Llama 3
Meta
partial detail 12 method fields published
Evaluation
CyberSecEval insecure coding testing framework
Propensity to generate insecure code measured by CyberSecEval's insecure coding testing framework (Bhatt et al. 2023) for Llama 3 8B/70B/405B; the model card additionally measures propensity to suggest insecure code as a coding assistant
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.