{
  "best_metric": 0.1153767853975296,
  "best_model_checkpoint": "./checkpoint-86",
  "epoch": 4.977491961414791,
  "eval_steps": 500,
  "global_step": 86,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06,
      "learning_rate": 2.9996470588235293e-05,
      "loss": 0.1951,
      "step": 1
    },
    {
      "epoch": 0.12,
      "learning_rate": 2.9992941176470588e-05,
      "loss": 0.3401,
      "step": 2
    },
    {
      "epoch": 0.17,
      "learning_rate": 2.9989411764705883e-05,
      "loss": 0.267,
      "step": 3
    },
    {
      "epoch": 0.23,
      "learning_rate": 2.9985882352941178e-05,
      "loss": 0.4651,
      "step": 4
    },
    {
      "epoch": 0.29,
      "learning_rate": 2.998235294117647e-05,
      "loss": 0.6765,
      "step": 5
    },
    {
      "epoch": 0.35,
      "learning_rate": 2.9978823529411765e-05,
      "loss": 0.4293,
      "step": 6
    },
    {
      "epoch": 0.41,
      "learning_rate": 2.997529411764706e-05,
      "loss": 0.2214,
      "step": 7
    },
    {
      "epoch": 0.46,
      "learning_rate": 2.9971764705882355e-05,
      "loss": 0.4923,
      "step": 8
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.9968235294117647e-05,
      "loss": 0.2441,
      "step": 9
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.9964705882352942e-05,
      "loss": 0.5719,
      "step": 10
    },
    {
      "epoch": 0.64,
      "learning_rate": 2.9961176470588237e-05,
      "loss": 0.2243,
      "step": 11
    },
    {
      "epoch": 0.69,
      "learning_rate": 2.9957647058823532e-05,
      "loss": 0.7649,
      "step": 12
    },
    {
      "epoch": 0.75,
      "learning_rate": 2.9954117647058824e-05,
      "loss": 0.1913,
      "step": 13
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.9950588235294116e-05,
      "loss": 0.1986,
      "step": 14
    },
    {
      "epoch": 0.87,
      "learning_rate": 2.9947058823529414e-05,
      "loss": 0.2042,
      "step": 15
    },
    {
      "epoch": 0.93,
      "learning_rate": 2.9943529411764706e-05,
      "loss": 0.3077,
      "step": 16
    },
    {
      "epoch": 0.98,
      "learning_rate": 2.994e-05,
      "loss": 0.2246,
      "step": 17
    },
    {
      "epoch": 0.98,
      "eval_loss": 0.15935464203357697,
      "eval_runtime": 17.1948,
      "eval_samples_per_second": 11.224,
      "eval_steps_per_second": 1.919,
      "step": 17
    },
    {
      "epoch": 1.04,
      "learning_rate": 2.9936470588235293e-05,
      "loss": 0.2669,
      "step": 18
    },
    {
      "epoch": 1.1,
      "learning_rate": 2.9932941176470588e-05,
      "loss": 0.1503,
      "step": 19
    },
    {
      "epoch": 1.16,
      "learning_rate": 2.9929411764705883e-05,
      "loss": 0.1519,
      "step": 20
    },
    {
      "epoch": 1.22,
      "learning_rate": 2.992588235294118e-05,
      "loss": 0.251,
      "step": 21
    },
    {
      "epoch": 1.27,
      "learning_rate": 2.992235294117647e-05,
      "loss": 0.1592,
      "step": 22
    },
    {
      "epoch": 1.33,
      "learning_rate": 2.9918823529411765e-05,
      "loss": 0.2098,
      "step": 23
    },
    {
      "epoch": 1.39,
      "learning_rate": 2.991529411764706e-05,
      "loss": 0.1687,
      "step": 24
    },
    {
      "epoch": 1.45,
      "learning_rate": 2.9911764705882352e-05,
      "loss": 0.2123,
      "step": 25
    },
    {
      "epoch": 1.5,
      "learning_rate": 2.9908235294117648e-05,
      "loss": 0.1677,
      "step": 26
    },
    {
      "epoch": 1.56,
      "learning_rate": 2.990470588235294e-05,
      "loss": 0.3109,
      "step": 27
    },
    {
      "epoch": 1.62,
      "learning_rate": 2.9901176470588238e-05,
      "loss": 0.6994,
      "step": 28
    },
    {
      "epoch": 1.68,
      "learning_rate": 2.989764705882353e-05,
      "loss": 1.4716,
      "step": 29
    },
    {
      "epoch": 1.74,
      "learning_rate": 2.9894117647058825e-05,
      "loss": 0.378,
      "step": 30
    },
    {
      "epoch": 1.79,
      "learning_rate": 2.9890588235294116e-05,
      "loss": 1.4485,
      "step": 31
    },
    {
      "epoch": 1.85,
      "learning_rate": 2.9887058823529415e-05,
      "loss": 0.2106,
      "step": 32
    },
    {
      "epoch": 1.91,
      "learning_rate": 2.9883529411764707e-05,
      "loss": 0.237,
      "step": 33
    },
    {
      "epoch": 1.97,
      "learning_rate": 2.9880000000000002e-05,
      "loss": 0.3392,
      "step": 34
    },
    {
      "epoch": 1.97,
      "eval_loss": 0.21225062012672424,
      "eval_runtime": 17.2008,
      "eval_samples_per_second": 11.22,
      "eval_steps_per_second": 1.919,
      "step": 34
    },
    {
      "epoch": 2.03,
      "learning_rate": 2.9876470588235294e-05,
      "loss": 0.1883,
      "step": 35
    },
    {
      "epoch": 2.08,
      "learning_rate": 2.9872941176470592e-05,
      "loss": 0.1791,
      "step": 36
    },
    {
      "epoch": 2.14,
      "learning_rate": 2.9869411764705884e-05,
      "loss": 0.1241,
      "step": 37
    },
    {
      "epoch": 2.2,
      "learning_rate": 2.9865882352941176e-05,
      "loss": 0.1891,
      "step": 38
    },
    {
      "epoch": 2.26,
      "learning_rate": 2.986235294117647e-05,
      "loss": 0.1484,
      "step": 39
    },
    {
      "epoch": 2.32,
      "learning_rate": 2.9858823529411763e-05,
      "loss": 0.1614,
      "step": 40
    },
    {
      "epoch": 2.37,
      "learning_rate": 2.985529411764706e-05,
      "loss": 0.1053,
      "step": 41
    },
    {
      "epoch": 2.43,
      "learning_rate": 2.9851764705882353e-05,
      "loss": 0.1314,
      "step": 42
    },
    {
      "epoch": 2.49,
      "learning_rate": 2.9848235294117648e-05,
      "loss": 0.1592,
      "step": 43
    },
    {
      "epoch": 2.55,
      "learning_rate": 2.984470588235294e-05,
      "loss": 0.123,
      "step": 44
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.984117647058824e-05,
      "loss": 0.0935,
      "step": 45
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.983764705882353e-05,
      "loss": 0.1293,
      "step": 46
    },
    {
      "epoch": 2.72,
      "learning_rate": 2.9834117647058825e-05,
      "loss": 0.1652,
      "step": 47
    },
    {
      "epoch": 2.78,
      "learning_rate": 2.9830588235294117e-05,
      "loss": 0.1558,
      "step": 48
    },
    {
      "epoch": 2.84,
      "learning_rate": 2.9827058823529416e-05,
      "loss": 0.0893,
      "step": 49
    },
    {
      "epoch": 2.89,
      "learning_rate": 2.9823529411764707e-05,
      "loss": 0.1105,
      "step": 50
    },
    {
      "epoch": 2.95,
      "learning_rate": 2.982e-05,
      "loss": 0.1162,
      "step": 51
    },
    {
      "epoch": 2.95,
      "eval_loss": 0.12937574088573456,
      "eval_runtime": 17.1968,
      "eval_samples_per_second": 11.223,
      "eval_steps_per_second": 1.919,
      "step": 51
    },
    {
      "epoch": 3.01,
      "learning_rate": 2.9816470588235294e-05,
      "loss": 0.1067,
      "step": 52
    },
    {
      "epoch": 3.07,
      "learning_rate": 2.981294117647059e-05,
      "loss": 0.1177,
      "step": 53
    },
    {
      "epoch": 3.13,
      "learning_rate": 2.9809411764705885e-05,
      "loss": 0.0819,
      "step": 54
    },
    {
      "epoch": 3.18,
      "learning_rate": 2.9805882352941176e-05,
      "loss": 0.0578,
      "step": 55
    },
    {
      "epoch": 3.24,
      "learning_rate": 2.980235294117647e-05,
      "loss": 0.0917,
      "step": 56
    },
    {
      "epoch": 3.3,
      "learning_rate": 2.9798823529411763e-05,
      "loss": 0.1044,
      "step": 57
    },
    {
      "epoch": 3.36,
      "learning_rate": 2.9795294117647062e-05,
      "loss": 0.0585,
      "step": 58
    },
    {
      "epoch": 3.41,
      "learning_rate": 2.9791764705882354e-05,
      "loss": 0.0468,
      "step": 59
    },
    {
      "epoch": 3.47,
      "learning_rate": 2.978823529411765e-05,
      "loss": 0.0567,
      "step": 60
    },
    {
      "epoch": 3.53,
      "learning_rate": 2.978470588235294e-05,
      "loss": 0.0923,
      "step": 61
    },
    {
      "epoch": 3.59,
      "learning_rate": 2.9781176470588236e-05,
      "loss": 0.0594,
      "step": 62
    },
    {
      "epoch": 3.65,
      "learning_rate": 2.977764705882353e-05,
      "loss": 0.042,
      "step": 63
    },
    {
      "epoch": 3.7,
      "learning_rate": 2.9774117647058822e-05,
      "loss": 0.0584,
      "step": 64
    },
    {
      "epoch": 3.76,
      "learning_rate": 2.9770588235294118e-05,
      "loss": 0.0982,
      "step": 65
    },
    {
      "epoch": 3.82,
      "learning_rate": 2.9767058823529413e-05,
      "loss": 0.0886,
      "step": 66
    },
    {
      "epoch": 3.88,
      "learning_rate": 2.9763529411764708e-05,
      "loss": 0.094,
      "step": 67
    },
    {
      "epoch": 3.94,
      "learning_rate": 2.976e-05,
      "loss": 0.1225,
      "step": 68
    },
    {
      "epoch": 3.99,
      "learning_rate": 2.9756470588235295e-05,
      "loss": 0.1261,
      "step": 69
    },
    {
      "epoch": 3.99,
      "eval_loss": 0.11959400027990341,
      "eval_runtime": 17.2044,
      "eval_samples_per_second": 11.218,
      "eval_steps_per_second": 1.918,
      "step": 69
    },
    {
      "epoch": 4.05,
      "learning_rate": 2.975294117647059e-05,
      "loss": 0.0772,
      "step": 70
    },
    {
      "epoch": 4.11,
      "learning_rate": 2.9749411764705885e-05,
      "loss": 0.0504,
      "step": 71
    },
    {
      "epoch": 4.17,
      "learning_rate": 2.9745882352941177e-05,
      "loss": 0.0226,
      "step": 72
    },
    {
      "epoch": 4.23,
      "learning_rate": 2.9742352941176472e-05,
      "loss": 0.0213,
      "step": 73
    },
    {
      "epoch": 4.28,
      "learning_rate": 2.9738823529411764e-05,
      "loss": 0.051,
      "step": 74
    },
    {
      "epoch": 4.34,
      "learning_rate": 2.973529411764706e-05,
      "loss": 0.0478,
      "step": 75
    },
    {
      "epoch": 4.4,
      "learning_rate": 2.9731764705882354e-05,
      "loss": 0.0219,
      "step": 76
    },
    {
      "epoch": 4.46,
      "learning_rate": 2.9728235294117646e-05,
      "loss": 0.0185,
      "step": 77
    },
    {
      "epoch": 4.51,
      "learning_rate": 2.972470588235294e-05,
      "loss": 0.0571,
      "step": 78
    },
    {
      "epoch": 4.57,
      "learning_rate": 2.9721176470588236e-05,
      "loss": 0.0488,
      "step": 79
    },
    {
      "epoch": 4.63,
      "learning_rate": 2.971764705882353e-05,
      "loss": 0.0274,
      "step": 80
    },
    {
      "epoch": 4.69,
      "learning_rate": 2.9714117647058823e-05,
      "loss": 0.0202,
      "step": 81
    },
    {
      "epoch": 4.75,
      "learning_rate": 2.9710588235294118e-05,
      "loss": 0.0779,
      "step": 82
    },
    {
      "epoch": 4.8,
      "learning_rate": 2.9707058823529413e-05,
      "loss": 0.0442,
      "step": 83
    },
    {
      "epoch": 4.86,
      "learning_rate": 2.970352941176471e-05,
      "loss": 0.022,
      "step": 84
    },
    {
      "epoch": 4.92,
      "learning_rate": 2.97e-05,
      "loss": 0.0237,
      "step": 85
    },
    {
      "epoch": 4.98,
      "learning_rate": 2.9696470588235295e-05,
      "loss": 0.0364,
      "step": 86
    },
    {
      "epoch": 4.98,
      "eval_loss": 0.1153767853975296,
      "eval_runtime": 17.2133,
      "eval_samples_per_second": 11.212,
      "eval_steps_per_second": 1.917,
      "step": 86
    }
  ],
  "logging_steps": 1,
  "max_steps": 8500,
  "num_train_epochs": 500,
  "save_steps": 500,
  "total_flos": 1.1713583065507922e+18,
  "trial_name": null,
  "trial_params": null
}