Steering CLIP's vision transformer with sparse autoencoders
Journal:
arXiv
Published Date:
Apr 11, 2025
Abstract
While vision models are highly capable, their internal mechanisms remain
poorly understood -- a challenge which sparse autoencoders (SAEs) have 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 on 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.
