John's Unnamed Lab

AI & ML interests

Safety, Interpretability, Control, Representations, Understanding

Recent Activity

LorenaYannnnn  updated a model about 3 hours ago

hewittlab/monitor_cot_for_hack-baseline-length-explicit_intent-nemotron-3-nano-30b-a3b

LorenaYannnnn  updated a collection about 3 hours ago

Reward Misspecification Degrades CoT Monitorability

LorenaYannnnn  published a model about 3 hours ago

hewittlab/monitor_cot_for_hack-baseline-length-explicit_intent-nemotron-3-nano-30b-a3b

View all activity

Team members 2