OctoMed/GLM-Multiturn-CoT
收藏Hugging Face2026-04-20 更新2026-04-26 收录
下载链接:
https://hf-mirror.com/datasets/OctoMed/GLM-Multiturn-CoT
下载链接
链接失效反馈官方服务:
资源简介:
---
license: apache-2.0
task_categories:
- question-answering
- text-generation
language:
- en
tags:
- reasoning
- chain-of-thought
- multi-turn
---
# OctoMed/GLM-Multiturn-CoT
Multi-turn chain-of-thought conversations converted to OctoMed format for SFT training.
## Source
Derived from [Jackrong/glm-4.7-multiturn-CoT](https://huggingface.co/datasets/Jackrong/glm-4.7-multiturn-CoT)
by [Jackrong](https://huggingface.co/Jackrong). All credit for the original data collection and
distillation goes to the original authors.
## Format
Each example contains:
- `conversations`: list of `{from, value}` turns (human / gpt), with `<think>` reasoning blocks in gpt turns
- `responses`: the final gpt turn repeated for compatibility with the OctoMed pipeline
- `question`, `answer`: empty strings (full context is in `conversations`)
许可证:Apache-2.0
任务类别:
- 问答(Question Answering)
- 文本生成(Text Generation)
语言:英语
标签:
- 推理(Reasoning)
- 思维链(Chain-of-Thought,CoT)
- 多轮对话(Multi-turn)
# OctoMed/GLM-Multiturn-CoT 数据集
本数据集将多轮思维链对话转换为OctoMed格式,以用于监督微调(Supervised Fine-Tuning,SFT)训练。
## 来源
本数据集衍生自数据集[Jackrong/glm-4.7-multiturn-CoT](https://huggingface.co/datasets/Jackrong/glm-4.7-multiturn-CoT),原作者为[Jackrong](https://huggingface.co/Jackrong)。所有原始数据采集与蒸馏工作的权属与荣誉均归属于原作者。
## 数据格式
每个数据样本包含以下字段:
- `conversations`:由若干条采用`{from, value}`格式的对话轮次构成的列表,其中GPT生成的轮次内包含以`<think>`标记包裹的思维链推理模块。
- `responses`:为兼容OctoMed训练流水线,重复了最终的GPT生成轮次内容。
- `question`与`answer`:均为空字符串(完整上下文信息均存储于`conversations`字段中)。
提供机构:
OctoMed


