Steering the top SAE features selected by target-language contrast improves multilingual benchmark accuracy in Gemma-3-12B-it without parameter updates.
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
-
Strengthening Target-Language Features: SAE-Based Steering for Multilingual Inference
Steering the top SAE features selected by target-language contrast improves multilingual benchmark accuracy in Gemma-3-12B-it without parameter updates.