1d:I[2890,["266","static/chunks/870fdd6f-4fcc92c129a57d11.js","970","static/chunks/970-91a3a37cc668faba.js","857","static/chunks/857-3b2a1421afa266d2.js","965","static/chunks/app/podcasts/episode/%5BpodcastEpisodeGuid%5D/page-e939c6b91b4a25d4.js"],"default"] 1e:I[1318,["266","static/chunks/870fdd6f-4fcc92c129a57d11.js","970","static/chunks/970-91a3a37cc668faba.js","857","static/chunks/857-3b2a1421afa266d2.js","965","static/chunks/app/podcasts/episode/%5BpodcastEpisodeGuid%5D/page-e939c6b91b4a25d4.js"],"default"] 1f:T40c,

Dev Rishi is the founder and CEO of Predibase, the company behind Ludwig and LoRAX. Predibase just released LoRA Land, a technical report showing 310 models that can outcompete GPT-4 on specific tasks through fine-tuning. In this episode, Dev tries (pretty successfully) to convince me that fine-tuning is the future, while answering a bunch of interesting questions, like:

Is fine-tuning hard?
If LoRAX is a competitive advantage for you, why open-source it?
Is model hosting becoming commoditized? If so, how can anyone compete?
What are people actually fine-tuning language models for?
How worried are you about OpenAI eating your lunch?

I had a ton of fun with Dev on this one. Also, check out Predibase’s newsletter called fine-tuned (great name!) and LoRA Land.

20:T536,[{"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Podcasts","item":"https://metacast.app/podcasts"},{"@type":"ListItem","position":2,"name":"Technology","item":"https://metacast.app/podcasts/categories/technology"},{"@type":"ListItem","position":3,"name":"Thinking Machines: AI & Philosophy","item":"https://metacast.app/podcasts/a02a3ceb-3418-5549-9514-d67c8101eb2c"}]},{"@context":"https://schema.org","@typ

‌

Episode description

The Future is Fine Tuned (with Dev Rishi, Predibase)

Episode description