Ryan
@ryanbloom.xyz
The weights for the autoencoder come from this paper by Aditya Cowsik, Alex Infanger, and Kfir Dolev: arxiv.org/abs/2410.12101
The Persian Rug: solving toy models of superposition using large-scale symmetries
We present a complete mechanistic description of the algorithm learned by a minimal non-linear sparse data autoencoder in the limit of large input dimension. The model, originally presented in…
arxiv.org
April 8, 2025 at 12:29 AM
The weights for the autoencoder come from this paper by Aditya Cowsik, Alex Infanger, and Kfir Dolev: arxiv.org/abs/2410.12101