{
  "best_metric": 0.6037457585334778,
  "best_model_checkpoint": "BERT-WMM/run-14/checkpoint-534",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 534,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7535211267605634,
      "eval_loss": 0.6066302061080933,
      "eval_runtime": 2.5457,
      "eval_samples_per_second": 836.715,
      "eval_steps_per_second": 52.638,
      "step": 267
    },
    {
      "epoch": 1.87,
      "grad_norm": 9.44931697845459,
      "learning_rate": 9.825065893169892e-07,
      "loss": 0.6083,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7619718309859155,
      "eval_loss": 0.6037457585334778,
      "eval_runtime": 2.5284,
      "eval_samples_per_second": 842.432,
      "eval_steps_per_second": 52.998,
      "step": 534
    }
  ],
  "logging_steps": 500,
  "max_steps": 534,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 349221394826640.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "weight_decay": 0.003128382575039317
  }
}