Benchmark

MASK

MASK is a collection of 1000 questions measuring whether models faithfully report their beliefs when pressured to lie. It operationalizes deception as the…

Modality
text
Categories
reasoning, safety
Openness
unknown
Source
llm_stats
Reported scores
1

Reported scores

llm_stats

ModelOrganizationReported valueReportedEvidence
Grok-4.1 ThinkingxAI0.512025-11-17evidence

Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.