Ir para o conteúdo
Global sound
Tech made Easy

Tech made Easy

Claude 3 Sonnet: Scaling Monosemanticity in LLMs

· 13 min · Temporada 1 Episódio 18

Notas do episódio

This research paper explores the use of sparse autoencoders to extract interpretable features from Anthropic's Claude 3 Sonnet language model. The authors successfully scale this method to a large model, uncovering a diverse range of abstract features, including those related to safety concerns like bias, deception, and dangerous content. They investigate feature interpretability through examples and experiments, demonstrating that these features not only reflect but also causally influence model behavior. The study also examines the relationship between feature frequency and dictionary…

Site: podcasters.spotify.com

Ao vivo, última hora: 6.139 músicas em 2.173 rádios de 116 países

Descubra o mundo pelo rádio

Mais formas de descobrir (9)

Ouviu algo legal? Salve ou abra direto no Spotify.

★ Artist of the week The fastest riser on the radio this week
Your song here? From € 9 per week →
PuntalRadioContinuar ouvindo