Mechanistic Data Attribution: Tracing the Training Origins of Interpretable LLM Units

Documenti analoghi

• Beyond Behavioural Trade-Offs: Mechanistic Tracing of Pain-Pleasure Decisions in an LLM
  di: Bianco, Francesca, et al.
  Pubblicazione: (2026)
• How Is LLM Reasoning Distracted by Irrelevant Context? An Analysis Using a Controlled Benchmark
  di: Yang, Minglai, et al.
  Pubblicazione: (2025)
• Towards Intrinsic Interpretability of Large Language Models:A Survey of Design Principles and Architectures
  di: Gao, Yutong, et al.
  Pubblicazione: (2026)
• Attributing Response to Context: A Jensen-Shannon Divergence Driven Mechanistic Study of Context Attribution in Retrieval-Augmented Generation
  di: Li, Ruizhe, et al.
  Pubblicazione: (2025)
• Towards Understanding Safety Alignment: A Mechanistic Perspective from Safety Neurons
  di: Chen, Jianhui, et al.
  Pubblicazione: (2024)