{
  "best_metric": 0.8371559633027523,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-38/checkpoint-527",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 1054,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 6.650190353393555,
      "learning_rate": 0.0003977381702368286,
      "loss": 0.8907,
      "step": 527
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8371559633027523,
      "eval_loss": 1.0675474405288696,
      "eval_runtime": 2.812,
      "eval_samples_per_second": 310.097,
      "eval_steps_per_second": 2.489,
      "step": 527
    },
    {
      "epoch": 2.0,
      "grad_norm": 7.2698974609375,
      "learning_rate": 0.0003181905361894629,
      "loss": 0.4134,
      "step": 1054
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8084862385321101,
      "eval_loss": 1.2476831674575806,
      "eval_runtime": 2.8048,
      "eval_samples_per_second": 310.9,
      "eval_steps_per_second": 2.496,
      "step": 1054
    }
  ],
  "logging_steps": 500,
  "max_steps": 3162,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 500,
  "total_flos": 16141525436100.0,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.63744345806538,
    "learning_rate": 0.00047728580428419435,
    "num_train_epochs": 6,
    "temperature": 29
  }
}