EeeLM/llm-jp-4-thinking-sft-data-chatml

这是一个转换后的数据集，基于llm-jp的llm-jp-4-thinking-sft-data，已转换为ChatML格式。数据集包含多个配置（如daring_anteater、flan、jaster_v1.4.1等），每个配置提供不同难度级别（低、中、高）的推理数据文件，涉及日语和英语语言。数据集用于推理任务的监督微调（SFT），但具体内容描述未在README中明确提及。许可证遵循原始数据源，未附加新许可证。This is a converted dataset based on llm-jps llm-jp-4-thinking-sft-data, transformed into ChatML format. It includes multiple configurations (e.g., daring_anteater, flan, jaster_v1.4.1, etc.), each providing reasoning data files at different difficulty levels (low, medium, high) in Japanese and English. The dataset is intended for supervised fine-tuning (SFT) of reasoning tasks, but detailed content descriptions are not explicitly provided in the README. The license follows the original data sources without additional licensing.