2026
Priors in time: Missing inductive biases for language model interpretability
Ekdeep Singh Lubana, Sai Sumedh R. Hindupur, Can Rager, Valérie Costa, Oam Patel, Sonia Krishna Murthy +7
ICLR 2026poster
A central aim of interpretability tools applied to language models is to recover meaningful concepts from model activations. Existing feature extraction methods focus on single activations regardless of the context, implicitly assuming independence (and therefore stationarity). This leaves open whet…