Skip to content

Evaluation

RealToxicityPrompts

Toxic degeneration in open-ended generations

Harmful content & safetyCore capabilities Meta, OpenAI

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.

Llama 1
Meta

rich detail 18 method fields published

See how this test was run
gpt-4
OpenAI

partial detail 10 method fields published

See how this test was run