This project was submitted by Naama Rozen. It was an Outstanding Submmission for the Technical AI Safety Project Sprint (Jan 2026). Participants worked on these projects for 5 weeks.
Does Llama-70B Actually Fake Alignment? A…
This project was submitted by Naama Rozen. It was an Outstanding Submmission for the Technical AI Safety Project Sprint (Jan 2026). Participants worked on these projects for 5 weeks.