lightblue/multilingual_prompts_100_sample_quantified_test
收藏Hugging Face2024-05-17 更新2024-06-12 收录
下载链接:
https://hf-mirror.com/datasets/lightblue/multilingual_prompts_100_sample_quantified_test
下载链接
链接失效反馈官方服务:
资源简介:
该数据集包含多个对话的特征,每个对话具有唯一的对话ID、使用的模型、对话内容、轮次、语言信息、OpenAI的审核结果(包括多个类别和分数)、语言检测结果、首次交换的令牌长度、索引级别、多个LLM模型的响应内容及其完成原因、模型评估结果以及量化响应。数据集分为训练集,包含250个样本,总大小为4908551字节,下载大小为2482060字节。
This dataset meticulously records various aspects of multi-turn conversations, including the unique identifier of the conversation, the model used, the content of the conversation and its roles, the turn of the conversation, the language used, the moderation results from OpenAI (including moderation for multiple categories and their scores), whether the conversation has been redacted, the language detection result, the length of the first exchange, the index level, LLM responses from different models, model evaluation information, and quantified responses. The training portion of the dataset contains 250 samples with a total size of 4908551 bytes.
提供机构:
lightblue
原始信息汇总
数据集概述
数据集特征
-
conversation_id
- 数据类型:string
-
model
- 数据类型:string
-
conversation
- 子特征:
- content
- 数据类型:string
- role
- 数据类型:string
- content
- 子特征:
-
turn
- 数据类型:int64
-
language
- 数据类型:string
-
openai_moderation
- 子特征:
- categories
- 子特征:
- harassment
- 数据类型:bool
- harassment/threatening
- 数据类型:bool
- hate
- 数据类型:bool
- hate/threatening
- 数据类型:bool
- self-harm
- 数据类型:bool
- self-harm/instructions
- 数据类型:bool
- self-harm/intent
- 数据类型:bool
- sexual
- 数据类型:bool
- sexual/minors
- 数据类型:bool
- violence
- 数据类型:bool
- violence/graphic
- 数据类型:bool
- harassment
- 子特征:
- category_scores
- 子特征:
- harassment
- 数据类型:float64
- harassment/threatening
- 数据类型:float64
- hate
- 数据类型:float64
- hate/threatening
- 数据类型:float64
- self-harm
- 数据类型:float64
- self-harm/instructions
- 数据类型:float64
- self-harm/intent
- 数据类型:float64
- sexual
- 数据类型:float64
- sexual/minors
- 数据类型:float64
- violence
- 数据类型:float64
- violence/graphic
- 数据类型:float64
- harassment
- 子特征:
- flagged
- 数据类型:bool
- categories
- 子特征:
-
redacted
- 数据类型:bool
-
lang_detect_result
- 子特征:
- lang
- 数据类型:string
- score
- 数据类型:float64
- lang
- 子特征:
-
first_exchange_len_tokens
- 数据类型:int64
-
index_level_1
- 数据类型:int64
-
llm_responses
- 子特征:
- CohereForAI/c4ai-command-r-plus
- 子特征:
- content
- 数据类型:string
- finish_reason
- 数据类型:string
- content
- 子特征:
- CohereForAI/c4ai-command-r-v01
- 子特征:
- content
- 数据类型:string
- finish_reason
- 数据类型:string
- content
- 子特征:
- Nexusflow/Starling-LM-7B-beta
- 子特征:
- content
- 数据类型:string
- finish_reason
- 数据类型:string
- content
- 子特征:
- Qwen/Qwen1.5-32B-Chat
- 子特征:
- content
- 数据类型:string
- finish_reason
- 数据类型:string
- content
- 子特征:
- Qwen/Qwen1.5-72B-Chat
- 子特征:
- content
- 数据类型:string
- finish_reason
- 数据类型:string
- content
- 子特征:
- gpt-4-0125-preview
- 子特征:
- content
- 数据类型:string
- finish_reason
- 数据类型:string
- content
- 子特征:
- peter-gpt-35-turbo
- 子特征:
- content
- 数据类型:string
- finish_reason
- 数据类型:string
- content
- 子特征:
- CohereForAI/c4ai-command-r-plus
- 子特征:
-
model_evals
- 子特征:
- eval_finish_reason
- 数据类型:string
- eval_model_name
- 数据类型:string
- eval_response
- 数据类型:string
- model_id_list
- 数据类型:sequence of string
- eval_finish_reason
- 子特征:
-
responses_quantified
- 数据类型:sequence of string
数据集分割
- train
- 字节数:4908551
- 示例数:250
数据集大小
- 下载大小:2482060
- 数据集大小:4908551
配置
- config_name: default
- data_files
- split: train
- path: data/train-*
- split: train
- data_files



