Search: “jan-leike--x”

Jan Leike announced he is beginning a new research project at Anthropic focused on contributing to safe and beneficial AGI development. The post expresses enthusiasm about the initiative and indicates

When I started to work on the alignment problem more than 10 years ago, we had no idea how AGI was going to be built or how to make it safe. The field had maybe a dozen people who were working on it a

I'm really excited about this as a new tool in our interpretability tool kit In a new paper, we present NLAs, an unsupervised method for converting an LLM's internal state into human-readable text. I'

Jan Leike praised collaborative work by researchers Jiaxin Wen, Liang Qiu, Joe Benton, and Jan Kirchner, directing followers to an Anthropic blog post for further details. The post appears to highligh

However, most alignment research is not very crisp and requires research taste when evaluating. This is why we chose to point the AAR at this scalable oversight problem! Progress would let AARs work o

New research result: we use Claude to make fully autonomous progress on scalable oversight research, as measured by performance gap recovered (PGR). Claude iterates on a number of different techniques