
5. Charlie Snell on DALL-E and CLIP
The Inside View
The Image and Better Is a Transformer
Clip text and better is essentially doing like war to vaguer like something similar or something specific to clip. It's basically a transformer it takes in the text it's transformer sort of like GPT and it takes the sort of a contextual embedding for the text from that transformer. So you just take sort of like one of the tokens the last token and the last layer of the transformer and use that to represent the sentence. And so that's like the more technical let's say justification on when I look at this.
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.