Equation 29 · Probing, Sparse Autoencoders, Patching, and Steering: The Main Interpretability Methods, Compared
What does this equation mean?
Read the formula alongside the article passage below. Each part has a deeper page with its role in the equation, the supporting passage and nearby citations.
This mathematical expression combines the displayed quantities; its precise role follows from the surrounding article text. Read the equation part by part below; each part has a contextual explanation and a link to its mathematical background.
Read it piece by piece
How to interpret it
Read this expression with the definitions, units, and assumptions supplied by the article.
What the article says around this equation
where P and N index matched positive and negative examples of the target behaviour, is the residual-stream activation at layer , and is a coefficient the operator sets by hand. Nothing in this construction requires understanding why the network represents the concept along this direction, only that adding it produces the intended behavioural shift.
Sources cited in the article section
- [12] Representation Engineering: A Top-Down Approach to AI Transparency ↗
- [13] Steering Language Models With Activation Engineering ↗
- [14] Steering Llama 2 via Contrastive Activation Addition ↗
- [15] Analyzing the Generalization and Reliability of Steering Vectors ↗
These citations give research context. Read each source to check which claims it supports.