Dogacel/nemotron-post-training-v2-qwen-3.5-0.8b-regen
收藏Hugging Face2026-04-29 更新2026-05-03 收录
下载链接：
https://hf-mirror.com/datasets/Dogacel/nemotron-post-training-v2-qwen-3.5-0.8b-regen
下载链接
链接失效反馈官方服务：
资源简介：
该数据集是通过使用Qwen3.5 0.8B模型重新生成nvidia/Nemotron-Post-Training-Dataset-v2数据集的响应而创建的。数据集仅包含Nemotron Post Training Dataset中的英文样本，其中85%的聊天提示启用了推理功能，其他类别则禁用了推理功能。数据集创建过程中使用了sglang和SpecForge工具，并在8xH200集群上运行了10小时。数据集的来源是Nemotron-Post-Training-Dataset-v2，保留了原始许可证。
Regenerated responses from nvidia/Nemotron-Post-Training-Dataset-v2 dataset using Qwen3.5 0.8B model. Dataset consists only the english samples from the Nemotron Post Training Dataset. 85% of the chat prompts have reasoning enabled, every other category has reasoning disabled. Dataset creation used sglang and SpecForge tools, and ran on an 8xH200 cluster for 10 hours. The source of the dataset is Nemotron-Post-Training-Dataset-v2, retaining the original licenses.
提供机构：
Dogacel
