A global workspace in language models

·LessWrong··

[This is the blog post for our new paper Verbalizable Representations Form a Global Workspace in Language ModelsReaders might also be interested in: the Public commentary, Github and Neuronpedia]As you read this sentence, circuits in your brain are adjusting your posture, controlling your breathing, and transforming lines and curves on the screen into recognizable words. Most of this processing is invisible to you. But some of what takes place in your brain you do have access to—an image that po...

Read full article →

Related Articles

Improper redaction reveals Google Data Center water and electricity usage
sensanaty · Hacker News · 6h ago
Run Qwen 3.8 Flash Next (125B) on consumer hardware (RTX 4090) at 100T/s
snehesht · Hacker News · 13h ago
Car is a smartphone on wheels. Here's who's listening
longhaul · Hacker News · 10h ago
Federal judge calls Flock 'indiscriminate mass surveillance'
sbulaev · Hacker News · 1d ago
Kolibri: A Sovereign Open-Weight Model
bastitx · Hacker News · 1d ago