Extracting Steering Vector for Contrastive Activation Addition
- Clean run
- Corrupted run
- Patched run
Example
- Clean input “What city is the Eiffel Tower in?” → Save clean activation
- Corrupted input “What city is the Colosseum in?” → Save corrupted output
- Patch activation on the corrupted input from the clean activation → observe which activation layer or attention head is important for producing the correct answer, “Paris”
Neel nanda method
Contrastive Activation Addition SVs operate unstably for specific inputs within the data in-distribution and have limitations in generalization OOD data