Zhao Pengfei

When we look at the downstream change in activations, relative to the base model, when steering with the B vector and mean-diff direction individually, we find that the differences induced reach a cosine similarity of 0.42 (this is significantly higher than occurs with random vectors).

At which point are those downstream changes measured ?

When we ablate the mean-diff direction from the B vector of a single rank-1 adapter model at layer 24, misalignment

What if you use the B vector to ablate from the residual activations ?

(0.04 cosine similarity)

Couldn't it possible that 0.04 is actually high in high dimensional space ? One could compare it to samples of random directions.