Author(s): Mia Dwyer

Originally published on Towards AI.

Fine-Tuning AI Responses with Savvy Prompt Engineering

I don’t know about you, but as much as I love ChatGPT, specifically GPT 4, one thing I cannot stand is the tone in which it writes, and its habit of inundating me with reams of text when I ask a basic question. In this article, we’ll be going over some tips + tricks I’ve found to be invaluable for dramatically improving the quality of output from Large Language Models.

Image created by me, using DALL-E.

The following additions to the end of your prompt can dramatically change the output of the LLM and shows how you can leverage prompt engineering to finesse the type of responses you get.

Here we can see how this approach to the prompt impacts the length and depth of the response, which in many contexts might be more what we want as a response, especially on a new topic. I regularly use this when I’m asking ChatGPT to explain concepts to me or write a short paragraph.

Note the significant difference in output between the original prompt (left) and with the addition of “no yapping” (right)

With the addition of “please don’t overwhelm me” to our query “Can you explain what RAG is in AI” we are showcasing how… Read the full blog for free on Medium.

Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming a sponsor.

Published via Towards AI