OALL/details_riotu-lab__ArabianGPT-01B_v2

该数据集是在评估模型 riotu-lab/ArabianGPT-01B 的过程中自动创建的。数据集由 116 个配置组成，每个配置对应一个评估任务。数据集是从一次运行中创建的，每次运行都有一个特定的分割，分割的名称使用运行的日期时间。train 分割始终指向最新的结果。还有一个额外的配置 results 存储了运行的所有汇总结果。README 中包括了一个 Python 代码片段，用于从特定的运行加载详细信息。最后，它提供了 2025-10-20T13:33:20.949472 运行的最新结果，并说明了如何访问这些结果。

Dataset automatically created during the evaluation run of model [riotu-lab/ArabianGPT-01B](https://huggingface.co/riotu-lab/ArabianGPT-01B). The dataset is composed of 116 configurations, each corresponding to one of the evaluated tasks. The dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run. The "train" split is always pointing to the latest results. An additional configuration "results" store all the aggregated results of the run. To load the details from a run, you can for instance do the following: python from datasets import load_dataset data = load_dataset("OALL/details_riotu-lab__ArabianGPT-01B_v2", "results", split="train") These are the [latest results from run 2025-10-20T13:33:20.949472](https://huggingface.co/datasets/OALL/details_riotu-lab__ArabianGPT-01B_v2/blob/main/results_2025-10-20T13-33-20.949472.json) (note that there might be results for other tasks in the repos if successive evals didnt cover the same tasks. You find each in the results and the "latest" split for each eval): python { "all": { "acc_norm": 0.2802020262807966, "acc_norm_stderr": 0.03209759259990315 }, "community|alghafa:meta_ar_dialects|0": { "acc_norm": 0.26005560704355885, "acc_norm_stderr": 0.005972789123713404 }, ... }