Anthropic discovered a small section of Claude's internal reasoning they call the "J-space." I explore what it is, and why it's so damn cool.
---
title: "The J-Space: How I Learned To Read An LLM's Mind"
published: true
description: "Anthropic discovered a small section of Claude's internal reasoning they call the \"J-space.\" I explore what it is, and why it's so damn cool."
tags: ai, LLM, anthropic, modeltraining
cover_image: https://www.maiatalksabout.ai/images/j-space/cover.png
canonical_url: https://www.maiatalksabout.ai/blog/j-space
---
Last week, Anthropic published a paper on a discovery they made regarding Claude's internal reasoning they call the **J-space**. It seems to be the steps that Claude works through before it commits to a final word: the closest thing we've come to seeing the inside of an LLM's "brain."
Although it holds a median of only 6–7% of a concept's representation inside the model and never more than about a tenth of the model's activity at any layer, if you switch it off, Claude's multi-step reasoning collapses to almost nothing. Fluent speech and simple recall remain intact.
I quite liked the [video that Anthropic released](https://www.youtube.com/watch?v=rKV5JcALQoQ) with the [research post](https://www.anthropic.com/research/global-workspace). It's a long research post though, so I thought I'd write a summary of the parts I considered the coolest and how I interpreted the mathematics of the J-space.
---
*This is a preview of a post from my blog.* [**Read the full post with the interactive charts →**](https://www.maiatalksabout.ai/blog/j-space)