Llama 3
Meta
partial detail 12 method fields published
Evaluation
Multilingual safety evaluation
Safety knowledge transfer across languages on an internal per-language benchmark: Llama 405B with and without Llama Guard vs two competing systems, plus violation/false-refusal comparison vs a competing standalone open-source model
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.