secmlr/VD-DS-Clean-8k_VD-QWQ-Clean-8k_Qwen2.5-7B-Instruct_full_sft_1e-5_train_dpo

该数据集包含了代码片段及其相关信息，如索引、输出、CWE分类、正确性、编程语言、数据集名称、代码文本、简短和长段修复的正确性、对话信息、修复建议的选择与拒绝情况、预测的最短和最长修复的得分和漏洞类型以及真实得分和漏洞类型。数据集分为训练集，提供了训练集的字节大小和示例数量。

The dataset includes code snippets and their associated information such as index, output, CWE classification, correctness, programming language, dataset name, code text, correctness of short and long fix suggestions, conversational information, chosen and rejected fix suggestions, predicted scores and vulnerability types for the shortest and longest fixes, as well as the true scores and vulnerability types. The dataset is split into a training set, with the byte size and number of examples provided for the training set.