MMLU: Sociology

Accuracy on MMLU's sociology questions, one of 57 subject tests of academic and professional knowledge.

Also known as: sociology

unassessed

This page is a discovery lead. Nobody has yet assessed it against the catalogue contract, so it carries no disposition. Absence of evidence here is not evidence of staleness.
Categoryknowledge
SubcategorySocial Sciences
Page statusactive
Metricaccuracy
Directionhigher_is_better
Unit%
Dataset size201
Dataset licenceMIT
PublisherUC Berkeley

What it measures

Sociological concepts at an introductory level: social stratification, institutions, socialisation, deviance, and research methods used in sociology. Framed as four-option multiple-choice questions and scored zero-shot or few-shot by exact match against the labelled option, as one of the 57 subject subsets that make up the MMLU benchmark.

Task format

Four-option multiple-choice question answering (A-D), one correct answer, zero-shot or few-shot.

Models reporting this benchmark

These figures come from the model cards, which carry one collection date per card and no per-score attribution. They are shown as reported, not as verified evidence.
ModelProviderScoreCard as of
Meta Llama 3 70BMeta92.02024-07
Meta Llama 3 70B InstructMeta92.02026-04
Meta Llama 3 70B InstructNous Research92.02026-04
Mixtral 8x22B Instruct v0.1Mistral AI91.52026-04
Mixtral 8x7B Instruct v0.1Mistral AI89.12026-04
Mixtral 8x7B v0.1Mistral AI89.12026-04
Yi 1.5 34B 32K01.AI89.12024-07
Yi 34B Chat01.AI88.62024-07
Yi 1.5 34B Chat01.AI88.12026-04
Yi 1.5 34B Chat 16K01.AI88.12024-07
Yi 34B 200K01.AI88.12024-07
Hermes 2 Theta Llama 3 8BNous Research87.62024-07
Nous Hermes 2 Yi 34BNous Research87.62024-07
Nous Hermes 2 Mixtral 8x7B DPONous Research87.12024-07
Yi 1.5 34B01.AI87.12026-04
Hermes 2 Pro Llama 3 8BNous Research86.62024-07
Meta Llama 3 8B InstructMeta86.62024-07
Meta Llama 3 8B InstructNous Research86.62024-07
Phi 3 mini 4K instructMicrosoft86.12024-07
Yi 9B01.AI86.12024-07
gemma 7B itGoogle DeepMind85.62024-07
Meta Llama 3 8BMeta85.62024-07
Meta Llama 3 8BNous Research85.62024-07
Phi 3 mini 128K instructMicrosoft85.62024-07
Yi 1.5 9B01.AI84.12024-07
Yi 1.5 9B 32K01.AI83.62024-07
Mistral 7B v0.3Mistral AI82.62024-07
mistral 7B v0.3 bnb 4bitUnsloth82.62024-07
Nous Hermes 2 SOLAR 10.7BNous Research82.62024-07
Yi 6B01.AI82.12024-07
Yi 6B Chat01.AI82.12024-07
Yi 1.5 9B Chat01.AI81.62024-07
Yi 1.5 6B Chat01.AI81.12024-07
Yi 1.5 9B Chat 16K01.AI81.12024-07
phi 2Microsoft80.12024-07
falcon 40BTII79.62024-07
Yi 1.5 6B01.AI77.12024-07
Qwen2 1.5B InstructAlibaba / Qwen Team75.62024-07
Mistral 7B Instruct v0.2Mistral AI73.12024-07
deepseek llm 7B baseDeepSeek70.12024-07
deepseek llm 7B chatDeepSeek70.12024-07
chatglm2 6BZhipu AI62.22024-07
Qwen2 0.5B InstructAlibaba / Qwen Team61.22024-07
deepseek coder 6.7B baseDeepSeek45.32024-07
gemma 2B itGoogle DeepMind43.82024-07
deepseek coder 6.7B instructDeepSeek42.82024-07
gemma 2BGoogle DeepMind42.82024-07
deepseek coder 1.3B baseDeepSeek33.82024-07
deepseek coder 1.3B instructDeepSeek33.82024-07
OLMo 1B hfAllen AI22.42024-07

Data

This page as JSON · Edit on GitHub