From superposition to sparse codes: interpretable representations in neural networks

Documenti analoghi

• Compute Optimal Inference and Provable Amortisation Gap in Sparse Autoencoders
  di: O'Neill, Charles, et al.
  Pubblicazione: (2024)
• Position: An Empirically Grounded Identifiability Theory Will Accelerate Self-Supervised Learning Research
  di: Reizinger, Patrik, et al.
  Pubblicazione: (2025)
• Causality is Key for Interpretability Claims to Generalise
  di: Joshi, Shruti, et al.
  Pubblicazione: (2026)
• Superposition disentanglement of neural representations reveals hidden alignment
  di: Longon, André, et al.
  Pubblicazione: (2025)
• Self-Attention as a Parametric Endofunctor: A Categorical Framework for Transformer Architectures
  di: O'Neill, Charles
  Pubblicazione: (2025)