AXRP - the AI X-risk Research Podcast cover image

17 - Training for Very High Reliability with Daniel Ziegler

AXRP - the AI X-risk Research Podcast

CHAPTER

How Do We Get More Out of Our Time?

The first part was just, like, people writing things and seeng if they seeing if they got throeh. And then sort of the question was, you know, how do we get more out of some fixed amount of human time? The two main things we did there are, one is, we did a sort of paraphrasing thing where we wer te used gpt three. Two, we like, few shot at gpety three to like, take some existing adversale example, or some sum, like some injurious example, and rewrite it in a prettyFew pr divers ways to generate more candidate adversal examples. Then, of the ones of the classifier missed or

00:00
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