Alex Xi's Notes

Tag: mechanistic-interpretability

7 items with this tag.

  • Jul 06, 2026

    neel-nanda-problems-for-fixing

    • mechanistic-interpretability
    • research-practice
    • debugging
  • Jul 06, 2026

    representation-engineering-mistral

    • representation-engineering
    • mechanistic-interpretability
    • steering
    • activation
  • Jul 06, 2026

    the-continual-learning-problem

    • mechanistic-interpretability
    • claude
    • training
    • representation
  • Jul 06, 2026

    anthropic-emotion-concepts

    • interpretability
    • emotions
    • ai-character
    • alignment
    • anthropic
    • mechanistic-interpretability
  • Jul 06, 2026

    bytebytego-how-claude-thinks

    • interpretability
    • mechanistic-interpretability
    • claude
    • anthropic
    • features
    • polysemanticity
  • Jul 06, 2026

    can-open-source-ai-introspect

    • open-source
    • introspection
    • mechanistic-interpretability
    • llm
  • Jul 06, 2026

    mechanistic-interpretability

    • mechanistic-interpretability
    • features
    • polysemanticity
    • anthropic
    • interpretability