{
  "best_metric": 0.9208715596330275,
  "best_model_checkpoint": "electra-distilled-sst\\run-11\\checkpoint-1581",
  "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": 14.338537216186523,
      "learning_rate": 5.0174854029835415e-05,
      "loss": 0.5621,
      "step": 527
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9071100917431193,
      "eval_loss": 0.39338329434394836,
      "eval_runtime": 1.4401,
      "eval_samples_per_second": 605.519,
      "eval_steps_per_second": 4.861,
      "step": 527
    },
    {
      "epoch": 2.0,
      "grad_norm": 21.374542236328125,
      "learning_rate": 4.302061894724029e-05,
      "loss": 0.3067,
      "step": 1054
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9185779816513762,
      "eval_loss": 0.45042580366134644,
      "eval_runtime": 1.461,
      "eval_samples_per_second": 596.848,
      "eval_steps_per_second": 4.791,
      "step": 1054
    },
    {
      "epoch": 3.0,
      "grad_norm": 5.032501697540283,
      "learning_rate": 3.5852782657263795e-05,
      "loss": 0.2271,
      "step": 1581
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9208715596330275,
      "eval_loss": 0.42918458580970764,
      "eval_runtime": 1.4441,
      "eval_samples_per_second": 603.85,
      "eval_steps_per_second": 4.847,
      "step": 1581
    }
  ],
  "logging_steps": 500,
  "max_steps": 4216,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "total_flos": 561038813831964.0,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9346338538228292,
    "learning_rate": 5.73426903198119e-05,
    "num_train_epochs": 8,
    "temperature": 23
  }
}