other

Cross-Layer Transcoder

Sparse feature-learning setup in which an encoder reads a layer's activations and decoders write the resulting features into the MLP outputs of that layer and later ones.

Known aliases

CLT features
cross-layer transcoders

Relationships

No evidence-backed relationships are recorded.

Current clusters

build1 publisher

Steering 64 CLT features moves Gemma-2-2B through the ordered days at weaker token probabilities

A replication of Wurgaft et al.'s manifold steering on Anthropic's pre-trained CLT features reproduces the cyclical weekday transition in Gemma-2-2B. The author attributes the weaker push to the MLP sublayers holding only part of the signal.

Publishers:lesswrong.com

Reality

Evidence42
Adoption: Insufficient
Hype gap−5
Incentives25
Confidence45