tdooms

Karma: 66

When Are Two Networks the Same? Tensor Similarity for Mechanistic Interpretability

Logan Riggs, tdooms, Conflux, lwroe, MLNissenGonzalez and mel83

29 May 2026 15:53 UTC

37 points

3 comments4 min readLW link

Tokenized SAEs: Infusing per-token biases.

tdooms and danwil

4 Aug 2024 9:17 UTC

20 points

20 comments15 min readLW link