This repository provides a comprehensive, continuously updated guide to prompt engineering for large language models (LLMs). It targets researchers and developers seeking to understand and leverage LLMs for various applications, offering a curated collection of papers, tutorials, lectures, and tools to improve LLM performance and interaction design.

How It Works

The guide systematically breaks down prompt engineering into foundational and advanced concepts, exploring applications, specific models like ChatGPT, adversarial techniques, and reliability. It emphasizes practical application through lectures with accompanying code notebooks and slides, aiming to demystify LLM capabilities and limitations.

Quick Start & Requirements

• Access to the guide is primarily through the GitHub repository and a new web version.
• No specific software installation is required to read the guide.

Highlighted Details

• Comprehensive coverage of prompt engineering topics, including introductions, basics, advanced techniques, applications, ChatGPT, adversarial prompts, reliability, and miscellaneous subjects.
• Curated lists of relevant papers, tools, and datasets for prompt engineering.
• Includes a lecture with video, code notebooks, and slides detailing prompt techniques, applications, and tools.
• Actively maintained with updates and new content, including a new course on prompt engineering for LLMs.

Maintenance & Community

• The project is actively maintained and welcomes contributions via Pull Requests.
• Community engagement is encouraged through GitHub Issues and a Discord server.
• Follows on Twitter and offers a newsletter subscription for updates.

Licensing & Compatibility

• The repository content is generally available for use, with a specific citation format provided for academic or professional work.
• No explicit license is stated in the README, but the nature of the content suggests it is for informational and educational purposes.

Limitations & Caveats

The guide is a work in progress, with sections still under development. While comprehensive, it may not cover every niche aspect of prompt engineering, and users are encouraged to contribute missing information.