lm-sys / lm-sys/FastChat

Does vicuna-13b/vicuna-7b provides flash-attention implementation during inference using this repo. If yes, where is the implementation?

Open
#1,610 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

Not an issue exactly. But couldn't figure. In training scrip, I believe it replaces all the multi-head attn with flash attn but not sure about what's happening in inference. Any help or explanation is highly appreciated.

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 locating the training script and the inference entry point referenced in the issue, then trace where multi-head attention and flash attention are selected. Done means documenting whether Vicuna inference uses flash attention and identifying the relevant implementation location.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
ai, machine-learning
Issue type
Documentation
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
20/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.