Monitoring Latent World States in Language Models with Propositional Probes

Documenti analoghi

• Extractive Structures Learned in Pretraining Enable Generalization on Finetuned Facts
  di: Feng, Jiahai, et al.
  Pubblicazione: (2024)
• How do Language Models Bind Entities in Context?
  di: Feng, Jiahai, et al.
  Pubblicazione: (2023)
• Learning a Generative Meta-Model of LLM Activations
  di: Luo, Grace, et al.
  Pubblicazione: (2026)
• LatentQA: Teaching LLMs to Decode Activations Into Natural Language
  di: Pan, Alexander, et al.
  Pubblicazione: (2024)
• Discovering Latent Knowledge in Language Models Without Supervision
  di: Burns, Collin, et al.
  Pubblicazione: (2022)