6 citations · 6 across the 4 of their papers we have counts for
4 papers
A Shared Subcircuit Lets LLMs Count Down Across Tasks
Jacob Dunefsky, Wes Gurnee, Emmanuel Ameisen
Writing a sentence of exactly twelve words; ending a DNA sequence at the right codon; formatting an ASCII table. These are all tasks that language models can do that requires track…
One-shot Optimized Steering Vectors Mediate Safety-relevant Behaviors in LLMs
Jacob Dunefsky, Arman Cohan
Steering vectors (SVs) have emerged as a promising approach for interpreting and controlling LLMs, but current methods typically require large contrastive datasets that are often i…
Transcoders Find Interpretable LLM Feature Circuits
Jacob Dunefsky, Philippe Chlenski, Neel Nanda
A key goal in mechanistic interpretability is circuit analysis: finding sparse subgraphs of models corresponding to specific behaviors or capabilities. However, MLP sublayers make…
Observable Propagation: Uncovering Feature Vectors in Transformers
Jacob Dunefsky, Arman Cohan
A key goal of current mechanistic interpretability research in NLP is to find linear features (also called "feature vectors") for transformers: directions in activation space corre…