SAEVerbalizer fine-tunes an LLM so that injecting an SAE decoder direction into its hidden states makes it generate a natural-language explanation, and this learned skill transfers to unseen features, new SAE dictionaries, and other LLMs through a small adapter.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
SAEVerbalizer: Generating Explanations for Sparse Autoencoder Features via Representation Verbalization
SAEVerbalizer fine-tunes an LLM so that injecting an SAE decoder direction into its hidden states makes it generate a natural-language explanation, and this learned skill transfers to unseen features, new SAE dictionaries, and other LLMs through a small adapter.