Sonia Joseph

Steering CLIP's vision transformer with sparse autoencoders

Sonia Joseph

Praneet Suresh

Ethan Goldfarb

Lorenz Hufe

Yossi Gandelsman

Robert Graham

Danilo Bzdok

Wojciech Samek

Blake Richards

While vision models are highly capable, their internal mechanisms remain poorly understood-- a challenge which sparse autoencoders (SAEs) ha… (voir plus)ve helped address in language, but which remains underexplored in vision. We address this gap by training SAEs on CLIP's vision transformer and uncover key differences between vision and language processing, including distinct sparsity patterns for SAEs trained across layers and token types. We then provide the first systematic analysis of the steerability of CLIP's vision transformer by introducing metrics to quantify how precisely SAE features can be steered to affect the model's output. We find that 10-15% of neurons and features are steerable, with SAEs providing thousands more steerable features than the base model. Through targeted suppression of SAE features, we then demonstrate improved performance on three vision disentanglement tasks (CelebA, Waterbirds, and typographic attacks), finding optimal disentanglement in middle model layers, and achieving state-of-the-art performance on defense against typographic attacks. We release our CLIP SAE models and code to support future research in vision transformer interpretability.

2025-03-30

thecvf.com/CVPR/2025/Workshop/MIV (poster)

openreview.net

Interpretability in Action: Exploratory Analysis of VPT, a Minecraft Agent

Karolis Jucys

George Adamopoulos

Mehrab Hamidi

Stephanie Milani

Mohammad Reza Samsami

Özgür Şimşek

Understanding the mechanisms behind decisions taken by large foundation models in sequential tasks is critical to ensuring that such systems… (voir plus) operate transparently and safely. However, interpretability methods have not yet been applied extensively to large-scale agents based on reinforcement learning. In this work, we perform exploratory analysis on the Video PreTraining (VPT) Minecraft playing agent, one of the largest open-source vision-based agents. We try to illuminate its reasoning mechanisms by applying various interpretability techniques. First, we analyze the attention mechanism while the agent solves its training task --- crafting a diamond pickaxe. The agent seems to pay attention to the 4 last frames and several key-frames further back. This provides clues as to how it maintains coherence in the task that takes 3-10 minutes, despite the agent's short memory span of only six seconds. Second, we perform various interventions, which help us uncover a worrying case of goal misgeneralization: VPT mistakenly identifies a villager wearing brown clothes as a tree trunk and punches it to death, when positioned stationary under green tree leaves. We demonstrate similar misbehavior in a related agent (STEVE-1), which motivates the use of VPT as a model organism for large-scale vision-based agent interpretability.

2024-06-24

ICML.cc/2024/Workshop/MI (poster)

doi.org

openreview.net

On the Information Geometry of Vision Transformers

Sonia Joseph

Kumar Krishna Agrawal

Arna Ghosh

Blake Richards

2023-11-28

NeurIPS.cc/2023/Workshop/NeurReps (poster)

openreview.net