{
  "best_metric": 0.5207920792079208,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-1/checkpoint-475",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 2850,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.4949035942554474,
      "learning_rate": 1.755924298388821e-05,
      "loss": 0.2107,
      "step": 475
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5207920792079208,
      "eval_f1": 0.1597222222222222,
      "eval_loss": 0.19466964900493622,
      "eval_precision": 0.6388888888888888,
      "eval_recall": 0.09126984126984126,
      "eval_runtime": 15.6911,
      "eval_samples_per_second": 32.184,
      "eval_steps_per_second": 1.02,
      "step": 475
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.5680577754974365,
      "learning_rate": 1.4047394387110566e-05,
      "loss": 0.1957,
      "step": 950
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5168316831683168,
      "eval_f1": 0.14685314685314682,
      "eval_loss": 0.19141286611557007,
      "eval_precision": 0.6176470588235294,
      "eval_recall": 0.08333333333333333,
      "eval_runtime": 15.1911,
      "eval_samples_per_second": 33.243,
      "eval_steps_per_second": 1.053,
      "step": 950
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.45545563101768494,
      "learning_rate": 1.0535545790332925e-05,
      "loss": 0.1943,
      "step": 1425
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5148514851485149,
      "eval_f1": 0.14634146341463414,
      "eval_loss": 0.1912413090467453,
      "eval_precision": 0.6,
      "eval_recall": 0.08333333333333333,
      "eval_runtime": 15.2913,
      "eval_samples_per_second": 33.025,
      "eval_steps_per_second": 1.046,
      "step": 1425
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.33228880167007446,
      "learning_rate": 7.023697193555283e-06,
      "loss": 0.1936,
      "step": 1900
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5168316831683168,
      "eval_f1": 0.14685314685314682,
      "eval_loss": 0.190897598862648,
      "eval_precision": 0.6176470588235294,
      "eval_recall": 0.08333333333333333,
      "eval_runtime": 15.1905,
      "eval_samples_per_second": 33.244,
      "eval_steps_per_second": 1.053,
      "step": 1900
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.3122839331626892,
      "learning_rate": 3.5118485967776416e-06,
      "loss": 0.1933,
      "step": 2375
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5148514851485149,
      "eval_f1": 0.14634146341463414,
      "eval_loss": 0.19050733745098114,
      "eval_precision": 0.6,
      "eval_recall": 0.08333333333333333,
      "eval_runtime": 15.1983,
      "eval_samples_per_second": 33.227,
      "eval_steps_per_second": 1.053,
      "step": 2375
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.29782089591026306,
      "learning_rate": 0.0,
      "loss": 0.193,
      "step": 2850
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5128712871287129,
      "eval_f1": 0.13986013986013984,
      "eval_loss": 0.19041746854782104,
      "eval_precision": 0.5882352941176471,
      "eval_recall": 0.07936507936507936,
      "eval_runtime": 15.2095,
      "eval_samples_per_second": 33.203,
      "eval_steps_per_second": 1.052,
      "step": 2850
    }
  ],
  "logging_steps": 500,
  "max_steps": 2850,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 500,
  "total_flos": 4667059698840.0,
  "train_batch_size": 20,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.22818780511125447,
    "learning_rate": 2.107109158066585e-05,
    "num_train_epochs": 6,
    "per_device_train_batch_size": 20,
    "temperature": 6
  }
}