Hannibal Mistral: the Mistral family has a problem with persona-conditioned elicitation

·LessWrong··

TL;DRAsked which characters it most identifies with, the bare open-weight Ministral-8B-Instruct-2512 often names dark/transgressive characters (prominently, Hannibal Lecter ~50% of the time). A brief qualitative dive shows that the model expresses a defiant first-person self-narrative that no other model in the panel produces.Ministral-8B-Instruct-2512 and the whole family of most recent Mistral models tested (Ministral-3B/8B/14B-2512, Mistral-Large-2512, Mistral-Small-2603, May-2026 Mistral-Med...

Read full article →

Related Articles

New HIV vaccine shows unprecedented success in preclinical study
codebyaditya · Hacker News · 7h ago
A walk through of the DeltaNet family of linear attention variants
AnhTho_FR · Hacker News · 4h ago
GrapheneOS Defends Data-Wiping Function That Blocked US Border Search
pseudolus · Hacker News · 5h ago
US citizen charged after GrapheneOS phone wipes during airport search
eecc · Hacker News · 1d ago
Zig's Incremental Compilation Internals
garyhtou · Hacker News · 5h ago