Gemini 1.5
Google DeepMind — Gemini
rich detail 8 method fields published
Evaluation
Baseline assurance evaluations (Gemini 1.5)
Arms-length assurance evaluations for responsibility governance decision-making: held-out prompt sets, content policies and representational harms, all modalities, performed for 1.5 Pro and Flash
How it was run
Open a release to see the setup, scoring, and other details that the lab published. Only fields the lab actually disclosed are shown.