{
  "best_metric": 0.9174311926605505,
  "best_model_checkpoint": "electra-distilled-sst\\run-26\\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": 25.633333206176758,
      "learning_rate": 5.249775471708889e-05,
      "loss": 0.9686,
      "step": 527
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9105504587155964,
      "eval_loss": 0.6331567168235779,
      "eval_runtime": 1.4527,
      "eval_samples_per_second": 600.271,
      "eval_steps_per_second": 4.819,
      "step": 527
    },
    {
      "epoch": 2.0,
      "grad_norm": 32.01615905761719,
      "learning_rate": 4.666590043043083e-05,
      "loss": 0.4917,
      "step": 1054
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.911697247706422,
      "eval_loss": 0.6638227105140686,
      "eval_runtime": 1.4467,
      "eval_samples_per_second": 602.739,
      "eval_steps_per_second": 4.838,
      "step": 1054
    },
    {
      "epoch": 3.0,
      "grad_norm": 19.456010818481445,
      "learning_rate": 4.0834046143772765e-05,
      "loss": 0.3597,
      "step": 1581
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9174311926605505,
      "eval_loss": 0.6752657294273376,
      "eval_runtime": 1.4552,
      "eval_samples_per_second": 599.222,
      "eval_steps_per_second": 4.81,
      "step": 1581
    }
  ],
  "logging_steps": 500,
  "max_steps": 5270,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 561038813831964.0,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.8150986578770638,
    "learning_rate": 5.831854286658062e-05,
    "num_train_epochs": 10,
    "temperature": 30
  }
}