Benjamin Wright

Karma: 877

Towards training-time mitigations for alignment faking in RL

Vlad Mikulik, gasteigerjo, Hoagy, Joe Benton, Benjamin Wright, Jonathan Uesato, Monte M, Fabien Roger and evhub

16 Dec 2025 21:01 UTC

33 points

1 comment5 min readLW link

(alignment.anthropic.com)

Natural emergent misalignment from reward hacking in production RL

evhub, Monte M, Benjamin Wright and Jonathan Uesato

21 Nov 2025 20:00 UTC

256 points

32 comments9 min readLW link

Agentic Misalignment: How LLMs Could be Insider Threats

Aengus Lynch, Benjamin Wright, Ethan Perez and evhub

20 Jun 2025 22:34 UTC

77 points

13 comments6 min readLW link

Alignment Faking in Large Language Models

ryan_greenblatt, evhub, Carson Denison, Benjamin Wright, Fabien Roger, Monte M, Sam Marks, Johannes Treutlein, Sam Bowman and Buck

18 Dec 2024 17:19 UTC

490 points

86 comments10 min readLW link 3 reviews

Evaluating Sparse Autoencoders with Board Game Models

Adam Karvonen, Sam Marks, Can, Benjamin Wright, Jannik Brinkmann, Logan Riggs and Rico Angell

2 Aug 2024 19:50 UTC

38 points

1 comment9 min readLW link

Addressing Feature Suppression in SAEs

Benjamin Wright and Lee Sharkey

16 Feb 2024 18:32 UTC

87 points

5 comments10 min readLW link