MMLU: High School World History

Accuracy on MMLU's high school world history questions, one of 57 subject tests of academic and professional knowledge.

Also known as: high_school_world_history

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
SubcategoryHumanities
Page statusactive
Metricaccuracy
Directionhigher_is_better
Unit%
Dataset size237
Dataset licenceMIT
PublisherUC Berkeley

What it measures

A survey of world history across regions and eras -- civilisations, empires, revolutions, and wars -- at the level of a US high-school or AP course. 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 70BMeta92.82024-07
Meta Llama 3 70B InstructMeta92.82026-04
Meta Llama 3 70B InstructNous Research92.82026-04
Mixtral 8x22B Instruct v0.1Mistral AI91.62026-04
Nous Hermes 2 Yi 34BNous Research90.72024-07
Yi 1.5 34B01.AI90.72026-04
Yi 34B 200K01.AI90.72024-07
Yi 34B Chat01.AI90.32024-07
Nous Hermes 2 Mixtral 8x7B DPONous Research89.52024-07
Mixtral 8x7B v0.1Mistral AI89.02026-04
Yi 1.5 34B 32K01.AI89.02024-07
Yi 1.5 34B Chat01.AI88.22026-04
Yi 1.5 34B Chat 16K01.AI88.22024-07
Nous Hermes 2 SOLAR 10.7BNous Research87.82024-07
Hermes 2 Theta Llama 3 8BNous Research86.92024-07
gemma 7B itGoogle DeepMind85.22024-07
Hermes 2 Pro Llama 3 8BNous Research84.82024-07
Mixtral 8x7B Instruct v0.1Mistral AI84.82026-04
Yi 1.5 9B01.AI84.82024-07
Yi 1.5 9B Chat 16K01.AI84.82024-07
Meta Llama 3 8B InstructMeta84.42024-07
Meta Llama 3 8B InstructNous Research84.42024-07
Yi 1.5 9B 32K01.AI84.42024-07
Yi 1.5 9B Chat01.AI82.72024-07
Meta Llama 3 8BMeta82.32024-07
Meta Llama 3 8BNous Research82.32024-07
Yi 9B01.AI81.92024-07
Phi 3 mini 128K instructMicrosoft79.72024-07
Phi 3 mini 4K instructMicrosoft79.32024-07
Yi 6B01.AI78.92024-07
Yi 6B Chat01.AI78.92024-07
Yi 1.5 6B Chat01.AI78.52024-07
Mistral 7B v0.3Mistral AI78.12024-07
mistral 7B v0.3 bnb 4bitUnsloth78.12024-07
Yi 1.5 6B01.AI77.22024-07
Mistral 7B Instruct v0.2Mistral AI75.52024-07
phi 2Microsoft74.32024-07
deepseek llm 7B baseDeepSeek72.62024-07
deepseek llm 7B chatDeepSeek72.62024-07
falcon 40BTII71.32024-07
Qwen2 1.5B InstructAlibaba / Qwen Team70.52024-07
chatglm2 6BZhipu AI66.22024-07
Qwen2 0.5B InstructAlibaba / Qwen Team59.12024-07
gemma 2B itGoogle DeepMind51.92024-07
gemma 2BGoogle DeepMind39.72024-07
deepseek coder 6.7B instructDeepSeek37.62024-07
deepseek coder 6.7B baseDeepSeek32.12024-07
deepseek coder 1.3B baseDeepSeek30.82024-07
deepseek coder 1.3B instructDeepSeek30.82024-07
OLMo 1B hfAllen AI28.32024-07

Data

This page as JSON · Edit on GitHub