{
  "best_metric": 0.9243119266055045,
  "best_model_checkpoint": "electra-distilled-sst\\run-49\\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": 8.770561218261719,
      "learning_rate": 4.952771397331368e-05,
      "loss": 0.3762,
      "step": 527
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.908256880733945,
      "eval_loss": 0.28651162981987,
      "eval_runtime": 1.4488,
      "eval_samples_per_second": 601.892,
      "eval_steps_per_second": 4.832,
      "step": 527
    },
    {
      "epoch": 2.0,
      "grad_norm": 11.216989517211914,
      "learning_rate": 3.714578547998526e-05,
      "loss": 0.2105,
      "step": 1054
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9048165137614679,
      "eval_loss": 0.3472459316253662,
      "eval_runtime": 1.4467,
      "eval_samples_per_second": 602.769,
      "eval_steps_per_second": 4.839,
      "step": 1054
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.215942621231079,
      "learning_rate": 2.476385698665684e-05,
      "loss": 0.1566,
      "step": 1581
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9243119266055045,
      "eval_loss": 0.30982208251953125,
      "eval_runtime": 1.448,
      "eval_samples_per_second": 602.203,
      "eval_steps_per_second": 4.834,
      "step": 1581
    }
  ],
  "logging_steps": 500,
  "max_steps": 2635,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 561038813831964.0,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.982936133975336,
    "learning_rate": 6.19096424666421e-05,
    "num_train_epochs": 5,
    "temperature": 22
  }
}