Automata from Agent Traces
🔁
Explore interactive FSMs from LLM agent traces
🏗️ Building on HF
Seonglae Cho
seonglae
AI & ML interests
NLP, Humanoid, AGI
Recent Activity
updated a Space 2 days ago
seonglae/AgentStateGraph updated a Space 4 days ago
seonglae/AgentRoom updated a Space 4 days ago
seonglae/ConfidenceManifoldOrganizations
updated a
Space 2 days ago
updated 5
Spaces 4 days ago
AgentRoom
🏠
Explore AI agents collaborating in a shared coding workspace
Confidence Manifold
🧭
OptimismBench
🎲
Control Reinforcement Learning
🎛
Explore LLM token decisions with feature‑driven visualizations
PaaT: Probe as a Tool for Proprioceptive Language Agents
🔀
Visualize model self‑risk probing and decision flow
updated 2
Spaces 6 days ago
published 3
Spaces 28 days ago
updated 3
Spaces 28 days ago
updated 6
Spaces about 1 month 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 language model output by clicking visual layers
Confidence Manifold
🧭
Control Reinforcement Learning
🎛
Explore LLM token decisions with feature‑driven visualizations