OASST1-pairwise-rlhf-reward

Name: OASST1-pairwise-rlhf-reward
Creator: OpenDataLab
License: 暂无描述

OpenXLab2026-04-18 收录

下载链接：

https://openxlab.org.cn/datasets/OpenDataLab/OASST1-pairwise-rlhf-reward

下载链接

链接失效反馈

官方服务：

资源简介：

OASST1 dataset preprocessed for reward modeling:

提供机构：

OpenDataLab

创建时间：

2024-04-30

tasksource/oasst1_pairwise_rlhf_reward

该数据集是基于OASST1数据集预处理得到的，专门用于奖励建模。数据集中包含了多种语言的对话数据，每个对话包括一个提示（prompt）、一个被选择的回复（chosen）和一个被拒绝的回复（rejected）。数据集分为训练集和验证集，分别包含17966和952个样本。数据集的下载大小为22371458字节，总大小为42888880字节。

Hugging Face2023-07-04 更新610

PJMixers/tasksource_oasst2_pairwise_rlhf_reward-PreferenceShareGPT

--- tags: - preference - preferences size_categories: - 10K<n<100K task_categories: - reinforcement-learning ---

Hugging Face2024-05-30 更新60

Dense Reward for Free in RLHF

The dataset used in the paper is not explicitly described, but it is mentioned that it is a preference dataset for language models.

DataCite Commons2024-12-16 更新60

tasksource/oasst2_pairwise_rlhf_reward

--- configs: - config_name: default data_files: - split: train path: data/train-* - split: validation path: data/validation-* dataset_info: features: - name: lang dtype: string

Hugging Face2024-01-09 更新80

d0rj/oasst1_pairwise_rlhf_reward-ru

--- dataset_info: features: - name: lang dtype: string - name: parent_id dtype: string - name: prompt dtype: string - name: chosen dtype: string - name: rejected dtype:

Hugging Face2024-07-02 更新170