{
  "best_metric": 0.6190994381904602,
  "best_model_checkpoint": "hBERTv2_new_pretrain_w_init_48_cola/checkpoint-268",
  "epoch": 9.0,
  "global_step": 603,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 3.9200000000000004e-05,
      "loss": 0.6301,
      "step": 67
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6292892098426819,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 1.8283,
      "eval_samples_per_second": 570.489,
      "eval_steps_per_second": 4.923,
      "step": 67
    },
    {
      "epoch": 2.0,
      "learning_rate": 3.8400000000000005e-05,
      "loss": 0.6238,
      "step": 134
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6254382133483887,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 1.8394,
      "eval_samples_per_second": 567.023,
      "eval_steps_per_second": 4.893,
      "step": 134
    },
    {
      "epoch": 3.0,
      "learning_rate": 3.76e-05,
      "loss": 0.6072,
      "step": 201
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6759347915649414,
      "eval_loss": 0.627076268196106,
      "eval_matthews_correlation": 0.033871931941817014,
      "eval_runtime": 1.8955,
      "eval_samples_per_second": 550.253,
      "eval_steps_per_second": 4.748,
      "step": 201
    },
    {
      "epoch": 4.0,
      "learning_rate": 3.680000000000001e-05,
      "loss": 0.5821,
      "step": 268
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6836050152778625,
      "eval_loss": 0.6190994381904602,
      "eval_matthews_correlation": 0.08208497144404353,
      "eval_runtime": 1.8857,
      "eval_samples_per_second": 553.096,
      "eval_steps_per_second": 4.773,
      "step": 268
    },
    {
      "epoch": 5.0,
      "learning_rate": 3.6e-05,
      "loss": 0.5262,
      "step": 335
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6510066986083984,
      "eval_loss": 0.7056761980056763,
      "eval_matthews_correlation": 0.11510140889517931,
      "eval_runtime": 1.8625,
      "eval_samples_per_second": 559.989,
      "eval_steps_per_second": 4.832,
      "step": 335
    },
    {
      "epoch": 6.0,
      "learning_rate": 3.52e-05,
      "loss": 0.4735,
      "step": 402
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6577181220054626,
      "eval_loss": 0.6755900382995605,
      "eval_matthews_correlation": 0.11809845732805219,
      "eval_runtime": 1.853,
      "eval_samples_per_second": 562.876,
      "eval_steps_per_second": 4.857,
      "step": 402
    },
    {
      "epoch": 7.0,
      "learning_rate": 3.44e-05,
      "loss": 0.4127,
      "step": 469
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6711409687995911,
      "eval_loss": 0.8493370413780212,
      "eval_matthews_correlation": 0.1229288843319562,
      "eval_runtime": 1.848,
      "eval_samples_per_second": 564.387,
      "eval_steps_per_second": 4.87,
      "step": 469
    },
    {
      "epoch": 8.0,
      "learning_rate": 3.3600000000000004e-05,
      "loss": 0.349,
      "step": 536
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6232023239135742,
      "eval_loss": 0.8918721675872803,
      "eval_matthews_correlation": 0.1434071727865922,
      "eval_runtime": 1.8365,
      "eval_samples_per_second": 567.926,
      "eval_steps_per_second": 4.901,
      "step": 536
    },
    {
      "epoch": 9.0,
      "learning_rate": 3.28e-05,
      "loss": 0.311,
      "step": 603
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.6768935918807983,
      "eval_loss": 0.9018412828445435,
      "eval_matthews_correlation": 0.13982593577610886,
      "eval_runtime": 1.8388,
      "eval_samples_per_second": 567.228,
      "eval_steps_per_second": 4.895,
      "step": 603
    },
    {
      "epoch": 9.0,
      "step": 603,
      "total_flos": 1.1262116140941312e+16,
      "train_loss": 0.5017338842894901,
      "train_runtime": 462.8664,
      "train_samples_per_second": 923.701,
      "train_steps_per_second": 7.238
    }
  ],
  "max_steps": 3350,
  "num_train_epochs": 50,
  "total_flos": 1.1262116140941312e+16,
  "trial_name": null,
  "trial_params": null
}