The Value Axis: Language Models Encode Whether They're on the Right Track
Nick Jiang, Isaac Kauvar, Jack Lindsey
TLDR: language models internally estimate how well they're doing towards their goals along a linear axis, akin to a value function in RL.
Interpretable Embeddings with Sparse Autoencoders: A Data Analysis Toolkit
Nick Jiang*, Xiaoqing Sun*, Lisa Dunlap, Lewis Smith, Neel Nanda
ICML 2026, NeurIPS Mech Interp Workshop 2025 (Spotlight ⭐)
paper | project page | code | X thread
TLDR: we show that sparse autoencoders outperform baselines on four data analysis tasks and find surprising model behaviors by analyzing training data and outputs.
Vision Transformers Don't Need Trained Registers
Nick Jiang*, Amil Dravid*, Alexei Efros, Yossi Gandelsman
NeurIPS 2025 (Spotlight ⭐)
paper | project page | code | X thread
TLDR: we find and remove a sparse mechanism that causes attention sinks in ViTs, leading to general performance gains.
Interpreting and Editing Vision-Language Representations to Mitigate Hallucinations
Nick Jiang*, Anish Kachinthaya*, Suzie Petryk, Yossi Gandelsman
ICLR 2025
TLDR: we use logit lens to identify and reduce hallucinations by 25% training-free from vision-language models.
*equal contribution