pith. sign in

Ollie Jaffe

Identifiers

  • name variant Ollie Jaffe 0.60 · backfill

Papers (1)

  1. AI Sandbagging: Language Models can Strategically Underperform on Evaluations cs.AI · 2024 · author #3

Mentions

  • 2406.07358 #3 · arxiv_oai · confidence 0.70 Ollie Jaffe

Frequent Coauthors