corpus.blog
56,966 blogs · 4,417,066 posts Claim your blog
Record · corpus.blog/posts/01a08219-ddaf-7384-bcb4-f2bf9d2cecbc

Natural Language Autoencoders Produce Unsupervised Explanations of LLM Activations

Transformer Circuits · published 7 May 2026

A record is what was published and who pointed at it. The text of the post is not held here: read it at the source, or at its Wayback capture.

Read the post at transformer-circuits.pub Wayback capture
7blogs citing
First cited
6 May 2026
Most recent
10 Sept 2026
Rank this month
not ranked
Links from this post
2
Words captured
4,634

Cited by

Blog
In the post
Date
Anthropic
An alignment assessment of recent cybersecurity incidents Read at the source ↗
10 Sept 2026
tylercrosse.com
Transformer Circuits Thread Notes Read at the source ↗
12 Jul 2026
turntrout.com
How Robust Are Natural Language Autoencoders to Initialization? Read at the source ↗
9 Jul 2026
yogesh.bearblog.dev
Can You Hide From a Natural Language Autoencoder? Read at the source ↗
17 Jun 2026
boxobark.ing
Claude prompts will soon directly poison lungs in South Memphis Read at the source ↗
18 May 2026
Philipp D. Dubach - Finance, Tech & Strategy
What Claude Thinks But Doesn't Say Read at the source ↗
10 May 2026
marktechpost.com
Anthropic Introduces Natural Language Autoencoders That Convert Claude’s Internal Activations Directly into Human-Readable Text Explanations Read at the source ↗
8 May 2026
Anthropic
Natural Language Autoencoders: Turning Claude’s thoughts into text Read at the source ↗
6 May 2026

Links from this post

Link
Host
https://github.com/kitft/natural_language_autoencoders
github.com
https://www.neuronpedia.org/nla
neuronpedia.org