AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
3min chapter
Raphaël Millière: The Vector Grounding Problem and Self-Consciousness
The Gradient: Perspectives on AI
How to Fine Tune a Language Model to Give Helpful, Harmless, and Honest Answers
Researchers call the three H's, namely, honesty, harmlessness, and helpfulness. The RRL-H8 fine tuning technique helps train models to use first person pronouns in a way that is somewhat systematic and reliable. When given certain prompts, such as explaining how to hotwire a car, you don't want the model to answer because it could be harmful. And so when you're not familiar with the nuts and bolts of training of a lens, it's easy to jump to conclusions.
00:00
Transcript
Episode notes
Get the Snipd
podcast app
Unlock the knowledge in podcasts with the podcast player of the future.
AI-powered
Discover
highlights
Listen to the best highlights from the podcasts you love and dive into the full episode
Save any
moment
Hear something you like? Tap your headphones to save it with AI-generated key takeaways
Share
& Export
Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more
AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
Discover
highlights
Listen to the best highlights from the podcasts you love and dive into the full episode