1 citations · 1 across the 10 of their papers we have counts for
Showing cs.LGShow all
2 papers · 1 filter
cs.LG2026
Size Doesn't Matter: Cosine-Scored Sparse Autoencoders
Silen Naihin, Lev Stambler
Sparse autoencoders (SAEs) detect features via inner product, so a feature's activation scales with both its directional alignment and the input's norm. Features that fire on token…
cs.LG2025
Provably Overwhelming Transformer Models with Designed Inputs
Lev Stambler, Seyed Sajjad Nezhadi, Matthew Coudron
We develop an algorithm which, given a trained transformer model as input, as well as a string of tokens of length and an integer , can genera…