neoneye/simon-arc-combine-v183

Simons ARC数据集是一个结合了多个数据集的版本，主要用于图像到文本和文本到图像的任务。数据集包含多个JSONL文件，每个文件代表不同的任务或数据类型，如颜色、旋转、平移、细胞自动机、形状、图像等。数据集的版本更新涉及添加新数据集、调整数据集状态（启用或禁用）、更改随机种子、调整图像大小和上下文长度等。

The Simons ARC dataset is a combination of multiple datasets, primarily used for image-to-text and text-to-image tasks. The dataset includes several JSONL files, each representing different tasks or data types such as color, rotation, translation, cellular automata, shapes, images, etc. The versions of the dataset involve adding new datasets, adjusting the status of datasets (enabled or disabled), changing random seeds, adjusting image sizes, and context lengths.