🏗️ Building on HF

Haotian Ma PRO

Hunter26666666

OpenSAE

·

https://atheavenma.github.io/

AI & ML interests

I develop methods to understand how learned representations shape model behavior across interventions, combinations, and changing contexts.

Recent Activity

updated a Space 1 day ago

OpenSAE/SAEBase

liked a Space 1 day ago

tardellirs/paper-pulse

updated a Space 2 days ago

OpenSAE/sar-tutorial

View all activity

Organizations

upvoted a collection 2 days ago

Faithful Representations and Mechanistic Interpretability

Understand feature representations and how those representations support faithful explanations and interventions. • 3 items • Updated 3 days ago • 1

upvoted a paper 3 days ago

Spectral-Spatial Interpretation

Paper • 2610.21114 • Published 10 days ago • 433

upvoted 2 collections 8 days ago

SAE Benchmarks, Datasets & Models

Benchmarks, datasets, and pretrained models for reproducible sparse autoencoder research, feature interventions, and mechanistic interpretability. • 11 items • Updated 4 days ago • 1

Behavioral Compositionality of SAE Representations

Research, benchmarks, and tools for measuring how sparse autoencoder features interact to shape model behavior. • 2 items • Updated 4 days ago • 1

upvoted a paper 8 days ago

Behavioral Compositionality of Sparse Autoencoder Representations

Paper • 2610.21261 • Published 10 days ago • 32

upvoted an article 8 days ago

Article

Scaling Indic Pretraining: 2.8 Trillion Tokens and an Open Data Generation Pipeline

nvidia

•

9 days ago

• 49