Article Source

• Keeping AI under control through mechanistic interpretability

Keeping AI under control through mechanistic interpretability

• The Impact of chatGPT and other large language models on physics research and education (2023)
• Event organizers: Kevin Burdge, Joshua Borrow, Mark Vogelsberger
• Session 1: The computer science underlying large language models
• “Keeping AI under control through mechanistic interpretability”
• Speaker: Prof. Max Tegmark (MIT)*