6 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…
Reducing Political Manipulation with Consistency Training
Long Phan, Devin Kim, Alexander Pan +3
Large language models (LLMs) exhibit systematic political bias across a variety of sensitive contexts. We find that LLMs handle counterpart topics from opposing political sides asy…
LatentQA: Teaching LLMs to Decode Activations Into Natural Language
Alexander Pan, Lijie Chen, Jacob Steinhardt
Top-down transparency typically analyzes language model activations using probes with scalar or single-token outputs, limiting the range of behaviors that can be captured. To allev…
Context Is Not Comprehension
Alex Pan, Mary-Anne Williams
The dominant way of judging Large Language Models (LLMs) has been to ask how well they can recall explicit facts from very long inputs. While today's best models achieve near perfe…
Representation Engineering: A Top-Down Approach to AI Transparency
Andy Zou, Long Phan, Sarah Chen +18
In this paper, we identify and characterize the emerging area of representation engineering (RepE), an approach to enhancing the transparency of AI systems that draws on insights f…
Safetywashing: Do AI Safety Benchmarks Actually Measure Safety Progress?
Richard Ren, Steven Basart, Adam Khoja +9
As artificial intelligence systems grow more powerful, there has been increasing interest in "AI safety" research to address emerging and future risks. However, the field of AI saf…