AI Safety Fundamentals: Alignment
En podkast av BlueDot Impact
Kategorier:
83 Episoder
-
Public by Default: How We Manage Information Visibility at Get on Board
Publisert: 12.5.2024 -
Writing, Briefly
Publisert: 12.5.2024 -
Being the (Pareto) Best in the World
Publisert: 4.5.2024 -
How to Succeed as an Early-Stage Researcher: The “Lean Startup” Approach
Publisert: 23.4.2024 -
Become a Person who Actually Does Things
Publisert: 17.4.2024 -
Planning a High-Impact Career: A Summary of Everything You Need to Know in 7 Points
Publisert: 16.4.2024 -
Working in AI Alignment
Publisert: 14.4.2024 -
Computing Power and the Governance of AI
Publisert: 7.4.2024 -
Emerging Processes for Frontier AI Safety
Publisert: 7.4.2024 -
Challenges in Evaluating AI Systems
Publisert: 7.4.2024 -
AI Control: Improving Safety Despite Intentional Subversion
Publisert: 7.4.2024 -
AI Watermarking Won’t Curb Disinformation
Publisert: 7.4.2024 -
Interpretability in the Wild: A Circuit for Indirect Object Identification in GPT-2 Small
Publisert: 1.4.2024 -
Towards Monosemanticity: Decomposing Language Models With Dictionary Learning
Publisert: 31.3.2024 -
Zoom In: An Introduction to Circuits
Publisert: 31.3.2024 -
Weak-To-Strong Generalization: Eliciting Strong Capabilities With Weak Supervision
Publisert: 26.3.2024 -
Can We Scale Human Feedback for Complex AI Tasks?
Publisert: 26.3.2024 -
Machine Learning for Humans: Supervised Learning
Publisert: 13.5.2023 -
Four Background Claims
Publisert: 13.5.2023 -
Biological Anchors: A Trick That Might Or Might Not Work
Publisert: 13.5.2023
Listen to resources from the AI Safety Fundamentals: Alignment course!https://aisafetyfundamentals.com/alignment