{
  "best_metric": 0.8302752293577982,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-0/checkpoint-1054",
  "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": 14.515488624572754,
      "learning_rate": 0.00014545600250112875,
      "loss": 1.8614,
      "step": 527
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8176605504587156,
      "eval_loss": 1.578995704650879,
      "eval_runtime": 2.8296,
      "eval_samples_per_second": 308.168,
      "eval_steps_per_second": 2.474,
      "step": 527
    },
    {
      "epoch": 2.0,
      "grad_norm": 19.460262298583984,
      "learning_rate": 0.00012121333541760731,
      "loss": 0.862,
      "step": 1054
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8302752293577982,
      "eval_loss": 1.6238842010498047,
      "eval_runtime": 2.8061,
      "eval_samples_per_second": 310.75,
      "eval_steps_per_second": 2.495,
      "step": 1054
    }
  ],
  "logging_steps": 500,
  "max_steps": 3689,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "total_flos": 16141525436100.0,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.24080905069528658,
    "learning_rate": 0.00016969866958465023,
    "num_train_epochs": 7,
    "temperature": 25
  }
}