{
  "best_metric": 70.84639498432603,
  "best_model_checkpoint": "outputs/bitfit/t5-base/superglue-wic/checkpoint-1700",
  "epoch": 20.0,
  "global_step": 3400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.59,
      "eval_accuracy": 64.57680250783699,
      "eval_average_metrics": 64.57680250783699,
      "eval_loss": 0.2697466313838959,
      "eval_runtime": 1.3478,
      "eval_samples_per_second": 236.682,
      "step": 100
    },
    {
      "epoch": 1.18,
      "eval_accuracy": 65.51724137931035,
      "eval_average_metrics": 65.51724137931035,
      "eval_loss": 0.3369597792625427,
      "eval_runtime": 1.5122,
      "eval_samples_per_second": 210.954,
      "step": 200
    },
    {
      "epoch": 1.76,
      "eval_accuracy": 67.39811912225704,
      "eval_average_metrics": 67.39811912225704,
      "eval_loss": 0.29254111647605896,
      "eval_runtime": 1.6714,
      "eval_samples_per_second": 190.86,
      "step": 300
    },
    {
      "epoch": 2.35,
      "eval_accuracy": 65.51724137931035,
      "eval_average_metrics": 65.51724137931035,
      "eval_loss": 0.30159813165664673,
      "eval_runtime": 1.399,
      "eval_samples_per_second": 228.019,
      "step": 400
    },
    {
      "epoch": 2.94,
      "learning_rate": 0.0002558823529411764,
      "loss": 0.2549,
      "step": 500
    },
    {
      "epoch": 2.94,
      "eval_accuracy": 69.90595611285266,
      "eval_average_metrics": 69.90595611285266,
      "eval_loss": 0.30272728204727173,
      "eval_runtime": 1.7009,
      "eval_samples_per_second": 187.551,
      "step": 500
    },
    {
      "epoch": 3.53,
      "eval_accuracy": 67.39811912225704,
      "eval_average_metrics": 67.39811912225704,
      "eval_loss": 0.3448639512062073,
      "eval_runtime": 1.6869,
      "eval_samples_per_second": 189.1,
      "step": 600
    },
    {
      "epoch": 4.12,
      "eval_accuracy": 69.27899686520375,
      "eval_average_metrics": 69.27899686520375,
      "eval_loss": 0.37638065218925476,
      "eval_runtime": 1.3564,
      "eval_samples_per_second": 235.188,
      "step": 700
    },
    {
      "epoch": 4.71,
      "eval_accuracy": 68.02507836990596,
      "eval_average_metrics": 68.02507836990596,
      "eval_loss": 0.3278528153896332,
      "eval_runtime": 1.436,
      "eval_samples_per_second": 222.137,
      "step": 800
    },
    {
      "epoch": 5.29,
      "eval_accuracy": 65.8307210031348,
      "eval_average_metrics": 65.8307210031348,
      "eval_loss": 0.37711796164512634,
      "eval_runtime": 1.7367,
      "eval_samples_per_second": 183.683,
      "step": 900
    },
    {
      "epoch": 5.88,
      "learning_rate": 0.00021176470588235295,
      "loss": 0.1531,
      "step": 1000
    },
    {
      "epoch": 5.88,
      "eval_accuracy": 68.96551724137932,
      "eval_average_metrics": 68.96551724137932,
      "eval_loss": 0.37187883257865906,
      "eval_runtime": 1.5128,
      "eval_samples_per_second": 210.871,
      "step": 1000
    },
    {
      "epoch": 6.47,
      "eval_accuracy": 66.14420062695925,
      "eval_average_metrics": 66.14420062695925,
      "eval_loss": 0.4265880882740021,
      "eval_runtime": 1.6439,
      "eval_samples_per_second": 194.053,
      "step": 1100
    },
    {
      "epoch": 7.06,
      "eval_accuracy": 67.0846394984326,
      "eval_average_metrics": 67.0846394984326,
      "eval_loss": 0.4190385937690735,
      "eval_runtime": 1.694,
      "eval_samples_per_second": 188.316,
      "step": 1200
    },
    {
      "epoch": 7.65,
      "eval_accuracy": 67.7115987460815,
      "eval_average_metrics": 67.7115987460815,
      "eval_loss": 0.4279927611351013,
      "eval_runtime": 1.6705,
      "eval_samples_per_second": 190.965,
      "step": 1300
    },
    {
      "epoch": 8.24,
      "eval_accuracy": 68.96551724137932,
      "eval_average_metrics": 68.96551724137932,
      "eval_loss": 0.4631330370903015,
      "eval_runtime": 1.6505,
      "eval_samples_per_second": 193.272,
      "step": 1400
    },
    {
      "epoch": 8.82,
      "learning_rate": 0.0001676470588235294,
      "loss": 0.0995,
      "step": 1500
    },
    {
      "epoch": 8.82,
      "eval_accuracy": 67.7115987460815,
      "eval_average_metrics": 67.7115987460815,
      "eval_loss": 0.46376463770866394,
      "eval_runtime": 1.4266,
      "eval_samples_per_second": 223.612,
      "step": 1500
    },
    {
      "epoch": 9.41,
      "eval_accuracy": 68.96551724137932,
      "eval_average_metrics": 68.96551724137932,
      "eval_loss": 0.509791374206543,
      "eval_runtime": 1.5279,
      "eval_samples_per_second": 208.785,
      "step": 1600
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 70.84639498432603,
      "eval_average_metrics": 70.84639498432603,
      "eval_loss": 0.4690648913383484,
      "eval_runtime": 1.3636,
      "eval_samples_per_second": 233.934,
      "step": 1700
    },
    {
      "epoch": 10.59,
      "eval_accuracy": 69.27899686520375,
      "eval_average_metrics": 69.27899686520375,
      "eval_loss": 0.5513712763786316,
      "eval_runtime": 1.6761,
      "eval_samples_per_second": 190.323,
      "step": 1800
    },
    {
      "epoch": 11.18,
      "eval_accuracy": 68.65203761755487,
      "eval_average_metrics": 68.65203761755487,
      "eval_loss": 0.5640075206756592,
      "eval_runtime": 1.498,
      "eval_samples_per_second": 212.948,
      "step": 1900
    },
    {
      "epoch": 11.76,
      "learning_rate": 0.00012352941176470587,
      "loss": 0.0627,
      "step": 2000
    },
    {
      "epoch": 11.76,
      "eval_accuracy": 68.3385579937304,
      "eval_average_metrics": 68.3385579937304,
      "eval_loss": 0.5400956273078918,
      "eval_runtime": 1.7505,
      "eval_samples_per_second": 182.237,
      "step": 2000
    },
    {
      "epoch": 12.35,
      "eval_accuracy": 66.77115987460816,
      "eval_average_metrics": 66.77115987460816,
      "eval_loss": 0.6046501994132996,
      "eval_runtime": 1.6979,
      "eval_samples_per_second": 187.882,
      "step": 2100
    },
    {
      "epoch": 12.94,
      "eval_accuracy": 70.53291536050156,
      "eval_average_metrics": 70.53291536050156,
      "eval_loss": 0.6463834047317505,
      "eval_runtime": 1.3771,
      "eval_samples_per_second": 231.652,
      "step": 2200
    },
    {
      "epoch": 13.53,
      "eval_accuracy": 69.27899686520375,
      "eval_average_metrics": 69.27899686520375,
      "eval_loss": 0.6423152089118958,
      "eval_runtime": 1.6672,
      "eval_samples_per_second": 191.343,
      "step": 2300
    },
    {
      "epoch": 14.12,
      "eval_accuracy": 68.96551724137932,
      "eval_average_metrics": 68.96551724137932,
      "eval_loss": 0.6988397240638733,
      "eval_runtime": 1.6913,
      "eval_samples_per_second": 188.613,
      "step": 2400
    },
    {
      "epoch": 14.71,
      "learning_rate": 7.941176470588235e-05,
      "loss": 0.037,
      "step": 2500
    },
    {
      "epoch": 14.71,
      "eval_accuracy": 68.3385579937304,
      "eval_average_metrics": 68.3385579937304,
      "eval_loss": 0.7036728858947754,
      "eval_runtime": 1.7114,
      "eval_samples_per_second": 186.402,
      "step": 2500
    },
    {
      "epoch": 15.29,
      "eval_accuracy": 68.65203761755487,
      "eval_average_metrics": 68.65203761755487,
      "eval_loss": 0.6903794407844543,
      "eval_runtime": 1.6091,
      "eval_samples_per_second": 198.248,
      "step": 2600
    },
    {
      "epoch": 15.88,
      "eval_accuracy": 68.96551724137932,
      "eval_average_metrics": 68.96551724137932,
      "eval_loss": 0.7062700390815735,
      "eval_runtime": 1.7338,
      "eval_samples_per_second": 183.989,
      "step": 2700
    },
    {
      "epoch": 16.47,
      "eval_accuracy": 70.21943573667711,
      "eval_average_metrics": 70.21943573667711,
      "eval_loss": 0.7178221940994263,
      "eval_runtime": 1.3196,
      "eval_samples_per_second": 241.74,
      "step": 2800
    },
    {
      "epoch": 17.06,
      "eval_accuracy": 68.96551724137932,
      "eval_average_metrics": 68.96551724137932,
      "eval_loss": 0.7901595234870911,
      "eval_runtime": 1.6693,
      "eval_samples_per_second": 191.103,
      "step": 2900
    },
    {
      "epoch": 17.65,
      "learning_rate": 3.529411764705882e-05,
      "loss": 0.0252,
      "step": 3000
    },
    {
      "epoch": 17.65,
      "eval_accuracy": 69.27899686520375,
      "eval_average_metrics": 69.27899686520375,
      "eval_loss": 0.8428929448127747,
      "eval_runtime": 1.6724,
      "eval_samples_per_second": 190.74,
      "step": 3000
    },
    {
      "epoch": 18.24,
      "eval_accuracy": 68.96551724137932,
      "eval_average_metrics": 68.96551724137932,
      "eval_loss": 0.824032187461853,
      "eval_runtime": 1.1559,
      "eval_samples_per_second": 275.979,
      "step": 3100
    },
    {
      "epoch": 18.82,
      "eval_accuracy": 69.27899686520375,
      "eval_average_metrics": 69.27899686520375,
      "eval_loss": 0.8500473499298096,
      "eval_runtime": 1.7687,
      "eval_samples_per_second": 180.362,
      "step": 3200
    },
    {
      "epoch": 19.41,
      "eval_accuracy": 68.3385579937304,
      "eval_average_metrics": 68.3385579937304,
      "eval_loss": 0.8623896241188049,
      "eval_runtime": 1.7067,
      "eval_samples_per_second": 186.905,
      "step": 3300
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 68.3385579937304,
      "eval_average_metrics": 68.3385579937304,
      "eval_loss": 0.8675826787948608,
      "eval_runtime": 1.7047,
      "eval_samples_per_second": 187.128,
      "step": 3400
    },
    {
      "epoch": 20.0,
      "step": 3400,
      "total_flos": 7227776409587712.0,
      "train_loss": 0.0949506963000578,
      "train_runtime": 1454.9221,
      "train_samples_per_second": 74.616,
      "train_steps_per_second": 2.337
    }
  ],
  "max_steps": 3400,
  "num_train_epochs": 20,
  "total_flos": 7227776409587712.0,
  "trial_name": null,
  "trial_params": null
}