Community Discourse & Replication
Open peer critique, empirical replications, and alignment hypotheses.
-
research discussion 2026-10-12
Can sleeper agent backdoors be detected using Sparse Autoencoders before activation?
Given that Anthropic showed monosemantic dictionary learning can extract hidden features (Bricken et al., 2023), has anyone attempted to run SAEs on the decepti...
✎ 0 contributions Join Thread →