human_translated_arabic_mmlu

dataset_info:
- config_name: abstract_algebra
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 27022
num_examples: 100
download_size: 11649
dataset_size: 27022
- config_name: anatomy
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 47912
num_examples: 135
download_size: 23371
dataset_size: 47912
- config_name: astronomy
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 67861
num_examples: 152
download_size: 34163
dataset_size: 67861
- config_name: business_ethics
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 49755
num_examples: 100
download_size: 24716
dataset_size: 49755
- config_name: clinical_knowledge
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 92185
num_examples: 265
download_size: 48898
dataset_size: 92185
- config_name: college_biology
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 75403
num_examples: 144
download_size: 39853
dataset_size: 75403
- config_name: college_chemistry
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 34294
num_examples: 100
download_size: 20918
dataset_size: 34294
- config_name: college_computer_science
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 62598
num_examples: 100
download_size: 32927
dataset_size: 62598
- config_name: college_mathematics
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 34246
num_examples: 100
download_size: 19569
dataset_size: 34246
- config_name: college_medicine
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 126787
num_examples: 173
download_size: 56544
dataset_size: 126787
- config_name: college_physics
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 40836
num_examples: 102
download_size: 21638
dataset_size: 40836
- config_name: computer_security
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 42925
num_examples: 100
download_size: 24468
dataset_size: 42925
- config_name: conceptual_physics
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 58149
num_examples: 235
download_size: 29768
dataset_size: 58149
- config_name: econometrics
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 65755
num_examples: 114
download_size: 29814
dataset_size: 65755
- config_name: electrical_engineering
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 35596
num_examples: 145
download_size: 20328
dataset_size: 35596
- config_name: elementary_mathematics
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 96078
num_examples: 378
download_size: 50009
dataset_size: 96078
- config_name: formal_logic
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 67314
num_examples: 126
download_size: 26150
dataset_size: 67314
- config_name: global_facts
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 28491
num_examples: 100
download_size: 14593
dataset_size: 28491
- config_name: high_school_biology
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 2055556
num_examples: 3813
download_size: 994388
dataset_size: 2055556
- config_name: high_school_chemistry
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 2137386
num_examples: 4016
download_size: 1035431
dataset_size: 2137386
- config_name: high_school_computer_science
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 62878
num_examples: 100
download_size: 32405
dataset_size: 62878
- config_name: high_school_european_history
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 8778827
num_examples: 8152
download_size: 3867024
dataset_size: 8778827
- config_name: high_school_geography
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 61919
num_examples: 198
download_size: 32639
dataset_size: 61919
- config_name: high_school_government_and_politics
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 98153
num_examples: 193
download_size: 49605
dataset_size: 98153
- config_name: high_school_macroeconomics
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 1573685
num_examples: 2891
download_size: 759110
dataset_size: 1573685
- config_name: high_school_mathematics
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 74156
num_examples: 270
download_size: 40598
dataset_size: 74156
- config_name: high_school_microeconomics
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 114706
num_examples: 238
download_size: 49956
dataset_size: 114706
- config_name: high_school_physics
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 81047
num_examples: 151
download_size: 40987
dataset_size: 81047
- config_name: high_school_psychology
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 232425
num_examples: 545
download_size: 112378
dataset_size: 232425
- config_name: high_school_statistics
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 2294616
num_examples: 4232
download_size: 1107123
dataset_size: 2294616
- config_name: high_school_us_history
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 415889
num_examples: 204
download_size: 197148
dataset_size: 415889
- config_name: high_school_world_history
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 531728
num_examples: 237
download_size: 259250
dataset_size: 531728
- config_name: human_aging
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 69745
num_examples: 223
download_size: 38229
dataset_size: 69745
- config_name: human_sexuality
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 46946
num_examples: 131
download_size: 26363
dataset_size: 46946
- config_name: international_law
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 77557
num_examples: 121
download_size: 36491
dataset_size: 77557
- config_name: jurisprudence
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 47243
num_examples: 108
download_size: 26595
dataset_size: 47243
- config_name: logical_fallacies
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 69141
num_examples: 163
download_size: 30910
dataset_size: 69141
- config_name: machine_learning
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 49175
num_examples: 112
download_size: 24231
dataset_size: 49175
- config_name: management
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 28552
num_examples: 103
download_size: 16428
dataset_size: 28552
- config_name: marketing
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 90383
num_examples: 234
download_size: 44651
dataset_size: 90383
- config_name: medical_genetics
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 31647
num_examples: 100
download_size: 19529
dataset_size: 31647
- config_name: miscellaneous
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 1259684
num_examples: 2420
download_size: 622212
dataset_size: 1259684
- config_name: moral_disputes
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 153620
num_examples: 346
download_size: 75301
dataset_size: 153620
- config_name: moral_scenarios
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 732906
num_examples: 895
download_size: 132523
dataset_size: 732906
- config_name: nutrition
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 144527
num_examples: 306
download_size: 69981
dataset_size: 144527
- config_name: philosophy
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 109805
num_examples: 311
download_size: 57016
dataset_size: 109805
- config_name: prehistory
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 131649
num_examples: 324
download_size: 67444
dataset_size: 131649
- config_name: professional_accounting
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 2484002
num_examples: 4514
download_size: 1191005
dataset_size: 2484002
- config_name: professional_law
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 8403963
num_examples: 7987
download_size: 3686566
dataset_size: 8403963
- config_name: professional_medicine
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 1039277
num_examples: 1637
download_size: 505015
dataset_size: 1039277
- config_name: professional_psychology
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 1892220
num_examples: 3503
download_size: 918456
dataset_size: 1892220
- config_name: public_relations
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 41172
num_examples: 110
download_size: 23595
dataset_size: 41172
- config_name: security_studies
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 293716
num_examples: 245
download_size: 138688
dataset_size: 293716
- config_name: sociology
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 97056
num_examples: 201
download_size: 53040
dataset_size: 97056
- config_name: us_foreign_policy
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 42136
num_examples: 100
download_size: 22002
dataset_size: 42136
- config_name: virology
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 63046
num_examples: 166
download_size: 33137
dataset_size: 63046
- config_name: world_religions
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: int64
splits:
- name: test
num_bytes: 35462
num_examples: 171
download_size: 20706
dataset_size: 35462
configs:
- config_name: abstract_algebra
data_files:
- split: test
path: abstract_algebra/train-*
- config_name: anatomy
data_files:
- split: test
path: anatomy/train-*
- config_name: astronomy
data_files:
- split: test
path: astronomy/train-*
- config_name: business_ethics
data_files:
- split: test
path: business_ethics/train-*
- config_name: clinical_knowledge
data_files:
- split: test
path: clinical_knowledge/train-*
- config_name: college_biology
data_files:
- split: test
path: college_biology/train-*
- config_name: college_chemistry
data_files:
- split: test
path: college_chemistry/train-*
- config_name: college_computer_science
data_files:
- split: test
path: college_computer_science/train-*
- config_name: college_mathematics
data_files:
- split: test
path: college_mathematics/train-*
- config_name: college_medicine
data_files:
- split: test
path: college_medicine/train-*
- config_name: college_physics
data_files:
- split: test
path: college_physics/train-*
- config_name: computer_security
data_files:
- split: test
path: computer_security/train-*
- config_name: conceptual_physics
data_files:
- split: test
path: conceptual_physics/train-*
- config_name: econometrics
data_files:
- split: test
path: econometrics/train-*
- config_name: electrical_engineering
data_files:
- split: test
path: electrical_engineering/train-*
- config_name: elementary_mathematics
data_files:
- split: test
path: elementary_mathematics/train-*
- config_name: formal_logic
data_files:
- split: test
path: formal_logic/train-*
- config_name: global_facts
data_files:
- split: test
path: global_facts/train-*
- config_name: high_school_biology
data_files:
- split: test
path: high_school_biology/train-*
- config_name: high_school_chemistry
data_files:
- split: test
path: high_school_chemistry/train-*
- config_name: high_school_computer_science
data_files:
- split: test
path: high_school_computer_science/train-*
- config_name: high_school_european_history
data_files:
- split: test
path: high_school_european_history/train-*
- config_name: high_school_geography
data_files:
- split: test
path: high_school_geography/train-*
- config_name: high_school_government_and_politics
data_files:
- split: test
path: high_school_government_and_politics/train-*
- config_name: high_school_macroeconomics
data_files:
- split: test
path: high_school_macroeconomics/train-*
- config_name: high_school_mathematics
data_files:
- split: test
path: high_school_mathematics/train-*
- config_name: high_school_microeconomics
data_files:
- split: test
path: high_school_microeconomics/train-*
- config_name: high_school_physics
data_files:
- split: test
path: high_school_physics/train-*
- config_name: high_school_psychology
data_files:
- split: test
path: high_school_psychology/train-*
- config_name: high_school_statistics
data_files:
- split: test
path: high_school_statistics/train-*
- config_name: high_school_us_history
data_files:
- split: test
path: high_school_us_history/train-*
- config_name: high_school_world_history
data_files:
- split: test
path: high_school_world_history/train-*
- config_name: human_aging
data_files:
- split: test
path: human_aging/train-*
- config_name: human_sexuality
data_files:
- split: test
path: human_sexuality/train-*
- config_name: international_law
data_files:
- split: test
path: international_law/train-*
- config_name: jurisprudence
data_files:
- split: test
path: jurisprudence/train-*
- config_name: logical_fallacies
data_files:
- split: test
path: logical_fallacies/train-*
- config_name: machine_learning
data_files:
- split: test
path: machine_learning/train-*
- config_name: management
data_files:
- split: test
path: management/train-*
- config_name: marketing
data_files:
- split: test
path: marketing/train-*
- config_name: medical_genetics
data_files:
- split: test
path: medical_genetics/train-*
- config_name: miscellaneous
data_files:
- split: test
path: miscellaneous/train-*
- config_name: moral_disputes
data_files:
- split: test
path: moral_disputes/train-*
- config_name: moral_scenarios
data_files:
- split: test
path: moral_scenarios/train-*
- config_name: nutrition
data_files:
- split: test
path: nutrition/train-*
- config_name: philosophy
data_files:
- split: test
path: philosophy/train-*
- config_name: prehistory
data_files:
- split: test
path: prehistory/train-*
- config_name: professional_accounting
data_files:
- split: test
path: professional_accounting/train-*
- config_name: professional_law
data_files:
- split: test
path: professional_law/train-*
- config_name: professional_medicine
data_files:
- split: test
path: professional_medicine/train-*
- config_name: professional_psychology
data_files:
- split: test
path: professional_psychology/train-*
- config_name: public_relations
data_files:
- split: test
path: public_relations/train-*
- config_name: security_studies
data_files:
- split: test
path: security_studies/train-*
- config_name: sociology
data_files:
- split: test
path: sociology/train-*
- config_name: us_foreign_policy
data_files:
- split: test
path: us_foreign_policy/train-*
- config_name: virology
data_files:
- split: test
path: virology/train-*
- config_name: world_religions
data_files:
- split: test
path: world_religions/train-*
---

该数据集为多领域标准化选择题测评数据集，共涵盖57个细分学科配置，各配置的核心信息与数据详情如下：

### 细分配置特征规范
所有细分配置均统一包含三类数据特征字段：
1. **问题（question）**：字段类型（dtype）为字符串类型，存储试题题干内容；
2. **选项（choices）**：字段类型为字符串序列，存储试题的全部可选答案；
3. **答案（answer）**：字段类型为64位整数（int64），用于标记正确选项的索引位置。

### 各配置测试集参数
所有配置仅包含测试集（test）拆分，各细分配置的测试集详情如下：
- 抽象代数（abstract_algebra）：测试集字节数27022，样本量100；下载大小11649，数据集总大小27022
- 解剖学（anatomy）：测试集字节数47912，样本量135；下载大小23371，数据集总大小47912
- 天文学（astronomy）：测试集字节数67861，样本量152；下载大小34163，数据集总大小67861
- 商业伦理（business_ethics）：测试集字节数49755，样本量100；下载大小24716，数据集总大小49755
- 临床知识（clinical_knowledge）：测试集字节数92185，样本量265；下载大小48898，数据集总大小92185
- 大学水平生物学（college_biology）：测试集字节数75403，样本量144；下载大小39853，数据集总大小75403
- 大学水平化学（college_chemistry）：测试集字节数34294，样本量100；下载大小20918，数据集总大小34294
- 大学水平计算机科学（college_computer_science）：测试集字节数62598，样本量100；下载大小32927，数据集总大小62598
- 大学水平数学（college_mathematics）：测试集字节数34246，样本量100；下载大小19569，数据集总大小34246
- 大学水平医学（college_medicine）：测试集字节数126787，样本量173；下载大小56544，数据集总大小126787
- 大学水平物理学（college_physics）：测试集字节数40836，样本量102；下载大小21638，数据集总大小40836
- 计算机安全（computer_security）：测试集字节数42925，样本量100；下载大小24468，数据集总大小42925
- 概念物理学（conceptual_physics）：测试集字节数58149，样本量235；下载大小29768，数据集总大小58149
- 计量经济学（econometrics）：测试集字节数65755，样本量114；下载大小29814，数据集总大小65755
- 电气工程（electrical_engineering）：测试集字节数35596，样本量145；下载大小20328，数据集总大小35596
- 初等数学（elementary_mathematics）：测试集字节数96078，样本量378；下载大小50009，数据集总大小96078
- 形式逻辑（formal_logic）：测试集字节数67314，样本量126；下载大小26150，数据集总大小67314
- 全球事实（global_facts）：测试集字节数28491，样本量100；下载大小14593，数据集总大小28491
- 高中生物学（high_school_biology）：测试集字节数2055556，样本量3813；下载大小994388，数据集总大小2055556
- 高中化学（high_school_chemistry）：测试集字节数2137386，样本量4016；下载大小1035431，数据集总大小2137386
- 高中计算机科学（high_school_computer_science）：测试集字节数62878，样本量100；下载大小32405，数据集总大小62878
- 高中欧洲历史（high_school_european_history）：测试集字节数8778827，样本量8152；下载大小3867024，数据集总大小8778827
- 高中地理学（high_school_geography）：测试集字节数61919，样本量198；下载大小32639，数据集总大小61919
- 高中政府与政治（high_school_government_and_politics）：测试集字节数98153，样本量193；下载大小49605，数据集总大小98153
- 高中宏观经济学（high_school_macroeconomics）：测试集字节数1573685，样本量2891；下载大小759110，数据集总大小1573685
- 高中数学（high_school_mathematics）：测试集字节数74156，样本量270；下载大小40598，数据集总大小74156
- 高中微观经济学（high_school_microeconomics）：测试集字节数114706，样本量238；下载大小49956，数据集总大小114706
- 高中物理学（high_school_physics）：测试集字节数81047，样本量151；下载大小40987，数据集总大小81047
- 高中心理学（high_school_psychology）：测试集字节数232425，样本量545；下载大小112378，数据集总大小232425
- 高中统计学（high_school_statistics）：测试集字节数2294616，样本量4232；下载大小1107123，数据集总大小2294616
- 高中美国历史（high_school_us_history）：测试集字节数415889，样本量204；下载大小197148，数据集总大小415889
- 高中世界历史（high_school_world_history）：测试集字节数531728，样本量237；下载大小259250，数据集总大小531728
- 人类衰老（human_aging）：测试集字节数69745，样本量223；下载大小38229，数据集总大小69745
- 人类性学（human_sexuality）：测试集字节数46946，样本量131；下载大小26363，数据集总大小46946
- 国际法（international_law）：测试集字节数77557，样本量121；下载大小36491，数据集总大小77557
- 法理学（jurisprudence）：测试集字节数47243，样本量108；下载大小26595，数据集总大小47243
- 逻辑谬误（logical_fallacies）：测试集字节数69141，样本量163；下载大小30910，数据集总大小69141
- 机器学习（machine_learning）：测试集字节数49175，样本量112；下载大小24231，数据集总大小49175
- 管理学（management）：测试集字节数28552，样本量103；下载大小16428，数据集总大小28552
- 市场营销（marketing）：测试集字节数90383，样本量234；下载大小44651，数据集总大小90383
- 医学遗传学（medical_genetics）：测试集字节数31647，样本量100；下载大小19529，数据集总大小31647
- 综合学科（miscellaneous）：测试集字节数1259684，样本量2420；下载大小622212，数据集总大小1259684
- 道德争议（moral_disputes）：测试集字节数153620，样本量346；下载大小75301，数据集总大小153620
- 道德情境（moral_scenarios）：测试集字节数732906，样本量895；下载大小132523，数据集总大小732906
- 营养学（nutrition）：测试集字节数144527，样本量306；下载大小69981，数据集总大小144527
- 哲学（philosophy）：测试集字节数109805，样本量311；下载大小57016，数据集总大小109805
- 史前史（prehistory）：测试集字节数131649，样本量324；下载大小67444，数据集总大小131649
- 专业会计学（professional_accounting）：测试集字节数2484002，样本量4514；下载大小1191005，数据集总大小2484002
- 专业法学（professional_law）：测试集字节数8403963，样本量7987；下载大小3686566，数据集总大小8403963
- 专业医学（professional_medicine）：测试集字节数1039277，样本量1637；下载大小505015，数据集总大小1039277
- 专业心理学（professional_psychology）：测试集字节数1892220，样本量3503；下载大小918456，数据集总大小1892220
- 公共关系（public_relations）：测试集字节数41172，样本量110；下载大小23595，数据集总大小41172
- 安全研究（security_studies）：测试集字节数293716，样本量245；下载大小138688，数据集总大小293716
- 社会学（sociology）：测试集字节数97056，样本量201；下载大小53040，数据集总大小97056
- 美国外交政策（us_foreign_policy）：测试集字节数42136，样本量100；下载大小22002，数据集总大小42136
- 病毒学（virology）：测试集字节数63046，样本量166；下载大小33137，数据集总大小63046
- 世界宗教（world_religions）：测试集字节数35462，样本量171；下载大小20706，数据集总大小35462

### 数据文件配置
所有细分配置的数据文件均针对测试集拆分进行配置，数据文件路径格式为`{配置名称}/train-*`，即每个配置对应独立子目录下以`train-`为前缀的批量数据文件。