    {
      "epoch": 1.125,
      "eval_accuracy": 0.6290322580645161,
      "eval_f1": 0.4369318181818182,
      "eval_loss": 0.8864430785179138,
      "eval_runtime": 1.5498,
      "eval_samples_per_second": 40.004,
      "eval_steps_per_second": 2.581,
      "step": 9
    },
    {
      "epoch": 1.25,
      "grad_norm": 7.78125,
      "learning_rate": 4.4871794871794874e-05,
      "loss": 0.6655,
      "step": 10
    },
    {
      "epoch": 1.25,
      "eval_accuracy": 0.6612903225806451,
      "eval_f1": 0.4548342267222553,
      "eval_loss": 0.8572250008583069,
      "eval_runtime": 1.5498,
      "eval_samples_per_second": 40.006,
      "eval_steps_per_second": 2.581,
      "step": 10
    },
    {
      "epoch": 1.375,
      "grad_norm": 27.0,
      "learning_rate": 4.423076923076923e-05,
      "loss": 0.9138,
      "step": 11
    },
    {
      "epoch": 1.375,
      "eval_accuracy": 0.6935483870967742,
      "eval_f1": 0.48024256651017216,
      "eval_loss": 0.8059869408607483,
      "eval_runtime": 1.5506,
      "eval_samples_per_second": 39.985,
      "eval_steps_per_second": 2.58,
      "step": 11
    },
    {
      "epoch": 1.5,
      "grad_norm": 8.375,
      "learning_rate": 4.358974358974359e-05,
      "loss": 0.443,
      "step": 12
    },
    {
      "epoch": 1.5,
      "eval_accuracy": 0.6612903225806451,
      "eval_f1": 0.4602089268755935,
      "eval_loss": 0.7958393692970276,
      "eval_runtime": 1.5502,
      "eval_samples_per_second": 39.994,
      "eval_steps_per_second": 2.58,
      "step": 12
    },
    {
      "epoch": 1.625,
      "grad_norm": 26.0,
      "learning_rate": 4.294871794871795e-05,
      "loss": 0.8109,
      "step": 13
    },
    {
      "epoch": 1.625,
      "eval_accuracy": 0.5806451612903226,
      "eval_f1": 0.40218826636483684,
      "eval_loss": 0.8772602677345276,
      "eval_runtime": 1.5492,
      "eval_samples_per_second": 40.021,
      "eval_steps_per_second": 2.582,
      "step": 13
    },
    {
      "epoch": 1.75,
      "grad_norm": 31.625,
      "learning_rate": 4.230769230769231e-05,
      "loss": 0.7076,
      "step": 14
    },
    {
      "epoch": 1.75,
      "eval_accuracy": 0.5806451612903226,
      "eval_f1": 0.40218826636483684,
      "eval_loss": 0.908179521560669,
      "eval_runtime": 1.5505,
      "eval_samples_per_second": 39.988,
      "eval_steps_per_second": 2.58,
      "step": 14
    },
    {
      "epoch": 1.875,
      "grad_norm": 27.25,
      "learning_rate": 4.166666666666667e-05,
      "loss": 0.6503,
      "step": 15
    },
    {
      "epoch": 1.875,
      "eval_accuracy": 0.6451612903225806,
      "eval_f1": 0.5400078064012491,
      "eval_loss": 0.8904320597648621,
      "eval_runtime": 1.5507,
      "eval_samples_per_second": 39.982,
      "eval_steps_per_second": 2.579,
      "step": 15
    },
    {
      "epoch": 2.0,
      "grad_norm": 24.5,
      "learning_rate": 4.1025641025641023e-05,
      "loss": 0.6473,
      "step": 16
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6935483870967742,
      "eval_f1": 0.5738151392632524,
      "eval_loss": 0.8587173819541931,
      "eval_runtime": 1.5497,
      "eval_samples_per_second": 40.007,
      "eval_steps_per_second": 2.581,
      "step": 16
    },
    {
      "epoch": 2.125,
      "grad_norm": 10.8125,
      "learning_rate": 4.038461538461539e-05,
      "loss": 0.608,
      "step": 17
    },
    {
      "epoch": 2.125,
      "eval_accuracy": 0.6774193548387096,
      "eval_f1": 0.4705882352941177,
      "eval_loss": 0.8885321021080017,
      "eval_runtime": 1.5511,
      "eval_samples_per_second": 39.971,
      "eval_steps_per_second": 2.579,
      "step": 17
    },
    {
      "epoch": 2.25,
      "grad_norm": 7.71875,
      "learning_rate": 3.974358974358974e-05,
      "loss": 0.6337,
      "step": 18
    },
    {
      "epoch": 2.25,
      "eval_accuracy": 0.6612903225806451,
      "eval_f1": 0.44877344877344877,
      "eval_loss": 0.9945423007011414,
      "eval_runtime": 1.5508,
      "eval_samples_per_second": 39.978,
      "eval_steps_per_second": 2.579,
      "step": 18
    },
    {
      "epoch": 2.375,
      "grad_norm": 46.0,
      "learning_rate": 3.9102564102564105e-05,
      "loss": 0.8558,
      "step": 19
    },
    {
      "epoch": 2.375,
      "eval_accuracy": 0.6129032258064516,
      "eval_f1": 0.40427350427350434,
      "eval_loss": 1.061297059059143,
      "eval_runtime": 1.5501,
      "eval_samples_per_second": 39.999,
      "eval_steps_per_second": 2.581,
      "step": 19
    },
    {
      "epoch": 2.5,
      "grad_norm": 42.0,
      "learning_rate": 3.846153846153846e-05,
      "loss": 0.8055,
      "step": 20
    },
    {
      "epoch": 2.5,
      "eval_accuracy": 0.5967741935483871,
      "eval_f1": 0.38834741173922466,
      "eval_loss": 1.065416932106018,
      "eval_runtime": 1.55,
      "eval_samples_per_second": 40.001,
      "eval_steps_per_second": 2.581,
      "step": 20
    },
    {
      "epoch": 2.625,
      "grad_norm": 47.0,
      "learning_rate": 3.782051282051282e-05,
      "loss": 0.8745,
      "step": 21
    },
    {
      "epoch": 2.625,
      "eval_accuracy": 0.6451612903225806,
      "eval_f1": 0.43443819053575156,
      "eval_loss": 1.0076254606246948,
      "eval_runtime": 1.5499,
      "eval_samples_per_second": 40.001,
      "eval_steps_per_second": 2.581,
      "step": 21
    },
    {
      "epoch": 2.75,
      "grad_norm": 38.25,
      "learning_rate": 3.717948717948718e-05,
      "loss": 0.6061,
      "step": 22
    },
    {
      "epoch": 2.75,
      "eval_accuracy": 0.6774193548387096,
      "eval_f1": 0.46266829865361075,
      "eval_loss": 0.9233418107032776,
      "eval_runtime": 1.5501,
      "eval_samples_per_second": 39.996,
      "eval_steps_per_second": 2.58,
      "step": 22
    },
    {
      "epoch": 2.875,
      "grad_norm": 19.375,
      "learning_rate": 3.653846153846154e-05,
      "loss": 0.5299,
      "step": 23
    },
    {
      "epoch": 2.875,
      "eval_accuracy": 0.7096774193548387,
      "eval_f1": 0.4925170068027211,
      "eval_loss": 0.8340729475021362,
      "eval_runtime": 1.5492,
      "eval_samples_per_second": 40.02,
      "eval_steps_per_second": 2.582,
      "step": 23
    },
    {
      "epoch": 3.0,
      "grad_norm": 5.84375,
      "learning_rate": 3.58974358974359e-05,
      "loss": 0.4845,
      "step": 24
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6935483870967742,
      "eval_f1": 0.4823957137389973,
      "eval_loss": 0.7939531803131104,
      "eval_runtime": 1.5494,
      "eval_samples_per_second": 40.015,
      "eval_steps_per_second": 2.582,
      "step": 24
    },
    {
      "epoch": 3.125,
      "grad_norm": 12.0625,
      "learning_rate": 3.525641025641026e-05,
      "loss": 0.5048,
      "step": 25
    },
    {
      "epoch": 3.125,
      "eval_accuracy": 0.6774193548387096,
      "eval_f1": 0.4705882352941177,
      "eval_loss": 0.7804250121116638,
      "eval_runtime": 1.5496,
      "eval_samples_per_second": 40.01,
      "eval_steps_per_second": 2.581,
      "step": 25
    },
    {
      "epoch": 3.25,
      "grad_norm": 10.1875,
      "learning_rate": 3.461538461538462e-05,
      "loss": 0.4168,
      "step": 26
    },
    {
      "epoch": 3.25,
      "eval_accuracy": 0.6935483870967742,
      "eval_f1": 0.5852111989074237,
      "eval_loss": 0.775275468826294,
      "eval_runtime": 1.5493,
      "eval_samples_per_second": 40.018,
      "eval_steps_per_second": 2.582,
      "step": 26
    },
    {
      "epoch": 3.375,
      "grad_norm": 19.75,
      "learning_rate": 3.397435897435898e-05,
      "loss": 0.672,
      "step": 27
    },
    {
      "epoch": 3.375,
      "eval_accuracy": 0.7096774193548387,
      "eval_f1": 0.5962091503267973,
      "eval_loss": 0.7586394548416138,
      "eval_runtime": 1.5491,
      "eval_samples_per_second": 40.024,
      "eval_steps_per_second": 2.582,
      "step": 27
    },
    {
      "epoch": 3.5,
      "grad_norm": 9.9375,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.4323,
      "step": 28
    },
    {
      "epoch": 3.5,
      "eval_accuracy": 0.7096774193548387,
      "eval_f1": 0.5971250971250971,
      "eval_loss": 0.7453081607818604,
      "eval_runtime": 1.5496,
      "eval_samples_per_second": 40.011,
      "eval_steps_per_second": 2.581,
      "step": 28
    },
    {
      "epoch": 3.625,
      "grad_norm": 6.34375,
      "learning_rate": 3.269230769230769e-05,
      "loss": 0.3613,
      "step": 29
    },
    {
      "epoch": 3.625,
      "eval_accuracy": 0.6935483870967742,
      "eval_f1": 0.586002257700371,
      "eval_loss": 0.7488570809364319,
      "eval_runtime": 1.5495,
      "eval_samples_per_second": 40.014,
      "eval_steps_per_second": 2.582,
      "step": 29
    },
    {
      "epoch": 3.75,
      "grad_norm": 10.25,
      "learning_rate": 3.205128205128206e-05,
      "loss": 0.5648,
      "step": 30
    },
    {
      "epoch": 3.75,
      "eval_accuracy": 0.6612903225806451,
      "eval_f1": 0.5636084747387594,
      "eval_loss": 0.7780008912086487,
      "eval_runtime": 1.5496,
      "eval_samples_per_second": 40.01,
      "eval_steps_per_second": 2.581,
      "step": 30
    },
    {
      "epoch": 3.875,
      "grad_norm": 17.75,
      "learning_rate": 3.141025641025641e-05,
      "loss": 0.3465,
      "step": 31
    },
    {
      "epoch": 3.875,
      "eval_accuracy": 0.6451612903225806,
      "eval_f1": 0.5513409961685825,
      "eval_loss": 0.7948296070098877,
      "eval_runtime": 1.549,
      "eval_samples_per_second": 40.025,
      "eval_steps_per_second": 2.582,
      "step": 31
    },
    {
      "epoch": 4.0,
      "grad_norm": 22.75,
      "learning_rate": 3.0769230769230774e-05,
      "loss": 0.4818,
      "step": 32
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6774193548387096,
      "eval_f1": 0.5750128008192524,
      "eval_loss": 0.7759857177734375,
      "eval_runtime": 1.5495,
      "eval_samples_per_second": 40.012,
      "eval_steps_per_second": 2.581,
      "step": 32
    },
    {
      "epoch": 4.125,
      "grad_norm": 20.625,
      "learning_rate": 3.012820512820513e-05,
      "loss": 0.3397,
      "step": 33
    },
    {
      "epoch": 4.125,
      "eval_accuracy": 0.6935483870967742,
      "eval_f1": 0.5863395863395864,
      "eval_loss": 0.739683210849762,
      "eval_runtime": 1.5501,
      "eval_samples_per_second": 39.998,
      "eval_steps_per_second": 2.581,
      "step": 33
    },
    {
      "epoch": 4.25,
      "grad_norm": 7.46875,
      "learning_rate": 2.948717948717949e-05,
      "loss": 0.3511,
      "step": 34
    },
    {
      "epoch": 4.25,
      "eval_accuracy": 0.7258064516129032,
      "eval_f1": 0.6082333430884792,
      "eval_loss": 0.7238311767578125,
      "eval_runtime": 1.5499,
      "eval_samples_per_second": 40.002,
      "eval_steps_per_second": 2.581,
      "step": 34
    },
    {
      "epoch": 4.375,
      "grad_norm": 22.0,
      "learning_rate": 2.8846153846153845e-05,
      "loss": 0.5464,
      "step": 35
    },
    {
      "epoch": 4.375,
      "eval_accuracy": 0.7096774193548387,
      "eval_f1": 0.5825569358178054,
      "eval_loss": 0.7330479621887207,
      "eval_runtime": 1.5002,
      "eval_samples_per_second": 41.329,
      "eval_steps_per_second": 2.666,
      "step": 35
    },
    {
      "epoch": 4.5,
      "grad_norm": 7.375,
      "learning_rate": 2.8205128205128207e-05,
      "loss": 0.4142,
      "step": 36
    },
    {
      "epoch": 4.5,
      "eval_accuracy": 0.7096774193548387,
      "eval_f1": 0.5825569358178054,
      "eval_loss": 0.7332301139831543,
      "eval_runtime": 1.5502,
      "eval_samples_per_second": 39.995,
      "eval_steps_per_second": 2.58,
      "step": 36
    },
    {
      "epoch": 4.625,
      "grad_norm": 10.375,
      "learning_rate": 2.756410256410257e-05,
      "loss": 0.3316,
      "step": 37
    },
    {
      "epoch": 4.625,
      "eval_accuracy": 0.7096774193548387,
      "eval_f1": 0.583994314143568,
      "eval_loss": 0.7186973094940186,
      "eval_runtime": 1.5504,
      "eval_samples_per_second": 39.991,
      "eval_steps_per_second": 2.58,
      "step": 37
    },
    {
      "epoch": 4.75,
      "grad_norm": 14.5,
      "learning_rate": 2.6923076923076923e-05,
      "loss": 0.3023,
      "step": 38
    },
    {
      "epoch": 4.75,
      "eval_accuracy": 0.6774193548387096,
      "eval_f1": 0.5748456790123457,
      "eval_loss": 0.718741774559021,
      "eval_runtime": 1.5491,
      "eval_samples_per_second": 40.023,
      "eval_steps_per_second": 2.582,
      "step": 38
    },
    {
      "epoch": 4.875,
      "grad_norm": 8.8125,
      "learning_rate": 2.6282051282051285e-05,
      "loss": 0.3108,
      "step": 39
    },
    {
      "epoch": 4.875,
      "eval_accuracy": 0.6774193548387096,
      "eval_f1": 0.5747126436781609,
      "eval_loss": 0.7788988947868347,
      "eval_runtime": 1.5485,
      "eval_samples_per_second": 40.038,
      "eval_steps_per_second": 2.583,
      "step": 39
    },
    {
      "epoch": 5.0,
      "grad_norm": 16.0,
      "learning_rate": 2.564102564102564e-05,
      "loss": 0.5096,
      "step": 40
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6774193548387096,
      "eval_f1": 0.5747126436781609,
      "eval_loss": 0.7997355461120605,
      "eval_runtime": 1.549,
      "eval_samples_per_second": 40.026,
      "eval_steps_per_second": 2.582,
      "step": 40
    },
    {
      "epoch": 5.125,
      "grad_norm": 24.75,
      "learning_rate": 2.5e-05,
      "loss": 0.6856,
      "step": 41
    },
    {
      "epoch": 5.125,
      "eval_accuracy": 0.6774193548387096,
      "eval_f1": 0.5747126436781609,
      "eval_loss": 0.7944700121879578,
      "eval_runtime": 1.5487,
      "eval_samples_per_second": 40.034,
      "eval_steps_per_second": 2.583,
      "step": 41
    },
    {
      "epoch": 5.25,
      "grad_norm": 17.75,
      "learning_rate": 2.435897435897436e-05,
      "loss": 0.3305,
      "step": 42
    },
    {
      "epoch": 5.25,
      "eval_accuracy": 0.6774193548387096,
      "eval_f1": 0.5750128008192524,
      "eval_loss": 0.7675864696502686,
      "eval_runtime": 1.5494,
      "eval_samples_per_second": 40.017,
      "eval_steps_per_second": 2.582,
      "step": 42
    },
    {
      "epoch": 5.375,
      "grad_norm": 10.5625,
      "learning_rate": 2.3717948717948718e-05,
      "loss": 0.3834,
      "step": 43
    },
    {
      "epoch": 5.375,
      "eval_accuracy": 0.6774193548387096,
      "eval_f1": 0.5748456790123457,
      "eval_loss": 0.7341117858886719,
      "eval_runtime": 1.5491,
      "eval_samples_per_second": 40.025,
      "eval_steps_per_second": 2.582,
      "step": 43
    },
    {
      "epoch": 5.5,
      "grad_norm": 8.1875,
      "learning_rate": 2.307692307692308e-05,
      "loss": 0.2876,
      "step": 44
    },
    {
      "epoch": 5.5,
      "eval_accuracy": 0.6774193548387096,
      "eval_f1": 0.5748456790123457,
      "eval_loss": 0.7020472884178162,
      "eval_runtime": 1.499,
      "eval_samples_per_second": 41.36,
      "eval_steps_per_second": 2.668,
      "step": 44
    },
    {
      "epoch": 5.625,
      "grad_norm": 7.0625,
      "learning_rate": 2.2435897435897437e-05,
      "loss": 0.2778,
      "step": 45
    },
    {
      "epoch": 5.625,
      "eval_accuracy": 0.6935483870967742,
      "eval_f1": 0.586002257700371,
      "eval_loss": 0.6907176375389099,
      "eval_runtime": 1.5487,
      "eval_samples_per_second": 40.034,
      "eval_steps_per_second": 2.583,
      "step": 45
    },
    {
      "epoch": 5.75,
      "grad_norm": 5.65625,
      "learning_rate": 2.1794871794871795e-05,
      "loss": 0.1332,
      "step": 46
    },
    {
      "epoch": 5.75,
      "eval_accuracy": 0.6935483870967742,
      "eval_f1": 0.5852111989074237,
      "eval_loss": 0.6734468936920166,
      "eval_runtime": 1.4986,
      "eval_samples_per_second": 41.372,
      "eval_steps_per_second": 2.669,
      "step": 46
    },
    {
      "epoch": 5.875,
      "grad_norm": 8.3125,
      "learning_rate": 2.1153846153846154e-05,
      "loss": 0.1979,
      "step": 47
    },
    {
      "epoch": 5.875,
      "eval_accuracy": 0.7096774193548387,
      "eval_f1": 0.5971250971250971,
      "eval_loss": 0.6837708353996277,
      "eval_runtime": 1.5495,
      "eval_samples_per_second": 40.014,
      "eval_steps_per_second": 2.582,
      "step": 47
    },
    {
      "epoch": 6.0,
      "grad_norm": 17.375,
      "learning_rate": 2.0512820512820512e-05,
      "loss": 0.429,
      "step": 48
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7096774193548387,
      "eval_f1": 0.5971250971250971,
      "eval_loss": 0.6831167936325073,
      "eval_runtime": 1.4996,
      "eval_samples_per_second": 41.344,
      "eval_steps_per_second": 2.667,
      "step": 48
    },
    {
      "epoch": 6.125,
      "grad_norm": 8.375,
      "learning_rate": 1.987179487179487e-05,
      "loss": 0.2167,
      "step": 49
    },
    {
      "epoch": 6.125,
      "eval_accuracy": 0.6774193548387096,
      "eval_f1": 0.5748456790123457,
      "eval_loss": 0.6853166222572327,
      "eval_runtime": 1.5491,
      "eval_samples_per_second": 40.022,
      "eval_steps_per_second": 2.582,
      "step": 49
    },
    {
      "epoch": 6.25,
      "grad_norm": 4.875,
      "learning_rate": 1.923076923076923e-05,
      "loss": 0.2167,
      "step": 50
    },
    {
      "epoch": 6.25,
      "eval_accuracy": 0.6935483870967742,
      "eval_f1": 0.5863395863395864,
      "eval_loss": 0.7052285075187683,
      "eval_runtime": 1.5486,
      "eval_samples_per_second": 40.036,
      "eval_steps_per_second": 2.583,
      "step": 50
    },
    {
      "epoch": 6.375,
      "grad_norm": 6.03125,
      "learning_rate": 1.858974358974359e-05,
      "loss": 0.1658,
      "step": 51
    },
    {
      "epoch": 6.375,
      "eval_accuracy": 0.6774193548387096,
      "eval_f1": 0.5750128008192524,
      "eval_loss": 0.7341977953910828,
      "eval_runtime": 1.5484,
      "eval_samples_per_second": 40.041,
      "eval_steps_per_second": 2.583,
      "step": 51
    },
    {
      "epoch": 6.5,
      "grad_norm": 8.375,
      "learning_rate": 1.794871794871795e-05,
      "loss": 0.2113,
      "step": 52
    },
    {
      "epoch": 6.5,
      "eval_accuracy": 0.6774193548387096,
      "eval_f1": 0.5750128008192524,
      "eval_loss": 0.7526947855949402,
      "eval_runtime": 1.549,
      "eval_samples_per_second": 40.027,
      "eval_steps_per_second": 2.582,
      "step": 52
    },
    {
      "epoch": 6.625,
      "grad_norm": 16.125,
      "learning_rate": 1.730769230769231e-05,
      "loss": 0.3413,
      "step": 53
    },
    {
      "epoch": 6.625,
      "eval_accuracy": 0.6774193548387096,
      "eval_f1": 0.5750128008192524,
      "eval_loss": 0.7378318309783936,
      "eval_runtime": 1.5486,
      "eval_samples_per_second": 40.036,
      "eval_steps_per_second": 2.583,
      "step": 53
    },
    {
      "epoch": 6.75,
      "grad_norm": 11.9375,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.2154,
      "step": 54
    },
    {
      "epoch": 6.75,
      "eval_accuracy": 0.6935483870967742,
      "eval_f1": 0.5863395863395864,
      "eval_loss": 0.7200038433074951,
      "eval_runtime": 1.5496,
      "eval_samples_per_second": 40.011,
      "eval_steps_per_second": 2.581,
      "step": 54
    },
    {
      "epoch": 6.875,
      "grad_norm": 10.375,
      "learning_rate": 1.602564102564103e-05,
      "loss": 0.2632,
      "step": 55
    },
    {
      "epoch": 6.875,
      "eval_accuracy": 0.6774193548387096,
      "eval_f1": 0.562610229276896,
      "eval_loss": 0.7117968797683716,
      "eval_runtime": 1.5485,
      "eval_samples_per_second": 40.038,
      "eval_steps_per_second": 2.583,
      "step": 55
    },
    {
      "epoch": 7.0,
      "grad_norm": 10.6875,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.2256,
      "step": 56
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.7096774193548387,
      "eval_f1": 0.649046620744734,
      "eval_loss": 0.7059355974197388,
      "eval_runtime": 1.5485,
      "eval_samples_per_second": 40.04,
      "eval_steps_per_second": 2.583,
      "step": 56
    },
    {
      "epoch": 7.125,
      "grad_norm": 6.4375,
      "learning_rate": 1.4743589743589745e-05,
      "loss": 0.1999,
      "step": 57
    },
    {
      "epoch": 7.125,
      "eval_accuracy": 0.6774193548387096,
      "eval_f1": 0.5924185104512973,
      "eval_loss": 0.7015820145606995,
      "eval_runtime": 1.5488,
      "eval_samples_per_second": 40.031,
      "eval_steps_per_second": 2.583,
      "step": 57
    },
    {
      "epoch": 7.25,
      "grad_norm": 7.84375,
      "learning_rate": 1.4102564102564104e-05,
      "loss": 0.1985,
      "step": 58
    },
    {
      "epoch": 7.25,
      "eval_accuracy": 0.6774193548387096,
      "eval_f1": 0.6150056839712011,
      "eval_loss": 0.7118080854415894,
      "eval_runtime": 1.5491,
      "eval_samples_per_second": 40.023,
      "eval_steps_per_second": 2.582,
      "step": 58
    },
    {
      "epoch": 7.375,
      "grad_norm": 10.125,
      "learning_rate": 1.3461538461538462e-05,
      "loss": 0.2274,
      "step": 59
    },
    {
      "epoch": 7.375,
      "eval_accuracy": 0.6774193548387096,
      "eval_f1": 0.6150056839712011,
      "eval_loss": 0.7169931530952454,
      "eval_runtime": 1.549,
      "eval_samples_per_second": 40.027,
      "eval_steps_per_second": 2.582,
      "step": 59
    },
    {
      "epoch": 7.5,
      "grad_norm": 7.1875,
      "learning_rate": 1.282051282051282e-05,
      "loss": 0.1642,
      "step": 60
    },
    {
      "epoch": 7.5,
      "eval_accuracy": 0.6935483870967742,
      "eval_f1": 0.6342191748828052,
      "eval_loss": 0.7017022967338562,
      "eval_runtime": 1.5484,
      "eval_samples_per_second": 40.041,
      "eval_steps_per_second": 2.583,
      "step": 60
    },
    {
      "epoch": 7.625,
      "grad_norm": 6.625,
      "learning_rate": 1.217948717948718e-05,
      "loss": 0.1695,
      "step": 61
    },
    {
      "epoch": 7.625,
      "eval_accuracy": 0.6935483870967742,
      "eval_f1": 0.6231532078526614,
      "eval_loss": 0.7116390466690063,
      "eval_runtime": 1.5485,
      "eval_samples_per_second": 40.037,
      "eval_steps_per_second": 2.583,
      "step": 61
    },
    {
      "epoch": 7.75,
      "grad_norm": 10.375,
      "learning_rate": 1.153846153846154e-05,
      "loss": 0.2007,
      "step": 62
    },
    {
      "epoch": 7.75,
      "eval_accuracy": 0.7096774193548387,
      "eval_f1": 0.6495280675608545,
      "eval_loss": 0.7067845463752747,
      "eval_runtime": 1.5486,
      "eval_samples_per_second": 40.037,
      "eval_steps_per_second": 2.583,
      "step": 62
    },
    {
      "epoch": 7.875,
      "grad_norm": 13.5625,
      "learning_rate": 1.0897435897435898e-05,
      "loss": 0.1979,
      "step": 63
    },
    {
      "epoch": 7.875,
      "eval_accuracy": 0.7096774193548387,
      "eval_f1": 0.5976702508960573,
      "eval_loss": 0.7046316266059875,
      "eval_runtime": 1.5484,
      "eval_samples_per_second": 40.042,
      "eval_steps_per_second": 2.583,
      "step": 63
    },
    {
      "epoch": 8.0,
      "grad_norm": 9.5,
      "learning_rate": 1.0256410256410256e-05,
      "loss": 0.1621,
      "step": 64
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7096774193548387,
      "eval_f1": 0.5976702508960573,
      "eval_loss": 0.6941011548042297,
      "eval_runtime": 1.5521,
      "eval_samples_per_second": 39.945,
      "eval_steps_per_second": 2.577,
      "step": 64
    },
    {
      "epoch": 8.125,
      "grad_norm": 10.5625,
      "learning_rate": 9.615384615384616e-06,
      "loss": 0.1881,
      "step": 65
    },
    {
      "epoch": 8.125,
      "eval_accuracy": 0.7096774193548387,
      "eval_f1": 0.5976702508960573,
      "eval_loss": 0.6907891035079956,
      "eval_runtime": 1.5016,
      "eval_samples_per_second": 41.29,
      "eval_steps_per_second": 2.664,
      "step": 65
    },
    {
      "epoch": 8.25,
      "grad_norm": 9.1875,
      "learning_rate": 8.974358974358976e-06,
      "loss": 0.174,
      "step": 66
    },
    {
      "epoch": 8.25,
      "eval_accuracy": 0.7096774193548387,
      "eval_f1": 0.5976702508960573,
      "eval_loss": 0.6863110661506653,
      "eval_runtime": 1.5523,
      "eval_samples_per_second": 39.942,
      "eval_steps_per_second": 2.577,
      "step": 66
    },
    {
      "epoch": 8.375,
      "grad_norm": 6.40625,
      "learning_rate": 8.333333333333334e-06,
      "loss": 0.1741,
      "step": 67
    },
    {
      "epoch": 8.375,
      "eval_accuracy": 0.7096774193548387,
      "eval_f1": 0.5976702508960573,
      "eval_loss": 0.67368084192276,
      "eval_runtime": 1.5484,
      "eval_samples_per_second": 40.041,
      "eval_steps_per_second": 2.583,
      "step": 67
    },
    {
      "epoch": 8.5,
      "grad_norm": 4.8125,
      "learning_rate": 7.692307692307694e-06,
      "loss": 0.0817,
      "step": 68
    },
    {
      "epoch": 8.5,
      "eval_accuracy": 0.7096774193548387,
      "eval_f1": 0.5976702508960573,
      "eval_loss": 0.6629729866981506,
      "eval_runtime": 1.5481,
      "eval_samples_per_second": 40.048,
      "eval_steps_per_second": 2.584,
      "step": 68
    },
    {
      "epoch": 8.625,
      "grad_norm": 7.875,
      "learning_rate": 7.051282051282052e-06,
      "loss": 0.1197,
      "step": 69
    },
    {
      "epoch": 8.625,
      "eval_accuracy": 0.6935483870967742,
      "eval_f1": 0.5863395863395864,
      "eval_loss": 0.6663243770599365,
      "eval_runtime": 1.5483,
      "eval_samples_per_second": 40.044,
      "eval_steps_per_second": 2.583,
      "step": 69
    },
    {
      "epoch": 8.75,
      "grad_norm": 7.34375,
      "learning_rate": 6.41025641025641e-06,
      "loss": 0.1191,
      "step": 70
    },
    {
      "epoch": 8.75,
      "eval_accuracy": 0.6612903225806451,
      "eval_f1": 0.5508198562443846,
      "eval_loss": 0.6636062860488892,
      "eval_runtime": 1.4994,
      "eval_samples_per_second": 41.351,
      "eval_steps_per_second": 2.668,
      "step": 70
    },
    {
      "epoch": 8.875,
      "grad_norm": 8.75,
      "learning_rate": 5.76923076923077e-06,
      "loss": 0.1876,
      "step": 71
    },
    {
      "epoch": 8.875,
      "eval_accuracy": 0.6774193548387096,
      "eval_f1": 0.5619047619047619,
      "eval_loss": 0.6584691405296326,
      "eval_runtime": 1.5027,
      "eval_samples_per_second": 41.259,
      "eval_steps_per_second": 2.662,
      "step": 71
    },
    {
      "epoch": 9.0,
      "grad_norm": 10.375,
      "learning_rate": 5.128205128205128e-06,
      "loss": 0.1564,
      "step": 72
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.6774193548387096,
      "eval_f1": 0.5619047619047619,
      "eval_loss": 0.6601389646530151,
      "eval_runtime": 1.5021,
      "eval_samples_per_second": 41.277,
      "eval_steps_per_second": 2.663,
      "step": 72
    },
    {
      "epoch": 9.125,
      "grad_norm": 9.375,
      "learning_rate": 4.487179487179488e-06,
      "loss": 0.1609,
      "step": 73
    },
    {
      "epoch": 9.125,
      "eval_accuracy": 0.6935483870967742,
      "eval_f1": 0.5738151392632524,
      "eval_loss": 0.6509959101676941,
      "eval_runtime": 1.5528,
      "eval_samples_per_second": 39.927,
      "eval_steps_per_second": 2.576,
      "step": 73
    },
    {
      "epoch": 9.25,
      "grad_norm": 5.90625,
      "learning_rate": 3.846153846153847e-06,
      "loss": 0.0808,
      "step": 74
    },
    {
      "epoch": 9.25,
      "eval_accuracy": 0.6774193548387096,
      "eval_f1": 0.562610229276896,
      "eval_loss": 0.6614367961883545,
      "eval_runtime": 1.5486,
      "eval_samples_per_second": 40.035,
      "eval_steps_per_second": 2.583,
      "step": 74
    },
    {
      "epoch": 9.375,
      "grad_norm": 4.5,
      "learning_rate": 3.205128205128205e-06,
      "loss": 0.123,
      "step": 75
    },
    {
      "epoch": 9.375,
      "eval_accuracy": 0.6774193548387096,
      "eval_f1": 0.562610229276896,
      "eval_loss": 0.6519985795021057,
      "eval_runtime": 1.5483,
      "eval_samples_per_second": 40.043,
      "eval_steps_per_second": 2.583,
      "step": 75
    },
    {
      "epoch": 9.5,
      "grad_norm": 10.9375,
      "learning_rate": 2.564102564102564e-06,
      "loss": 0.1828,
      "step": 76
    },
    {
      "epoch": 9.5,
      "eval_accuracy": 0.6774193548387096,
      "eval_f1": 0.562610229276896,
      "eval_loss": 0.6541938781738281,
      "eval_runtime": 1.5484,
      "eval_samples_per_second": 40.042,
      "eval_steps_per_second": 2.583,
      "step": 76
    },
    {
      "epoch": 9.625,
      "grad_norm": 4.09375,
      "learning_rate": 1.9230769230769234e-06,
      "loss": 0.1006,
      "step": 77
    },
    {
      "epoch": 9.625,
      "eval_accuracy": 0.6935483870967742,
      "eval_f1": 0.5742214774472839,
      "eval_loss": 0.6612020134925842,
      "eval_runtime": 1.5478,
      "eval_samples_per_second": 40.058,
      "eval_steps_per_second": 2.584,
      "step": 77
    },
    {
      "epoch": 9.75,
      "grad_norm": 5.0,
      "learning_rate": 1.282051282051282e-06,
      "loss": 0.0964,
      "step": 78
    },
    {
      "epoch": 9.75,
      "eval_accuracy": 0.6935483870967742,
      "eval_f1": 0.6377937076861807,
      "eval_loss": 0.6558251976966858,
      "eval_runtime": 1.5484,
      "eval_samples_per_second": 40.042,
      "eval_steps_per_second": 2.583,
      "step": 78
    },
    {
      "epoch": 9.875,
      "grad_norm": 4.875,
      "learning_rate": 6.41025641025641e-07,
      "loss": 0.0975,
      "step": 79
    },
    {
      "epoch": 9.875,
      "eval_accuracy": 0.6774193548387096,
      "eval_f1": 0.562610229276896,
      "eval_loss": 0.6551134586334229,
      "eval_runtime": 1.5484,
      "eval_samples_per_second": 40.042,
      "eval_steps_per_second": 2.583,
      "step": 79
    },
    {
      "epoch": 10.0,
      "grad_norm": 5.03125,
      "learning_rate": 0.0,
      "loss": 0.0893,
      "step": 80
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.6935483870967742,
      "eval_f1": 0.5742214774472839,
      "eval_loss": 0.6528465747833252,
      "eval_runtime": 1.5484,
      "eval_samples_per_second": 40.041,
      "eval_steps_per_second": 2.583,
      "step": 80
    },
    {
      "epoch": 10.0,
      "step": 80,
      "total_flos": 1.5526536052473856e+16,
      "train_loss": 0.5239600273780525,
      "train_runtime": 278.2601,
      "train_samples_per_second": 8.769,
      "train_steps_per_second": 0.288
    }
  ],
  "logging_steps": 1,
  "max_steps": 80,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.5526536052473856e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}