LocalResearchGroup/split-glaive-code-assistant-v3

该数据集包含四种不同规模的问题和答案对，分别为100k、10k、1k和完整版(full)。每种规模的数据集都包括训练集和测试集，用于训练和评估模型在问答任务上的表现。数据集中的字段包括问题(question)、答案(answer)、提示(prompt)和响应(response)，均为字符串类型。

The dataset consists of four different scales of question-answer pairs, namely 100k, 10k, 1k, and the full version. Each scale of the dataset includes a training set and a test set for training and evaluating the models performance on question answering tasks. The fields in the dataset include question, answer, prompt, and response, all of which are of string type.