Resurrecting the Salmon: Rethinking Mechanistic Interpretability with Domain-Specific Sparse Autoencoders

Documenti analoghi

• A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations
  di: O'Neill, Charles, et al.
  Pubblicazione: (2025)
• Sparse Autoencoders Enable Scalable and Reliable Circuit Identification in Language Models
  di: O'Neill, Charles, et al.
  Pubblicazione: (2024)
• Compute Optimal Inference and Provable Amortisation Gap in Sparse Autoencoders
  di: O'Neill, Charles, et al.
  Pubblicazione: (2024)
• Disentangling Dense Embeddings with Sparse Autoencoders
  di: O'Neill, Charles, et al.
  Pubblicazione: (2024)
• Self-Attention as a Parametric Endofunctor: A Categorical Framework for Transformer Architectures
  di: O'Neill, Charles
  Pubblicazione: (2025)