Refusal in Language Models Is Mediated by a Single Direction
Paper • 2406.11717 • Published • 9 •
2
Ahmad Der
ranea
·
AI & ML interests
None yet
Recent Activity
commentedon a paper 1 day ago
Refusal in Language Models Is Mediated by a Single Direction commentedon a paper 1 day ago
Refusal in Language Models Is Mediated by a Single Direction liked a Space 1 day ago
ThirdTimesTheCiarc/lora-scalerOrganizations
None yet
commented 2
papers 1 day ago