Appears On
Episode Appearances
The Political Scene | The New Yorker · Feb 12, 2026
Can Anthropic Control What It's Building?
“Stanford professor conducting interpretability research; argued safety requires addressing proximate harms”
AI Safety and Alignment ResearchMechanistic Interpretability in Neural NetworksLarge Language Model Capability vs. Safety Trade-offs
View Analysis
The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence) · Mar 24, 2025
Dynamic Token Merging for Efficient Byte-level Language Models with Julie Kallini - #724
TokenizationByte-level language modelsDynamic token merging
View Analysis