"The Cognitive Revolution" | AI Builders, Researchers, and Live Player Analysis cover image

E17: The Prompt Engineering Revolution with Riley Goodside

"The Cognitive Revolution" | AI Builders, Researchers, and Live Player Analysis

00:00

THIRD ERA: RLHF

The late chef phase of large language models focuses on automating instruction following and instruction tuning to reduce costs. Instead of having humans perform tasks, the model is allowed to generate examples. If humans agree that the model's output is indistinguishable from human performance, it is considered as being done by a human. This approach helps in increasing the number of correct task examples without incurring additional costs.

Transcript
Play full episode

Remember Everything You Learn from Podcasts

Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.
App store bannerPlay store banner
Get the app