learn.ifs.name

Scaling Monosemanticity: Extracting Interpretable Features from Claude 3 Sonnet

Templeton et al., 2024, Scaling Monosemanticity (Transformer Circuits Thread)

Subscriber deck — free during early access

1 cards in “Features as Computational Intermediates” · whole deck

Features as Computational Intermediates

  1. How can attribution reveal a model’s intermediate steps, according to Templeton et al.?
    By estimating each feature’s effect on the logit difference between candidate answers. For a question about the capital of the state where Kobe Bryant played, top features included Kobe Bryant, the Lakers, California and Los Angeles, though these were far from the most strongly active features.
    • Templeton et al., 2024, Scaling Monosemanticity (Transformer Circuits Thread), Features as Computational Intermediates

Source

  • Adly Templeton, Tom Conerly, Jonathan Marcus, Jack Lindsey, Trenton Bricken, Brian Chen, Adam Pearce, Craig Citro, Emmanuel Ameisen, Andy Jones, Hoagy Cunningham, Nicholas L. Turner, Callum McDougall, Monte MacDiarmid, Alex Tamkin, Esin Durmus, Tristan Hume, Francesco Mosconi, C. Daniel Freeman, Theodore R. Sumers, Edward Rees, Joshua Batson, Adam Jermyn, Shan Carter, Chris Olah and Tom Henighan. 2024. Scaling Monosemanticity: Extracting Interpretable Features from Claude 3 Sonnet. In Transformer Circuits Thread. https://transformer-circuits.pub/2024/scaling-monosemanticity/index.html