Blog Post··8 min read
Logit Lens: How Predictions Form Layer by Layer
Applying the unembedding matrix at intermediate layers to watch how a transformer's prediction evolves — and what direct logit attribution tells us about which components matter.
Tag: direct-logit-attribution
Blog Post··8 min read
Logit Lens: How Predictions Form Layer by Layer
Applying the unembedding matrix at intermediate layers to watch how a transformer's prediction evolves — and what direct logit attribution tells us about which components matter.