wassname
ActiveResearch focus
I focus on anything I think is doable and will move the needle. Mostly focused around steering and representation engineering as pragmatic tools for alignment and improving aI character.
Open to collaboration
No
Contact
Join the Damaqu community to see this team's full profile.
Papers on Damaqu (3)
- Can we use steering vectors to suppress reward-hacking? Somewhat — LessWrongother
- GitHub - wassname/w2schar-mini: Weak to strong character steering. Can a small model use weight steering to align a stronger model?other
- AntiPaSTO: Self-Supervised Honesty Steering via Anti-Parallel Representationsarxiv· 12 Jan 2026