Model responds normally despite prompt set to "answer only with questions"

[naikmubashir]

Select Topic Area

Bug

Body

Model responds normally despite prompt set to "answer only with questions"

I encountered an issue with the model where I set the prompt to "answer only with questions," but it didn’t follow that instruction. Instead of responding with only questions as expected, the model answered normally, which seems like a bug in the behavior.

Has anyone else experienced this? Could this be a limitation in how prompts are interpreted? Any thoughts on why the model didn’t adhere to the prompt?

[github-actions[bot]]

💬 Your Product Feedback Has Been Submitted 🎉

Thank you for taking the time to share your insights with us! Your feedback is invaluable as we build a better GitHub experience for all our users.

Here's what you can expect moving forward ⏩

• Your input will be carefully reviewed and cataloged by members of our product teams. 
  • Due to the high volume of submissions, we may not always be able to provide individual responses.
  • Rest assured, your feedback will help chart our course for product improvements.
• Other users may engage with your post, sharing their own perspectives or experiences. 
• GitHub staff may reach out for further clarification or insight. 
  • We may 'Answer' your discussion if there is a current solution, workaround, or roadmap/changelog post related to the feedback.

Where to look to see what's shipping 👀

• Read the Changelog for real-time updates on the latest GitHub features, enhancements, and calls for feedback.
• Explore our Product Roadmap, which details upcoming major releases and initiatives.

What you can do in the meantime 💻

• Upvote and comment on other user feedback Discussions that resonate with you.
• Add more information at any point! Useful details include: use cases, relevant labels, desired outcomes, and any accompanying screenshots.

As a member of the GitHub community, your participation is essential. While we can't promise that every suggestion will be implemented, we want to emphasize that your feedback is instrumental in guiding our decisions and priorities.

Thank you once again for your contribution to making GitHub even better! We're grateful for your ongoing support and collaboration in shaping the future of our platform. ⭐

[ogulcanaydogan]

It seems that the model is not strictly adhering to the prompt due to how instructions are interpreted. Models can be designed to process context and generate responses, but they're not always perfect at following highly specific or restrictive directives like "only respond with questions." This behavior might stem from the model's natural tendency to provide a complete answer or due to its training data, which emphasizes informative responses.

It's possible that this is a limitation of prompt control, where instructions like this are interpreted as suggestions rather than absolute rules. You could experiment with more explicit phrasing or system instructions, but this issue may arise due to inherent constraints in how the model processes language.