MPT-7B triton instead of torch
Open
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 39.5k
- Forks
- 4.8k
- PR merge metrics
- No merged PRs in 30d
Description
Hi! The attention mechanism should be set to triton for MPT right? I might be mistaken, but is it possible to do so? The warnings recommend that because of Alibi and faster performance
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
No file or test is named. Start by locating the MPT-7B attention implementation and the warnings about Alibi and performance, then determine whether Triton support is appropriate and how the existing behavior is tested. Done means the requested attention path is implemented or the question is resolved with supporting tests or measurements.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, pytorch
- Domain
- machine-learning
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100