lm-sys / lm-sys/FastChat

Why the output given by Mixtral 8*7B in chatbot arena is not the same as local inference?

Open
#2,954 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
39.5k
Forks
4.8k
PR merge metrics
No merged PRs in 30d

Description

I set temperature to 0.0 and followed the prompt format at here. But the result is never the same.

Another issue mentioned it might be caused by a different system prompt which I did not use for local inference. Just wonder if (or where) I can find the system prompt used in chatbot aerna?

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start with fastchat/conversation.py at the linked prompt-format location, then read issue #2850 for the discussion about differing system prompts. Compare the local and Chatbot Arena prompt paths and identify where the Arena system prompt is defined or exposed; done means the source or a documented explanation of the prompt difference is clear.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.