$\textit{Revelio}$: Interpreting and leveraging semantic information in diffusion models
This work advances interpretability of black-box diffusion models, which is crucial for researchers and practitioners in AI and computer vision.
The authors investigated how visual semantic information is represented across layers and timesteps in diffusion models, using k-sparse autoencoders to uncover interpretable features and demonstrating their effectiveness for representation learning on four datasets.
We study $\textit{how}$ rich visual semantic information is represented within various layers and denoising timesteps of different diffusion architectures. We uncover monosemantic interpretable features by leveraging k-sparse autoencoders (k-SAE). We substantiate our mechanistic interpretations via transfer learning using light-weight classifiers on off-the-shelf diffusion models' features. On $4$ datasets, we demonstrate the effectiveness of diffusion features for representation learning. We provide an in-depth analysis of how different diffusion architectures, pre-training datasets, and language model conditioning impacts visual representation granularity, inductive biases, and transfer learning capabilities. Our work is a critical step towards deepening interpretability of black-box diffusion models. Code and visualizations available at: https://github.com/revelio-diffusion/revelio