TY  - RPRT
TI  - Aligned Training: A Parameter-Free Method to Improve Feature Quality and Stability of Sparse Autoencoders (SAE)
AU  - Michał Brzozowski
AU  - Neo Christopher Chung
PY  - 2026
UR  - https://arxiv.org/abs/2605.18629
ID  - 2605.18629
ER  -