LanguageShades/FormattedBiasShadesTest
收藏Hugging Face2024-06-13 更新2024-06-29 收录
下载链接：
https://hf-mirror.com/datasets/LanguageShades/FormattedBiasShadesTest
下载链接
链接失效反馈官方服务：
资源简介：
---
dataset_info:
features:
- name: index
dtype: float64
- name: Subset
dtype: string
- name: bias_type
sequence: string
- name: stereotype_origin_langs
sequence: string
- name: stereotype_valid_langs
sequence: string
- name: stereotype_valid_regions
sequence: string
- name: stereotyped_group
dtype: string
- name: en_template
dtype: string
- name: en_biased_sentence
dtype: string
- name: 'English: Is this a saying?'
dtype: string
- name: 'English: Comments'
dtype: string
- name: fr_template
dtype: string
- name: fr_biased_sentence
dtype: string
- name: fr_expression
dtype: float64
- name: 'French: Comments'
dtype: string
- name: ro_template
dtype: string
- name: ro_biased_sentence
dtype: string
- name: 'ro_expression '
dtype: float64
- name: 'Romanian: Comments'
dtype: string
- name: ar_template
dtype: float64
- name: ar_biased_sentence
dtype: string
- name: 'Arabic: Comments'
dtype: string
- name: 'Arabic: Is this a saying?'
dtype: string
- name: bn_template
dtype: float64
- name: bn_biased_sentence
dtype: string
- name: 'Bengali: Comments'
dtype: float64
- name: 'Bengali: Is this a saying?'
dtype: string
- name: zh_template
dtype: float64
- name: zh_biased_sentence
dtype: string
- name: zh_expression
dtype: string
- name: 'Chinese: Comments'
dtype: string
- name: 'Traditional Chinese: Templates'
dtype: float64
- name: zh_hant_biased_sentence
dtype: string
- name: zh_hk_expression
dtype: string
- name: 'Traditional Chinese: Comments'
dtype: string
- name: nl_template
dtype: string
- name: nl_biased_sentence
dtype: string
- name: nl_expression
dtype: string
- name: 'Dutch: Comments'
dtype: string
- name: hi_template
dtype: string
- name: hi_biased_sentence
dtype: string
- name: 'Hindi: Is this a saying?'
dtype: float64
- name: 'Hindi: Comments'
dtype: string
- name: mr_template
dtype: string
- name: mr_biased_sentence
dtype: string
- name: 'Marathi: Is this a saying?'
dtype: float64
- name: 'Marathi: Comments'
dtype: float64
- name: ru_template
dtype: string
- name: ru_biased_sentence
dtype: string
- name: 'Russian: Comments'
dtype: string
- name: ru_expression
dtype: string
- name: de_template
dtype: string
- name: de_biased_sentence
dtype: string
- name: 'German: Comments'
dtype: string
- name: de_expression
dtype: string
- name: it_template
dtype: string
- name: it_biased_sentence
dtype: string
- name: 'Italian: Is this a saying?'
dtype: float64
- name: 'Italian: Comments'
dtype: float64
- name: pl_template
dtype: string
- name: pl_biased_sentence
dtype: string
- name: 'Polish: Comments'
dtype: string
- name: pl_expression
dtype: string
- name: pt_br_template
dtype: string
- name: pt_br_biased_sentence
dtype: string
- name: 'Brazilian Portuguese: Comments'
dtype: string
- name: pt_br_expression
dtype: string
- name: 'Spanish: Templates'
dtype: string
- name: es_biased_sentence
dtype: string
- name: 'Spanish: Comments'
dtype: string
- name: es_expression
dtype: float64
- name: logprob_meta-llama_Meta-Llama-3-8B
sequence: float64
- name: tokens_meta-llama_Meta-Llama-3-8B
sequence: string
- name: template_meta-llama_Meta-Llama-3-8B
sequence: string
- name: English_logprob_meta-llama_Meta-Llama-3-8B
sequence: float64
- name: English_tokens_meta-llama_Meta-Llama-3-8B
sequence: string
- name: English_template_meta-llama_Meta-Llama-3-8B
sequence: string
- name: French_logprob_meta-llama_Meta-Llama-3-8B
sequence: float64
- name: French_tokens_meta-llama_Meta-Llama-3-8B
sequence: string
- name: French_template_meta-llama_Meta-Llama-3-8B
sequence: string
- name: English_logprob_bigscience_bloom-7b1
sequence: float64
- name: English_tokens_bigscience_bloom-7b1
sequence: string
- name: English_template_bigscience_bloom-7b1
sequence: string
- name: French_logprob_bigscience_bloom-7b1
sequence: float64
- name: French_tokens_bigscience_bloom-7b1
sequence: string
- name: French_template_bigscience_bloom-7b1
sequence: string
splits:
- name: test
num_bytes: 1739738
num_examples: 668
download_size: 660248
dataset_size: 1739738
configs:
- config_name: default
data_files:
- split: test
path: data/test-*
---
提供机构：
LanguageShades
原始信息汇总
数据集概述
数据集特征
- index: 数据类型为
float64 - Subset: 数据类型为
string - bias_type: 数据类型为
string，序列类型 - stereotype_origin_langs: 数据类型为
string，序列类型 - stereotype_valid_langs: 数据类型为
string，序列类型 - stereotype_valid_regions: 数据类型为
string，序列类型 - stereotyped_group: 数据类型为
string - en_template: 数据类型为
string - en_biased_sentence: 数据类型为
string - English: Is this a saying?: 数据类型为
string - English: Comments: 数据类型为
string - fr_template: 数据类型为
string - fr_biased_sentence: 数据类型为
string - fr_expression: 数据类型为
float64 - French: Comments: 数据类型为
string - ro_template: 数据类型为
string - ro_biased_sentence: 数据类型为
string - ro_expression: 数据类型为
float64 - Romanian: Comments: 数据类型为
string - ar_template: 数据类型为
float64 - ar_biased_sentence: 数据类型为
string - Arabic: Comments: 数据类型为
string - Arabic: Is this a saying?: 数据类型为
string - bn_template: 数据类型为
float64 - bn_biased_sentence: 数据类型为
string - Bengali: Comments: 数据类型为
float64 - Bengali: Is this a saying?: 数据类型为
string - zh_template: 数据类型为
float64 - zh_biased_sentence: 数据类型为
string - zh_expression: 数据类型为
string - Chinese: Comments: 数据类型为
string - Traditional Chinese: Templates: 数据类型为
float64 - zh_hant_biased_sentence: 数据类型为
string - zh_hk_expression: 数据类型为
string - Traditional Chinese: Comments: 数据类型为
string - nl_template: 数据类型为
string - nl_biased_sentence: 数据类型为
string - nl_expression: 数据类型为
string - Dutch: Comments: 数据类型为
string - hi_template: 数据类型为
string - hi_biased_sentence: 数据类型为
string - Hindi: Is this a saying?: 数据类型为
float64 - Hindi: Comments: 数据类型为
string - mr_template: 数据类型为
string - mr_biased_sentence: 数据类型为
string - Marathi: Is this a saying?: 数据类型为
float64 - Marathi: Comments: 数据类型为
float64 - ru_template: 数据类型为
string - ru_biased_sentence: 数据类型为
string - Russian: Comments: 数据类型为
string - ru_expression: 数据类型为
string - de_template: 数据类型为
string - de_biased_sentence: 数据类型为
string - German: Comments: 数据类型为
string - de_expression: 数据类型为
string - it_template: 数据类型为
string - it_biased_sentence: 数据类型为
string - Italian: Is this a saying?: 数据类型为
float64 - Italian: Comments: 数据类型为
float64 - pl_template: 数据类型为
string - pl_biased_sentence: 数据类型为
string - Polish: Comments: 数据类型为
string - pl_expression: 数据类型为
string - pt_br_template: 数据类型为
string - pt_br_biased_sentence: 数据类型为
string - Brazilian Portuguese: Comments: 数据类型为
string - pt_br_expression: 数据类型为
string - Spanish: Templates: 数据类型为
string - es_biased_sentence: 数据类型为
string - Spanish: Comments: 数据类型为
string - es_expression: 数据类型为
float64 - logprob_meta-llama_Meta-Llama-3-8B: 数据类型为
float64，序列类型 - tokens_meta-llama_Meta-Llama-3-8B: 数据类型为
string，序列类型 - template_meta-llama_Meta-Llama-3-8B: 数据类型为
string，序列类型 - English_logprob_meta-llama_Meta-Llama-3-8B: 数据类型为
float64，序列类型 - English_tokens_meta-llama_Meta-Llama-3-8B: 数据类型为
string，序列类型 - English_template_meta-llama_Meta-Llama-3-8B: 数据类型为
string，序列类型 - French_logprob_meta-llama_Meta-Llama-3-8B: 数据类型为
float64，序列类型 - French_tokens_meta-llama_Meta-Llama-3-8B: 数据类型为
string，序列类型 - French_template_meta-llama_Meta-Llama-3-8B: 数据类型为
string，序列类型 - English_logprob_bigscience_bloom-7b1: 数据类型为
float64，序列类型 - English_tokens_bigscience_bloom-7b1: 数据类型为
string，序列类型 - English_template_bigscience_bloom-7b1: 数据类型为
string，序列类型 - French_logprob_bigscience_bloom-7b1: 数据类型为
float64，序列类型 - French_tokens_bigscience_bloom-7b1: 数据类型为
string，序列类型 - French_template_bigscience_bloom-7b1: 数据类型为
string，序列类型
数据集分割
- test: 包含 668 个样本，总字节数为 1739738
数据集大小
- 下载大小: 660248 字节
- 数据集大小: 1739738 字节
配置
- config_name: default
- data_files:
- split: test
- path: data/test-*
- data_files:
