{
  "best_metric": 0.5967329740524292,
  "best_model_checkpoint": "BERT-WMM/run-9/checkpoint-1066",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 1066,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.94,
      "grad_norm": 11.453248977661133,
      "learning_rate": 9.307368434290008e-06,
      "loss": 0.7373,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7469483568075117,
      "eval_loss": 0.625262975692749,
      "eval_runtime": 2.1087,
      "eval_samples_per_second": 1010.08,
      "eval_steps_per_second": 63.545,
      "step": 533
    },
    {
      "epoch": 1.88,
      "grad_norm": 8.733488082885742,
      "learning_rate": 5.072896899126219e-06,
      "loss": 0.4866,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7615023474178404,
      "eval_loss": 0.5967329740524292,
      "eval_runtime": 2.1659,
      "eval_samples_per_second": 983.446,
      "eval_steps_per_second": 61.869,
      "step": 1066
    }
  ],
  "logging_steps": 500,
  "max_steps": 1599,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 348929503378848.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 1.3541839969453797e-05
  }
}