✳flâneur — a map of the web's best reading

Mor Geva on X: "What's in a neuron? 💫 (an atypically long, almost personal post) Neurons in LMs have always been a fascinating object to study. I've been studying them since 2020, viewing them as key-value memory cells, analyzing what they capture in vocabulary space, and how they compose https://t.co/179ELFnEw3" / X

x.com · 168 words · saved by 1 readers

What's in a neuron? 💫 (an atypically long, almost personal post) Neurons in LMs have always been a fascinating object to study. I've been studying them since 2020, viewing them as key-value memory cells, analyzing what they capture in vocabulary space, and how they compose https://t.co/179ELFnEw3

@megamor2: What's in a neuron? (an atypically long, almost personal post) Neurons in LMs have always been a fascinating object to study. I've been studying them since 2020, viewing them as key-value memory cells, analyzing what they capture in vocabulary space, and how they compose together to form features. https:// arxiv.org/abs/2012.14913 https:// arxiv.org/abs/2203.14680 https:// arxiv.org/abs/2501.08319 https:// arxiv.org/abs/2506.10920 But many neurons still remain opaque! They do many things. Our recent work led by @AsafAvrahamy tackles this challenge by decomposing neuron wei

Explore this link on the map →

saved by

related reading