Search papers, labs, and topics across Lattice.
1
20
3
18
Unlocking VLM interpretability, sparse autoencoders let you directly steer multimodal LLMs like LLaVA by intervening on CLIP's vision encoder.
