Reasoning
AGI-Elo
Item-level MMLU predictions released for AGI-Elo, a study of joint task difficulty and AI-system competency. This curation covers the released question-answering observations, not derived Elo ratings.
13,953items
20subjects
Apache-2.0license
generaldomain
reasoningdomain
textmodality
item-level responses released
Saturation status: Yes
Response matrix
Loading response matrix…
Correct (1)Incorrect (0)Unobserved
Scale: 1 = correct · 0 = incorrect