Nurture-intelligence/thinking_dataset_v1

这个数据集包含了用于制作思考模型的基础质问数据。数据是通过在ollama上运行Qwen/Qwen2.5-32B-Instruct模型的q8_0/GGUF版本产生的。虽然对问题部分进行了清洗，但没有对答案进行清洗。数据集的答案是使用另一个名为mistral large的模型生成的，因此只有问题部分遵循Apache 2.0协议。This dataset consists of question data used as the basis for creating thinking models. The data was generated by running the Qwen/Qwen2.5-32B-Instruct models q8_0/GGUF version on ollama. Although there has been some cleaning on the question part, there is no cleaning on the answers. The answers in the dataset were generated using another model called mistral large, so only the question part follows the Apache 2.0 protocol.