Training sparse autoencoders on a language model's own generated text can improve seed stability and downstream probing relative to training on web text.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
FaithfulSAE: Towards Capturing Faithful Features with Sparse Autoencoders without External Dataset Dependencies
Training sparse autoencoders on a language model's own generated text can improve seed stability and downstream probing relative to training on web text.