Same prompt, same endpoint, different score | Hacker News - But Better

Same prompt, same endpoint, different score codecut.ai

Learn how prompt evaluation noise can make a small prompt tweak look better than it is, even with temperature 0 and provider pinning.

1 points•khuyentran1401•9 days ago•0 comments•

#prompt #endpoint #score #different #different score #tweak #evaluation #noise

0 comments

No comments yet.

Related stories

Show HN: Noise – Static Noise for Focusingnoise.ryan-schachte.com
Hacker News · 1 points · 1 day ago
Prompt Wars – An AI riff on Core Wars to prompt-inject each otherpromptwars.party
Hacker News · 2 points · 10 days ago
What Is a Parser Differential? How Can the Same Input Mean Different Things to Different Systems?dev.to
DEV Community · 1 points · 12 days ago
Bose’s first wired earbuds in 11 years add noise cancelingtheverge.com
The Verge · 0 points · 3 days ago
One model writes the code, a different model reviews itgithub.com
Hacker News · 1 points · 10 days ago
Partnering with Accenture on Embedded Evaluationanthropic.com
Hacker News · 3 points · 11 days ago