{
  "id": "scratchpad",
  "code": "PTL-0026",
  "term": "Scratchpad",
  "aliases": [
    "intermediate computation"
  ],
  "category": "reasoning",
  "definition": "A scratchpad is a region of model output reserved for intermediate computation, which the model writes before its final answer so that multi-step calculations can be carried out explicitly.",
  "description": "Nye et al. trained models to emit intermediate steps for tasks like long addition and program execution, an important precursor to chain-of-thought prompting. In practice the term also refers to private reasoning sections that are hidden from end users.",
  "example": null,
  "broader": [],
  "narrower": [],
  "related": [
    "chain-of-thought",
    "extended-thinking"
  ],
  "introduced": 2021,
  "sources": [
    {
      "title": "Show Your Work: Scratchpads for Intermediate Computation with Language Models",
      "authors": "Nye et al.",
      "year": 2021,
      "url": "https://arxiv.org/abs/2112.00114"
    }
  ],
  "url": "https://protologue.com/t/scratchpad/",
  "citation": "Protologue. (2026). Scratchpad. In Protologue: A Taxonomy of Prompting and LLM Techniques (v1.0.0, PTL-0026). https://protologue.com/t/scratchpad/"
}