{
  "best_global_step": 39,
  "best_metric": 0.6173285198555957,
  "best_model_checkpoint": "bert-base-uncased-finetuned-rte-run_14/run-0/checkpoint-39",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 39,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2564102564102564,
      "grad_norm": 1.662625789642334,
      "learning_rate": 9.487179487179487e-05,
      "loss": 0.696,
      "step": 10
    },
    {
      "epoch": 0.5128205128205128,
      "grad_norm": 2.0300142765045166,
      "learning_rate": 8.974358974358975e-05,
      "loss": 0.6793,
      "step": 20
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 4.4956440925598145,
      "learning_rate": 8.461538461538461e-05,
      "loss": 0.6499,
      "step": 30
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6173285198555957,
      "eval_loss": 0.6330550909042358,
      "eval_runtime": 0.6611,
      "eval_samples_per_second": 418.98,
      "eval_steps_per_second": 7.563,
      "step": 39
    }
  ],
  "logging_steps": 10,
  "max_steps": 195,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 194932403139840.0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": {
    "dropout_rate": 0.01,
    "learning_rate": 0.0001,
    "max_length": 32,
    "num_train_epochs": 5,
    "per_device_train_batch_size": 64
  }
}