Seonglae ChoWhen comparing MSE and Total Information while increasing the Dictionary size, a bounce point appears in the graph at the correct number of features. This suggests that we have found the most appropriate factorization without excessively splitting or merging features. It is used as a tool to visually confirm that the "simplest decomposition" on the Pareto frontier becomes the optimal solution.
Circuits Updates — May 2023
We report a number of developing ideas on the Anthropic interpretability team, which might be of interest to researchers working actively in this space. Some of these are emerging strands of research where we expect to publish more on in the coming months. Others are minor points we wish to share, since we're unlikely to ever write a paper about them.
https://transformer-circuits.pub/2023/may-update/index.html#simple-factorization