Comment by andy12_
3 days ago
This seems really interesting. While Anthropic tried to use dictionary learning over an existing model to try to extract concepts, this almost feels like training the model alongside the dictionary itself (or rather, the model and the dictionary are intertwined).
You are exactly right, it is guiding the model, during training, with concepts and the dictionary. This is important because dictionary learning for interpretability (post hoc) is not currently reliable: https://www.arxiv.org/abs/2602.14111