SARM - a Schrieffer Collection

Schrieffer 's Collections

SARM

updated 4 days ago

[AAAI 2026 Oral] open weights

Schrieffer/Llama-SARM-4B

Reinforcement Learning • 5B • Updated Dec 11, 2025 • 46 • 2
Schrieffer/Llama-SARM-4B-PostSAEPretrain

Feature Extraction • 5B • Updated Dec 11, 2025 • 30 • 2
Interpretable Reward Model via Sparse Autoencoder

Paper • 2508.08746 • Published Aug 12, 2025 • 2