Naval Amplifies DeepSeek Explainer: SFT Is a Bike Manual, RL Is Learning to Ride

McDonaghMatthew · x · 2026-09-06

Naval quoted Matthew McDonagh's ELI5 explainer on DeepSeek: "Imagine teaching a child to ride a bike. You could give them a detailed manual (Supervised Fine Tuning), but they'll likely learn better by trying it themselves (Reinforcement Learning), falling, getting up, and gradually improving." McDonagh celebrated being quoted, urging everyone to keep writing words and analogies that make hard technical concepts click.

Original post →

More from Models

Models channel →