argilla/magpie-ultra-v0.1

Magpie Ultra v0.1是一个合成的监督微调数据集，基于Llama 3.1 405B-Instruct模型创建。它包含挑战性的指令和响应，适用于多种任务，如编程调试、数学、数据分析、创意写作等。数据集通过Magpie配方生成，包含50000个示例，数据大小为837917458字节。

Magpie Ultra v0.1 is a synthetic supervised fine-tuning dataset created based on the Llama 3.1 405B-Instruct model. It contains challenging instructions and responses suitable for various tasks such as programming debugging, mathematics, data analysis, creative writing, etc. The dataset was generated using the Magpie recipe and includes 50,000 examples with a total size of 837,917,458 bytes.