AI Safety Fundamentals: Alignment

83 Épisodes

1. Public by Default: How We Manage Information Visibility at Get on Board

   Publié: 12/05/2024
2. Writing, Briefly

   Publié: 12/05/2024
3. Being the (Pareto) Best in the World

   Publié: 04/05/2024
4. How to Succeed as an Early-Stage Researcher: The “Lean Startup” Approach

   Publié: 23/04/2024
5. Become a Person who Actually Does Things

   Publié: 17/04/2024
6. Planning a High-Impact Career: A Summary of Everything You Need to Know in 7 Points

   Publié: 16/04/2024
7. Working in AI Alignment

   Publié: 14/04/2024
8. Computing Power and the Governance of AI

   Publié: 07/04/2024
9. Emerging Processes for Frontier AI Safety

   Publié: 07/04/2024
10. Challenges in Evaluating AI Systems

    Publié: 07/04/2024
11. AI Control: Improving Safety Despite Intentional Subversion

    Publié: 07/04/2024
12. AI Watermarking Won’t Curb Disinformation

    Publié: 07/04/2024
13. Interpretability in the Wild: A Circuit for Indirect Object Identification in GPT-2 Small

    Publié: 01/04/2024
14. Towards Monosemanticity: Decomposing Language Models With Dictionary Learning

    Publié: 31/03/2024
15. Zoom In: An Introduction to Circuits

    Publié: 31/03/2024
16. Weak-To-Strong Generalization: Eliciting Strong Capabilities With Weak Supervision

    Publié: 26/03/2024
17. Can We Scale Human Feedback for Complex AI Tasks?

    Publié: 26/03/2024
18. Machine Learning for Humans: Supervised Learning

    Publié: 13/05/2023
19. Four Background Claims

    Publié: 13/05/2023
20. Biological Anchors: A Trick That Might Or Might Not Work

    Publié: 13/05/2023