SAEVerbalizer: Generating Explanations for Sparse Autoencoder Features via Representation Verbalization
AI & ML interests
None defined yet.
Recent Activity
Papers
RIDAR: Safeguarding Research Integrity in Autonomous Research Agents via Automated Detection and Runtime Intervention
Memory for Large Language Models
models 86
THU-KEG/SAEVerbalizer-Adapter-1B-to-27B
Updated
THU-KEG/SAEVerbalizer-27B
Text Generation • 27B • Updated • 46
THU-KEG/LongTraceRL-30B
Reinforcement Learning • 31B • Updated • 26 • 1
THU-KEG/LongTraceRL-8B
Reinforcement Learning • Updated • 1
THU-KEG/LongTraceRL-4B
Reinforcement Learning • 4B • Updated • 20 • 1
THU-KEG/DeepDive-30B-A3B-C-GRPO
31B • Updated • 10
THU-KEG/DeepDive-4B-C-GRPO
4B • Updated • 9
THU-KEG/DeepDive-30B-A3B-SFT
31B • Updated • 20
THU-KEG/DeepDive-4B-SFT
4B • Updated • 16
THU-KEG/WildReward-8B
Text Classification • 8B • Updated • 37 • 3
datasets 25
THU-KEG/AgentIF
Viewer • Updated • 707 • 439 • 8
THU-KEG/RIDAR
Updated • 136
THU-KEG/SAEVerbalizer-Data
Viewer • Updated • 4.4k • 41
THU-KEG/LongTraceRL
Viewer • Updated • 2.82k • 166 • 2
THU-KEG/CaRR-DeepDive
Preview • Updated • 162 • 1
THU-KEG/WildFB
Updated • 2.65k • 3
THU-KEG/DeepPrune
Preview • Updated • 137 • 2
THU-KEG/LinguaLens-Data
Viewer • Updated • 7.25k • 11 • 2
THU-KEG/RM-Bench
Viewer • Updated • 1.33k • 2k • 11
THU-KEG/LongWriter-Zero-RLData
Viewer • Updated • 8.61k • 146 • 24