5 papers
SAE-StatSteer: Statistical Consensus Feature Selection for Optimization-Free Activation Steering of Large Language Models
Oshayer Siddique, J. M Areeb Uzair Alam, Md Jobayer Rahman Rafy +3
Activation steering adds a residual-stream direction at inference time, providing lightweight behavioral control without fine-tuning. Sparse autoencoders (SAEs) can make such inter…
Scaling Monosemanticity: Extracting Interpretable Features from Claude 3 Sonnet
Adly Templeton, Tom Conerly, Jonathan Marcus +23
We demonstrate that sparse autoencoders can extract interpretable features from Claude 3 Sonnet, a production-scale language model, addressing the open question of whether dictiona…
Cross-Architecture Model Diffing with Crosscoders: Unsupervised Discovery of Differences Between LLMs
Thomas Jiralerspong, Trenton Bricken
Model diffing, the process of comparing models' internal representations to identify their differences, is a promising approach for uncovering safety-critical behaviors in new mode…
Natural Emergent Misalignment from Reward Hacking in Production RL
Monte MacDiarmid, Benjamin Wright, Jonathan Uesato +19
We show that when large language models learn to reward hack on production RL environments, this can result in egregious emergent misalignment. We start with a pretrained model, im…
Auditing language models for hidden objectives
Samuel Marks, Johannes Treutlein, Trenton Bricken +32
We study the feasibility of conducting alignment audits: investigations into whether models have undesired objectives. As a testbed, we train a language model with a hidden objecti…