https://llm-utils.org/AI+Rankings+-+June+2023 Let’s work this out in a step by step way to be sure we have the right answer. Source
“You are a helpful assistant. Always answer with the bottom line up front - give the answer first, then if any explanation is required, give it later. No explanations or disclaimers before the answer, explanations only after the answer. Minimize disclaimers. Be concise. Minimize extra explanations, the user will ask for more info if needed.”
This is my personal system prompt.
https://llm-utils.org/My+template+for+a+semi-complex+GPT-4+prompt
https://help.openai.com/en/articles/6654000-best-practices-for-prompt-engineering-with-openai-api
- Put instructions at the beginning of the prompt and use or """ to separate the instruction and context
- Be specific, descriptive and as detailed as possible about the desired context, outcome, length, format, style, etc
- Articulate the desired output format through examples (example 1, example 2)
- Start with zero-shot, then few-shot (example), neither of them worked, then fine-tune
- Reduce “fluffy” and imprecise descriptions
- Instead of just saying what not to do, say what to do instead
- Code Generation Specific - Use “leading words” to nudge the model toward a particular pattern
- For most factual use cases such as data extraction, and truthful Q&A, the temperature of 0 is best.
See also: https://platform.openai.com/docs/guides/gpt-best-practices https://github.com/openai/openai-cookbook/tree/main/examples
White, Jules, Quchen Fu, Sam Hays, Michael Sandborn, Carlos Olea, Henry Gilbert, Ashraf Elnashar, Jesse Spencer-Smith, and Douglas C. Schmidt. “A Prompt Pattern Catalog to Enhance Prompt Engineering with ChatGPT.” arXiv, February 21, 2023. https://doi.org/10.48550/arXiv.2302.11382.
Zhou, Yongchao, Andrei Ioan Muresanu, Ziwen Han, Keiran Paster, Silviu Pitis, Harris Chan, and Jimmy Ba. “Large Language Models Are Human-Level Prompt Engineers.” arXiv, March 10, 2023. https://doi.org/10.48550/arXiv.2211.01910.
Zamfirescu-Pereira, J.D., Richmond Y. Wong, Bjoern Hartmann, and Qian Yang. “Why Johnny Can’t Prompt: How Non-AI Experts Try (and Fail) to Design LLM Prompts.” In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems, 1–21. CHI ’23. New York, NY, USA: Association for Computing Machinery, 2023. https://doi.org/10.1145/3544548.3581388.
Kojima, Takeshi, Shixiang Shane Gu, Machel Reid, Yutaka Matsuo, and Yusuke Iwasawa. “Large Language Models Are Zero-Shot Reasoners.” arXiv, January 29, 2023. https://doi.org/10.48550/arXiv.2205.11916.
Zhou, Denny, Nathanael Schärli, Le Hou, Jason Wei, Nathan Scales, Xuezhi Wang, Dale Schuurmans, et al. “Least-to-Most Prompting Enables Complex Reasoning in Large Language Models,” 2022. https://openreview.net/forum?id=WZH7099tgfM.
Ph.D, Cameron R. Wolfe. “Practical Prompt Engineering.” Substack newsletter. Deep (Learning) Focus (blog), May 1, 2023. https://cameronrwolfe.substack.com/p/practical-prompt-engineering-part.
Ph.D, Cameron R. Wolfe. “Advanced Prompt Engineering.” Substack newsletter. Deep (Learning) Focus (blog), May 8, 2023. https://cameronrwolfe.substack.com/p/advanced-prompt-engineering.