answerdotai/MLMMLU
收藏Hugging Face2024-07-09 更新2024-07-22 收录
下载链接:
https://hf-mirror.com/datasets/answerdotai/MLMMLU
下载链接
链接失效反馈官方服务:
资源简介:
该数据集包含四个不同的配置:Amateur、Reserve、Rookie和Semipro。每个配置包含多个特征,如问题ID、问题文本、选项、答案、答案索引、推理内容、类别、来源、LLM预测结果等。数据集分为训练集和测试集,每个配置的训练集和测试集的大小和样本数量也有所不同。
The dataset contains four different configurations: Amateur, Reserve, Rookie, and Semipro. Each configuration includes multiple features such as question ID, question text, options, answer, answer index, reasoning content, category, source, LLM prediction results, etc. The dataset is divided into training and test sets, with varying sizes and numbers of samples for each configuration.
提供机构:
answerdotai
原始信息汇总
数据集概述
数据集配置
Amateur
- 特征:
question_id:int64question:stringoptions:sequenceofstringanswer:stringanswer_index:int64cot_content:stringcategory:stringsrc:stringllama_pred:stringllama_correct:bool
- 分割:
train:- 字节数: 4423321
- 样本数: 6017
test:- 字节数: 1712259
- 样本数: 2415
- 下载大小: 3016041
- 数据集大小: 6135580
Reserve
- 特征:
question:stringcategory:stringchoices:sequenceofstringanswer:int64id_in_subset:int64question_id:stringllama_correct:bool
- 分割:
train:- 字节数: 3718345
- 样本数: 7022
test:- 字节数: 3713635
- 样本数: 7020
- 下载大小: 3722290
- 数据集大小: 7431980
Rookie
- 特征:
question:stringcategory:stringchoices:sequenceofstringanswer:int64id_in_subset:int64question_id:stringllama_correct:bool
- 分割:
train:- 字节数: 3718345
- 样本数: 7022
test:- 字节数: 2369448
- 样本数: 4623
- 下载大小: 3072212
- 数据集大小: 6087793
Semipro
- 特征:
question_id:int64question:stringoptions:sequenceofstringanswer:stringanswer_index:int64cot_content:stringcategory:stringsrc:stringllama_pred:stringllama_correct:bool
- 分割:
train:- 字节数: 4423321
- 样本数: 6017
test:- 字节数: 4360084
- 样本数: 6015
- 下载大小: 4287701
- 数据集大小: 8783405
数据文件路径
Amateur
train:Amateur/train-*test:Amateur/test-*
Reserve
train:Reserve/train-*test:Reserve/test-*
Rookie
train:Rookie/train-*test:Rookie/test-*
Semipro
train:Semipro/train-*test:Semipro/test-*



