Skip to content

Instantly share code, notes, and snippets.

@mark-wiemer
Last active April 15, 2023 21:56
Show Gist options
  • Save mark-wiemer/f245014a1ff63a815d298af4e92e2d99 to your computer and use it in GitHub Desktop.
Save mark-wiemer/f245014a1ff63a815d298af4e92e2d99 to your computer and use it in GitHub Desktop.
Prompt engineering samples

Prompt engineering article (ChatGPT Mar 23 Version, GPT-3.5)

Define prompt engineering

Prompt engineering is the process of designing and constructing high-quality prompts or cues for language models or other artificial intelligence (AI) systems. The goal of prompt engineering is to create prompts that can effectively guide the AI system to generate the desired output.

Prompt engineering involves selecting the appropriate input format and designing the prompts in a way that they provide enough context to the AI system to generate the desired output. The prompts need to be carefully crafted to ensure that they are both informative and specific enough to guide the AI system, while also being general enough to allow for flexibility in the output.

Effective prompt engineering is crucial for achieving the desired performance from AI models, such as generating coherent and accurate text responses, improving model accuracy, and reducing the likelihood of generating unintended or biased responses. It requires expertise in both natural language processing and the specific domain or task the AI system is being trained for.

Define prompt engineering in 20 words

Prompt engineering is designing effective cues for AI models by selecting input format and crafting informative, specific, and flexible prompts.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment