Back to article
transformer-circuits.pub

Natural Language Autoencoders Produce Unsupervised Explanations of LLM Activations (opens in new tab)

Covered by 3 sources See all sources covering this story including DEV Community, lesswrong.comDiscussed on Hacker News

Covered in 5 articles

DEV Community·

Anthropic NLAs: Reading Claude's Hidden Thoughts in 2026

Discussed on DEV
Feeds
lesswrong.com·

[Linkpost] How Transparent Is DiffusionGemma (and why it matters)

Feeds
lesswrong.com·

SAE It Across Models: Explaining Features With Foreign NLA Verbalizers

Feeds
lesswrong.com·

Logits as a new monitor for evaluation awareness

Discussed on Hacker News
Feeds

In other languages

habr.com·

Frontend Status: свежий дайджест фронтенда и AI — 04.06.2026

Feeds

Keyboard Shortcuts

Navigation

Next / previous post
j/k
Open post
oorEnter
Preview post
v

Post Actions

Love post
a
Like post
l
Dislike post
d
Undo reaction
u
Save / unsave
s

Recommendations

Add interest / feed
Enter
Not interested
x

Go to

Home
gh
Interests
gi
Feeds
gf
Likes
gl
History
gy
Changelog
gc
Settings
gs
Discover
gb
Search
/

General

Show this help
?
Submit feedback
!
Close modal / unfocus
Esc

Press ? anytime to show this help