Circuit Oracle: Automating Attribution Graph Analysis via Natural-Language Queries
arXiv:2611.00420v1 Announce Type: new Abstract: Attribution graphs, an emerging tool in mechanistic interpretability, use transcoders to decompose language model computations into sparse interpretable features connected by causal edges. However, turning a graph into a safety-relevant insight require
Der Artikeltext wird beim nächsten Sync nachgeladen. Original bei arXiv cs.CL lesen
Quelle: arXiv cs.CL (arXiv cs.CL). Original öffnen