{
  "best_metric": 0.9151376146788991,
  "best_model_checkpoint": "electra-distilled-sst\\run-0\\checkpoint-527",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 1581,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 32.83808898925781,
      "learning_rate": 5.9109468466799715e-05,
      "loss": 1.4461,
      "step": 527
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9151376146788991,
      "eval_loss": 0.8842347860336304,
      "eval_runtime": 1.4618,
      "eval_samples_per_second": 596.511,
      "eval_steps_per_second": 4.789,
      "step": 527
    },
    {
      "epoch": 2.0,
      "grad_norm": 52.87685775756836,
      "learning_rate": 2.96107091088419e-05,
      "loss": 0.7,
      "step": 1054
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8967889908256881,
      "eval_loss": 1.0896263122558594,
      "eval_runtime": 1.4481,
      "eval_samples_per_second": 602.175,
      "eval_steps_per_second": 4.834,
      "step": 1054
    },
    {
      "epoch": 3.0,
      "grad_norm": 13.138901710510254,
      "learning_rate": 1.1194975088409036e-07,
      "loss": 0.494,
      "step": 1581
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9151376146788991,
      "eval_loss": 1.0067952871322632,
      "eval_runtime": 1.4498,
      "eval_samples_per_second": 601.446,
      "eval_steps_per_second": 4.828,
      "step": 1581
    }
  ],
  "logging_steps": 500,
  "max_steps": 1581,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 561038813831964.0,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.1559830125797934,
    "learning_rate": 8.849627807387343e-05,
    "num_train_epochs": 3,
    "temperature": 3
  }
}