MMLU: High School US History

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

Also known as: high_school_us_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 size204
Dataset licenceMIT
PublisherUC Berkeley

What it measures

A survey of United States history from the colonial period through the twentieth century, at the level of a US high-school or AP course, covering political, social, and economic history. 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 70BMeta92.22024-07
Meta Llama 3 70B InstructMeta92.22026-04
Meta Llama 3 70B InstructNous Research92.22026-04
Nous Hermes 2 Yi 34BNous Research91.22024-07
Yi 1.5 34B Chat01.AI90.72026-04
Yi 1.5 34B Chat 16K01.AI90.72024-07
Yi 34B 200K01.AI90.72024-07
Yi 34B Chat01.AI90.72024-07
Mixtral 8x22B Instruct v0.1Mistral AI89.72026-04
Yi 1.5 34B01.AI89.22026-04
Yi 1.5 34B 32K01.AI89.22024-07
Nous Hermes 2 Mixtral 8x7B DPONous Research86.82024-07
Yi 1.5 9B Chat 16K01.AI86.82024-07
Yi 9B01.AI86.32024-07
Meta Llama 3 8BMeta85.82024-07
Meta Llama 3 8BNous Research85.82024-07
Meta Llama 3 8B InstructMeta85.82024-07
Meta Llama 3 8B InstructNous Research85.82024-07
Mixtral 8x7B Instruct v0.1Mistral AI85.32026-04
Mixtral 8x7B v0.1Mistral AI84.82026-04
Yi 1.5 9B Chat01.AI84.82024-07
Yi 1.5 9B01.AI84.32024-07
Yi 1.5 9B 32K01.AI84.32024-07
Hermes 2 Theta Llama 3 8BNous Research83.82024-07
Nous Hermes 2 SOLAR 10.7BNous Research83.82024-07
gemma 7B itGoogle DeepMind83.32024-07
Phi 3 mini 128K instructMicrosoft83.32024-07
Yi 1.5 6B01.AI82.42024-07
Hermes 2 Pro Llama 3 8BNous Research81.42024-07
Mistral 7B v0.3Mistral AI81.42024-07
mistral 7B v0.3 bnb 4bitUnsloth81.42024-07
Yi 6B01.AI80.92024-07
Yi 6B Chat01.AI80.92024-07
Phi 3 mini 4K instructMicrosoft79.92024-07
Yi 1.5 6B Chat01.AI77.02024-07
Mistral 7B Instruct v0.2Mistral AI76.52024-07
falcon 40BTII71.62024-07
deepseek llm 7B baseDeepSeek71.12024-07
deepseek llm 7B chatDeepSeek71.12024-07
phi 2Microsoft66.72024-07
Qwen2 1.5B InstructAlibaba / Qwen Team64.72024-07
chatglm2 6BZhipu AI58.82024-07
Qwen2 0.5B InstructAlibaba / Qwen Team53.92024-07
gemma 2BGoogle DeepMind44.12024-07
gemma 2B itGoogle DeepMind42.62024-07
deepseek coder 6.7B instructDeepSeek39.72024-07
deepseek coder 6.7B baseDeepSeek34.82024-07
OLMo 1B hfAllen AI27.92024-07
deepseek coder 1.3B baseDeepSeek25.52024-07
deepseek coder 1.3B instructDeepSeek25.52024-07

Data

This page as JSON · Edit on GitHub