lm-sys / lm-sys/FastChat

Does MT-Bench gen_model_answer.py accept adapter weights (adapter_config.json) file as model path?

Open
#3,453 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

Hi,

I have been passing the adapter weights path for the mistral 7b v0.3 model to gen_model_answer.py script as the model path. I obtained satisfactory results from it, but I wanted to be confirm if you guys merge the adapter weights yourself in the backend with the original base model?

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 by reading gen_model_answer.py and checking how the supplied adapter path is handled. Verify the behavior with the Mistral 7B v0.3 adapter, then document whether the backend merges it with the original base model.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Documentation
Difficulty
2/5
Estimated time
1-3 hours
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
30/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.