On the Origins of Linear Representations in Large Language Models

Ähnliche Einträge

• Do LLMs dream of elephants (when told not to)? Latent concept association and associative memory in transformers
  von: Jiang, Yibo, et al.
  Veröffentlicht: (2024)
• Learning Interpretable Concepts: Unifying Causal Representation Learning and Foundation Models
  von: Rajendran, Goutham, et al.
  Veröffentlicht: (2024)
• Identifying General Mechanism Shifts in Linear Causal Representations
  von: Chen, Tianyu, et al.
  Veröffentlicht: (2024)
• The Linear Representation Hypothesis and the Geometry of Large Language Models
  von: Park, Kiho, et al.
  Veröffentlicht: (2023)
• The Geometry of Categorical and Hierarchical Concepts in Large Language Models
  von: Park, Kiho, et al.
  Veröffentlicht: (2024)