OALL/details_AXCXEPT__EZO-Qwen2.5-72B-Instruct

该数据集是在评估模型 [AXCXEPT/EZO-Qwen2.5-72B-Instruct](https://huggingface.co/AXCXEPT/EZO-Qwen2.5-72B-Instruct) 时自动创建的。数据集由 136 个配置组成，每个配置对应一个评估任务。数据集的结构包括根据运行时间戳命名的分割，train 分割总是指向最新的结果。还有一个额外的 results 配置，用于存储所有运行结果的汇总。README 文件还包括一个 Python 代码片段，演示如何加载数据集的详细信息。最后，README 提供了特定运行的最新结果的摘要。

Dataset automatically created during the evaluation run of model [AXCXEPT/EZO-Qwen2.5-72B-Instruct](https://huggingface.co/AXCXEPT/EZO-Qwen2.5-72B-Instruct). The dataset is composed of 136 configurations, each one corresponding to one of the evaluated tasks. The dataset has been structured with splits named by run timestamps, with the train split always pointing to the latest results. An additional results configuration stores all the aggregated results of the run. The README also includes a Python code snippet demonstrating how to load dataset details. Lastly, the README provides a summary of the latest results from a specific run.