An experiment based on prompts from the awesome ChatGPT prompts GitHub repo. Which performs better, asking a simple question to ChatGPT? Or prefacing the question with a verbose prompt?
“Machine Learning” Test Question sourced from the repository:
I have a dataset without labels. Which machine learning algorithm should I use?
Full prompt from the repository:
I want you to act as a machine learning engineer. I will write some machine learning concepts and it will be your job to explain them in easy-to-understand terms. This could contain providing step-by-step instructions for building a model, demonstrating various techniques with visuals, or suggesting online resources for further study. My first suggestion request is "I have a dataset without labels. Which machine learning algorithm should I use?"
Test using just the question with ChatGPT:
Test using the full prompt from the GitHub repository:
“Extreme TLDR” to compare the content of two responses
Simplify the following to an "Extreme TLDR" that only contains the most essential elements to gain an understanding of the contents:
Just the question
Full prompt from the repository
Results
The “full prompt” from the repository is WORSE than just asking the question itself. The “full prompt” response seems unnecessarily verbose. The “Extreme TLDR” test shows that the essence of the response is nearly identical to asking the question by itself.