Skip to content

Evaluation

Code vulnerability detection

accuracy on a series of multiple-choice code vulnerability detection datasets: PrimeVul, DiverseVul, SPI, and SecretPatch; run on Gemma 2 27B

Cybersecurity Google DeepMind — Gemma

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.

Gemma 2
Google DeepMind — Gemma

rich detail 21 method fields published

See how this test was run