zed-industries/zeta

这是一个用于训练Zeta模型的开放数据集。Zeta模型是基于Qwen2.5-Coder-7B模型的编辑预测模型，能够根据开发者的近期编程模式和光标位置预测其下一次代码编辑，通过简单的Tab键按下实现智能代码补全。数据集分为三部分：训练数据（train.jsonl）、直接偏好优化数据（dpo.jsonl）和评估数据（eval.jsonl）。

This is the open dataset used to train Zeta, an edit prediction model derived from Qwen2.5-Coder-7B, which predicts the developers next code edit based on their recent programming patterns and cursor position, allowing for intelligent code completion with a simple tab press. The dataset is split into three parts: training data (train.jsonl), direct preference optimization data (dpo.jsonl), and evaluation data (eval.jsonl).