mbzuai-ugrip-statement-tuning/xstorycloze

---
dataset_info:
- config_name: ar
features:
- name: answer_right_ending
dtype: int32
- name: statement1
dtype: string
- name: statement2
dtype: string
splits:
- name: test
num_bytes: 1172623
num_examples: 1511
download_size: 581963
dataset_size: 1172623
- config_name: en
features:
- name: answer_right_ending
dtype: int32
- name: statement1
dtype: string
- name: statement2
dtype: string
splits:
- name: test
num_bytes: 822860
num_examples: 1511
download_size: 459654
dataset_size: 822860
- config_name: es
features:
- name: answer_right_ending
dtype: int32
- name: statement1
dtype: string
- name: statement2
dtype: string
splits:
- name: test
num_bytes: 894779
num_examples: 1511
download_size: 507534
dataset_size: 894779
- config_name: eu
features:
- name: answer_right_ending
dtype: int32
- name: statement1
dtype: string
- name: statement2
dtype: string
splits:
- name: test
num_bytes: 896518
num_examples: 1511
download_size: 485370
dataset_size: 896518
- config_name: hi
features:
- name: answer_right_ending
dtype: int32
- name: statement1
dtype: string
- name: statement2
dtype: string
splits:
- name: test
num_bytes: 1958974
num_examples: 1511
download_size: 739647
dataset_size: 1958974
- config_name: id
features:
- name: answer_right_ending
dtype: int32
- name: statement1
dtype: string
- name: statement2
dtype: string
splits:
- name: test
num_bytes: 913463
num_examples: 1511
download_size: 476175
dataset_size: 913463
- config_name: my
features:
- name: answer_right_ending
dtype: int32
- name: statement1
dtype: string
- name: statement2
dtype: string
splits:
- name: test
num_bytes: 2735802
num_examples: 1511
download_size: 849229
dataset_size: 2735802
- config_name: ru
features:
- name: answer_right_ending
dtype: int32
- name: statement1
dtype: string
- name: statement2
dtype: string
splits:
- name: test
num_bytes: 1415824
num_examples: 1511
download_size: 688117
dataset_size: 1415824
- config_name: sw
features:
- name: answer_right_ending
dtype: int32
- name: statement1
dtype: string
- name: statement2
dtype: string
splits:
- name: test
num_bytes: 879761
num_examples: 1511
download_size: 468032
dataset_size: 879761
- config_name: te
features:
- name: answer_right_ending
dtype: int32
- name: statement1
dtype: string
- name: statement2
dtype: string
splits:
- name: test
num_bytes: 2037222
num_examples: 1511
download_size: 745049
dataset_size: 2037222
- config_name: zh
features:
- name: answer_right_ending
dtype: int32
- name: statement1
dtype: string
- name: statement2
dtype: string
splits:
- name: test
num_bytes: 806541
num_examples: 1511
download_size: 485889
dataset_size: 806541
configs:
- config_name: ar
data_files:
- split: test
path: ar/test-*
- config_name: en
data_files:
- split: test
path: en/test-*
- config_name: es
data_files:
- split: test
path: es/test-*
- config_name: eu
data_files:
- split: test
path: eu/test-*
- config_name: hi
data_files:
- split: test
path: hi/test-*
- config_name: id
data_files:
- split: test
path: id/test-*
- config_name: my
data_files:
- split: test
path: my/test-*
- config_name: ru
data_files:
- split: test
path: ru/test-*
- config_name: sw
data_files:
- split: test
path: sw/test-*
- config_name: te
data_files:
- split: test
path: te/test-*
- config_name: zh
data_files:
- split: test
path: zh/test-*
---

原始信息汇总

数据集概述

数据集配置

• config_name: 包含多种语言配置，如ar, en, es, eu, hi, id, my, ru, sw, te, zh。
• features: 每种配置包含三个特征：
  • answer_right_ending: 数据类型为int32。
  • statement1: 数据类型为string。
  • statement2: 数据类型为string。

数据集分割

• split: 所有配置均包含名为test的分割。
• num_examples: 每个test分割包含1511个样本。
• num_bytes: 不同语言配置的test分割大小不同，具体如下：
  • ar: 1172623 bytes
  • en: 822860 bytes
  • es: 894779 bytes
  • eu: 896518 bytes
  • hi: 1958974 bytes
  • id: 913463 bytes
  • my: 2735802 bytes
  • ru: 1415824 bytes
  • sw: 879761 bytes
  • te: 2037222 bytes
  • zh: 806541 bytes

数据集大小与下载大小

• download_size: 不同语言配置的下载大小不同，具体如下：
  • ar: 581963 bytes
  • en: 459654 bytes
  • es: 507534 bytes
  • eu: 485370 bytes
  • hi: 739647 bytes
  • id: 476175 bytes
  • my: 849229 bytes
  • ru: 688117 bytes
  • sw: 468032 bytes
  • te: 745049 bytes
  • zh: 485889 bytes
• dataset_size: 与num_bytes相同，反映数据集的实际大小。

数据文件路径

• path: 每个配置的test分割数据文件路径格式为[语言代码]/test-*。