Tracking the Feature Dynamics in LLM Training: A Mechanistic Study

Documenti analoghi

• On Calibration of LLM-based Guard Models for Reliable Content Moderation
  di: Liu, Hongfu, et al.
  Pubblicazione: (2024)
• Feature Rivalry in Sparse Autoencoder Representations: A Mechanistic Study of Uncertainty-Driven Feature Competition in LLMs
  di: Harshavardhan
  Pubblicazione: (2026)
• Finite State Automata Inside Transformers with Chain-of-Thought: A Mechanistic Study on State Tracking
  di: Zhang, Yifan, et al.
  Pubblicazione: (2025)
• BayesAgent: Bayesian Agentic Reasoning Under Uncertainty via Verbalized Probabilistic Graphical Modeling
  di: Huang, Hengguan, et al.
  Pubblicazione: (2024)
• Rethinking Local Learning: A Cheaper and Faster Recipe for LLM Post-Training
  di: Shi, Hengyu, et al.
  Pubblicazione: (2026)