aidando73/llama3-recipe-3.2-evals-mmlu
收藏Hugging Face2024-11-23 更新2024-12-14 收录
下载链接:
https://hf-mirror.com/datasets/aidando73/llama3-recipe-3.2-evals-mmlu
下载链接
链接失效反馈官方服务:
资源简介:
该数据集包含用于重新计算MMLU(大规模多任务语言理解)分数的评估数据。这些数据是通过加载3.2 Eval Collection中的数据并再次通过MMLU处理来生成的。数据集包含两个配置,分别对应不同模型版本的评估数据。
This dataset contains evaluation data for recomputing MMLU scores, involving loading evaluation data from the 3.2 Eval Collection and passing it through MMLU again to recompute the scores. The dataset includes two configurations, corresponding to the evaluation data for 1B and 3B model sizes.
提供机构:
aidando73



