Publications

Georg Lange, Alex Makelov*, Neel Nanda (2025). Towards principled evaluations of sparse autoencoders for interpretability and control. In ICLR 2025.

Amber Brands, Georg Lange, Iris Groen (2025). Temporal adaptation aids object recognition in deep convolutional neural networks in suboptimal viewing scenario's. BioRxiv Preprint.

Georg Lange, Federico Gnazzo, Jeff Beeler (2025). Accumbal Dopamine and Acetylcholine Dynamics during Psychostimulant Sensitization. BioRxiv Preprint.

Georg Lange, Alex Makelov*, Neel Nanda (2024). Is This the Subspace You Are Looking For? An Interpretability Illusion for Subspace Activation Patching. In ICLR 2024.