ljvmiranda921/details_msde-allenai_Olmo-3-1025-7B-lora-4bit-msde-S1-id_Llama-3_1-70B-Instruct

该数据集是在模型ljvmiranda921/msde-sft-dev的评估运行过程中自动创建的。数据集包含14个配置，每个配置对应一个评估任务。数据集由14次运行的数据组成，每次运行作为每个配置中的一个特定分割存储，分割名称使用运行的时间戳。"train"分割始终指向最新的结果。另外，一个名为"results"的配置存储了所有运行的聚合结果。README还提供了加载数据集的Python代码示例，并展示了最新评估结果的JSON格式。

Dataset automatically created during the evaluation run of model ljvmiranda921/msde-sft-dev. The dataset is composed of 14 configurations, each corresponding to one of the evaluated tasks. The dataset has been created from 14 runs, with each run stored as a specific split in each configuration, named using the timestamp of the run. The "train" split always points to the latest results. An additional "results" configuration stores all the aggregated results of the runs. The README includes a Python code snippet for loading the dataset and displays the latest evaluation results in JSON format.