Mapping the Mind of a Neural Net: Goodfire’s Eric Ho on the Future of Interpretability

Interpretability tooling aims to let you inspect, steer and debug model behavior directly, a possible alternative to prompting and fine-tuning for reliability.

• fine-tuning — Taking a trained model and training it a bit more on your own examples so it gets better at one specific job.

• videoHow to Scale AI Application Inference 100x ft. Fireworks’ Lin Qiao
• videoOpenAI Researcher Dan Roberts on What Physics Can Teach Us About AI
• videoOpenAI Codex Team: From Coding Autocomplete to Asynchronous Autonomous Agents
• videoStarcloud's Philip Johnston: Why the Cheapest Compute Will Be in Space

• videoInterpretability: Understanding how AI models thinkAnthropic
• videoGoogle DeepMind's Logan Kilpatrick: Why the Model Eats the HarnessSequoia Capital
• videoBuilding the Automated AGI Lab: Core Automation's Jerry Tworek and Rohan AnilSequoia Capital

Checking sign-in…

Loading comments…