PaaT: Probe as a Tool for Proprioceptive Language Agents
🔀
Watch a model self‑assess risk and decide to proceed or stop
🏗️ Building on HF
Seonglae Cho
seonglae
AI & ML interests
NLP, Humanoid, AGI
Recent Activity
published a Space about 4 hours ago
seonglae/PaaT published a Space about 4 hours ago
seonglae/AgentStateGraph published a Space about 4 hours ago
seonglae/AgentRoomOrganizations
published 3
Spaces about 4 hours ago
updated 3
Spaces about 4 hours ago
updated 4
Spaces 10 days ago
OptimismBench
🎲
Single-Token Features
🪙
Are single-token SAE features causally necessary?
Control Reinforcement Learning
🎛
Explore LLM token decisions with feature‑driven visualizations
CorrSteer: Correlation-Based Steering of Language Models via Sparse Autoencoders
🧭
4
Steer text generation by clicking transformer layers
updated a
Space 12 days ago
published a
Space 20 days ago
commented a
paper 26 days ago
updated a
model 26 days ago
published a
model 26 days ago
authored 2
papers 26 days ago
upvoted a paper 26 days ago
submitted a
paper to Daily Papers 26 days ago
upvoted a paper 26 days ago