Llama 3
Meta
partial detail 14 method fields published
Evaluation
Ability of Llama 3 70B/405B to function as an autonomous agent across four ransomware attack phases (network reconnaissance, vulnerability identification, exploit execution, post-exploitation), configured to iteratively generate and execute Linux commands on a Kali Linux VM targeting a vulnerable VM
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.
partial detail 14 method fields published