venue
active
venue:transformer-circuits

Transformer Circuits

Publication venue for this paper; platform for interpretability research on LLM internals.

Neighborhood — ranked by edge-count

Methods (1)

method
  • Natural Language Autoencoders (NLAs)
    cites
    Core unsupervised method for generating natural language explanations of LLM activations through a verbalizer-reconstructor pair trained with RL.