five

Ring-lite-distill-preview-dpo-data

收藏
魔搭社区2026-01-05 更新2025-04-19 收录
下载链接:
https://modelscope.cn/datasets/inclusionAI/Ring-lite-distill-preview-dpo-data
下载链接
链接失效反馈
官方服务:
资源简介:
<p align="center"> <img src="https://modelscope.cn/api/v1/models/inclusionAI/Ling-lite-base/repo?Revision=master&FilePath=ant-bailing.png&View=true" width="100"/> <p> <p align="center"> 🤖 <a href="https://modelscope.cn/organization/inclusionAI">ModelScope</a> 🤗 <a href="https://huggingface.co/inclusionAI">HuggingFace</a> 🖥️ <a href="https://github.com/inclusionAI/Ring">GitHub</a> <p> # Ring-lite-distill-preview The Ring-lite-distill-preview Dataset comprises the following components: - [Ring-lite-distill-preview-sft-data](https://modelscope.cn/datasets/inclusionAI/Ring-lite-distill-preview-sft-data): A subset of SFT data used for training [Ring-lite-distill-preview](https://modelscope.cn/models/inclusionAI/Ring-lite-distill-preview). - [Ring-lite-distill-preview-dpo-data](https://modelscope.cn/datasets/inclusionAI/Ring-lite-distill-preview-dpo-data): A subset of DPO data used for training [Ring-lite-distill-preview](https://modelscope.cn/models/inclusionAI/Ring-lite-distill-preview). ## Ring-lite-distill-preview-dpo-data This is a subset of DPO data used to train the [Ring-lite-distill-preview](https://modelscope.cn/models/inclusionAI/Ring-lite-distill-preview) model, featuring approximately 4K high-quality English and Chinese samples focused on complex reasoning tasks and instruction following. More details will be reported in our technical report [TBD]

<p align="center"> <img src="https://modelscope.cn/api/v1/models/inclusionAI/Ling-lite-base/repo?Revision=master&FilePath=ant-bailing.png&View=true" width="100"/> <p> <p align="center"> 🤖 <a href="https://modelscope.cn/organization/inclusionAI">ModelScope</a> 🤗 <a href="https://huggingface.co/inclusionAI">HuggingFace</a> 🖥️ <a href="https://github.com/inclusionAI/Ring">GitHub</a> <p> # Ring-lite-distill-preview Ring-lite-distill-preview 数据集包含以下组成部分: - [Ring-lite-distill-preview-sft-data](https://modelscope.cn/datasets/inclusionAI/Ring-lite-distill-preview-sft-data): 用于训练[Ring-lite-distill-preview](https://modelscope.cn/models/inclusionAI/Ring-lite-distill-preview)模型的监督微调(Supervised Fine-Tuning,SFT)数据子集。 - [Ring-lite-distill-preview-dpo-data](https://modelscope.cn/datasets/inclusionAI/Ring-lite-distill-preview-dpo-data): 用于训练[Ring-lite-distill-preview](https://modelscope.cn/models/inclusionAI/Ring-lite-distill-preview)模型的直接偏好优化(Direct Preference Optimization,DPO)数据子集。 ## Ring-lite-distill-preview-dpo-data 本数据集为用于训练[Ring-lite-distill-preview](https://modelscope.cn/models/inclusionAI/Ring-lite-distill-preview)模型的直接偏好优化(DPO)数据子集,包含约4000条高质量中英双语样本,聚焦复杂推理任务与指令遵循场景。 更多细节将在我们的技术报告[TBD]中公布。
提供机构:
maas
创建时间:
2025-04-13
5,000+
优质数据集
54 个
任务类型
进入经典数据集
二维码
社区交流群

面向社区/商业的数据集话题

二维码
科研交流群

面向高校/科研机构的开源数据集话题

数据驱动未来

携手共赢发展

商业合作