Mechanistic evaluation of SAE-based activation interventions for correcting distractor-induced errors in RAG.

7 Open Issues Need Help Last updated: Jul 30, 2026

Open Issues Need Help

View All on GitHub
good first issue dataset spike

Mechanistic evaluation of SAE-based activation interventions for correcting distractor-induced errors in RAG.

Python
good first issue lat-defuse

Mechanistic evaluation of SAE-based activation interventions for correcting distractor-induced errors in RAG.

Python
good first issue sae-gemma

Mechanistic evaluation of SAE-based activation interventions for correcting distractor-induced errors in RAG.

Python
good first issue infra

Mechanistic evaluation of SAE-based activation interventions for correcting distractor-induced errors in RAG.

Python
good first issue science

Mechanistic evaluation of SAE-based activation interventions for correcting distractor-induced errors in RAG.

Python
good first issue science

Mechanistic evaluation of SAE-based activation interventions for correcting distractor-induced errors in RAG.

Python
good first issue infra

Mechanistic evaluation of SAE-based activation interventions for correcting distractor-induced errors in RAG.

Python