This chapter explores the phenomenon of polysementicity in transformer structures within neural networks, specifically focusing on the multi-layer perceptron (MLP). It discusses the challenges of interpreting and untangling the representations of neurons that fire for multiple different conceptual meanings. The chapter also delves into the issue of conceptual bleedovers caused by the reuse of neurons across distinct elements, leading to unexpected behaviors.

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