The network at most can, you know, just do interpolation in the range we train the on. That's maybe a good bridge to start to move into because we're still kind of climbing the ladder of complexity with these experiments. We see that amazingly it can learn to do that with just very few active neurons and we attribute that to the unreasonable effectiveness of the smooth activation curve. And then you've got the third one, which is compositionality, which starts to look a little bit more tangled but clearly has some parts that I see as kind of interpretable. So even if we cannot generalize, there's still a lot of interesting things we can say about at least where the task

Get the Snipd
podcast app

Unlock the knowledge in podcasts with the podcast player of the future.
App store bannerPlay store banner

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

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