lm-sys / lm-sys/FastChat

Minor typos in judge prompts

Open
#2,802 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

In llm_judge/data/judge_prompts.jsonl there are two "You evaluation" typos which should be "Your evaluation"

  1. "You evaluation should focus on the assistant's answer to the second question" in "single-math-v1-multi-turn"
  2. "You evaluation should focus on the assistant's answer to the second question" in "single-v1-multi-turn"

Doubt they make any difference to performance as this is an easy error for models to parse but worth correcting for completeness.

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

Open fastchat/llm_judge/data/judge_prompts.jsonl and locate the single-math-v1-multi-turn and single-v1-multi-turn entries. Check both occurrences of "You evaluation" and confirm they read "Your evaluation" when the two prompt strings are corrected.

Written by the indexing model from the issue text.

Assessment

Domain
content
Issue type
Documentation
Difficulty
1/5
Estimated time
Under an hour
Activity status
Stale
Clarity
Clearly specified
Newbie friendliness
58/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.