Hacker News new | ask | show | jobs
Aligning LLMs at inference time by suppressing internal concepts (guidelabs.ai)
2 points by adebayoj 87 days ago