Models for more layers
Open
- Dominant language
- Python
- Stars
- 66
- Forks
- 12
- PR merge metrics
- No merged PRs in 30d
Description
Hello Authors,
Thank you for releasing the codebase. Do you plan to release the SAEs at different layers of the backbone model?
Regards
Contributor guide
Research direction
No files, tests, or entry points are mentioned. First determine how the existing sparse autoencoders are tied to backbone layers and whether models for additional layers are intended; done would require an agreed scope and released models or a documented decision.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- computer-vision, machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 15/100