Interpretability

Feeds to Scour
SubscribedAll
Scoured 81 posts in 5.8 ms

Mechanistic Interpretability: The Key to Trusting Agentic AI

 🤖AI Agents  Content type: Discussion
bradenkelley.com·

Query Lens: Interpreting Sparse Key-Value Features with Indirect Effects

 🧠AI Research  Content type: Academic
arxiv.org·

You Can Catch Sleeper Agents by Teaching Another Model to Imitate Them

 💬LLMs
lesswrong.com·

Compositional and interpretable representation of histology using AI foundation models and sparse autoencoders

 📉Deep Learning  Content type: Academic
biorxiv.org·

Is the Space Pope Reptilian?

 🔄Transformers  Content type: News
tearsinrain.ai··Hacker News

princezuda/-RequiemGPT-: Fully open source and open weights built and trained by fable five with one prompt. An experience in how AI actually works

 🔥PyTorch  Content type: Code
github.com··Hacker News

Can You Hide From a Natural Language Autoencoder?

 ⚙️Model Training  Content type: Blog
yogesh.bearblog.dev·

Arithmetic Without Numbers – How LLMs Do Math

 💬LLMs
alvaro-videla.com··Hacker News

How LLMs work | Practical Leaders

 💬LLMs
practical-leaders.com··Hacker News

Interpreting and Steering a Text-to-Speech Language Model with Sparse Autoencoders

 💬LLMs  Content type: Academic
arxiv.org·

Playing with Vision Embeddings

 📐Scaling Laws
prestonbjensen.com··Hacker News

BioByte 162: The Hype of Virtual Cells, ESMC's AlphaFold3-Like Performance, and the Prediction of Antibody Non-Specificity

 🖥️ML Systems  Content type: Blog
decodingbio.substack.com··Substack

The Standard Interpretable Model: A general theory of interpretable machine learning to deductively design interpretable methods using Lagrangian mechanics

 🖥️ML Systems  Content type: Academic
arxiv.org·

Who Elected Anthropic?

 📐Scaling Laws  Content type: Blog
vizierprime.substack.com··Substack

Machinic Psychopharmacology: Do LLMs Self-Medicate?

 ⚙️Model Training
lesswrong.com··Hacker News

Trajectory Geometry of Transformer Representations Across Layers

 🔄Transformers  Content type: Academic
arxiv.org·

Coelho Mollo and Millière: The Vector Grounding Problem

 ⚙️Model Training
philosophyofbrains.com·

scMTG reconstructs single-cell temporal dynamics with Markov transition generators

 📐Scaling Laws  Content type: Academic
biorxiv.org·

FoldSAE: Learning to Steer Protein Folding Through Sparse Representations

 📉Deep Learning  Content type: Academic
arxiv.org·

How LLMs Actually Work: A Friendly Map for Humans • oreoro

 🔄Transformers
oreoro.github.io··Hacker News

Keyboard Shortcuts

Navigation

Next / previous item
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
Browse
gb
Search
/

General

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

Press ? anytime to show this help