Deepexi/glaive-function-calling-vicuna

---
license: cc-by-4.0
language:
- en
size_categories:
- 10K<n<100K
---

**数据集格式说明： [[glaiveai/glaive-function-calling · Datasets at Hugging Face](https://huggingface.co/datasets/glaiveai/glaive-function-calling)](glaiveai/glaive-function-calling) 的 SFT 格式**

我们高兴地宣布，数据集 "glaiveai/glaive-function-calling" 已经根据 SFT（Supervised Fine-Tuning）的需求进行了格式转换，以支持大型语言模型的训练。以下是有关这一新格式的简要说明：

1. **数据集概述:**

数据集 "glaiveai/glaive-function-calling" 基于 CC-BY-4.0 协议发布，原始数据集包含标识符和对话信息，这些数据已被转换为适应 SFT 训练的结构。

2. **数据格式：**
转换后的数据集格式包含以下关键信息：

- `id`: 整数类型的标识符，用于唯一标识每个数据样本。
- `conversations`: 一个数组，其中包含对话信息。每个对话可以由多个句子组成，以更好地呈现函数调用的上下文。

3. **数据集用途：**
转换后的数据集适用于 SFT 的训练，主要用途包括但不限于：

- 函数调用理解： 通过分析对话中的函数调用信息，让语言模型更好地理解函数之间的关系，从而提高其代码理解能力。
- 上下文感知性： 对话信息能够为模型提供更丰富的上下文，使其更准确地推断和生成代码片段。
- 代码生成与推荐： 基于对话中的函数调用上下文，模型可以更精确地生成代码，并提供更合适的函数建议。

通过将数据集 "glaiveai/glaive-function-calling" 转换为 SFT 格式，我们旨在为大型语言模型的训练提供更适合sft的函数调用数据，以提升其代码理解和生成的性能。

如有任何问题或需要进一步帮助，请随时联系我们。感谢您对函数调用数据集及其应用的兴趣与支持！

---
license: cc-by-4.0
language:
- en
size_categories:
- 10K<n<100K
---

**Dataset Format Specification: SFT format of [glaiveai/glaive-function-calling · Datasets at Hugging Face](https://huggingface.co/datasets/glaiveai/glaive-function-calling) (glaiveai/glaive-function-calling)**

We are delighted to announce that the dataset "glaiveai/glaive-function-calling" has been formatted for Supervised Fine-Tuning (SFT) to support the training of large language models. Below is a brief introduction to this new format:

1. **Dataset Overview**
The dataset "glaiveai/glaive-function-calling" is released under the CC-BY-4.0 license. The original dataset contains identifiers and dialogue information, which has been converted into a structure optimized for SFT training.

2. **Data Format**
The converted dataset includes the following key fields:
- `id`: An integer-type unique identifier for each data sample.
- `conversations`: An array containing dialogue information. Each conversation may consist of multiple sentences to better present the context of function calls.

3. **Dataset Use Cases**
The converted dataset is suitable for SFT training, with main applications including but not limited to:
- Function Call Understanding: By analyzing function call information in dialogues, enable language models to better understand the relationships between functions, thereby improving their code comprehension capabilities.
- Context Awareness: Dialogue information provides richer context for the model, allowing it to more accurately infer and generate code snippets.
- Code Generation and Recommendation: Based on the function call context in dialogues, the model can generate code more precisely and provide more appropriate function suggestions.

By converting the dataset "glaiveai/glaive-function-calling" into SFT format, we aim to provide function call data that is more suitable for SFT training of large language models, to enhance their code comprehension and generation performance.

If you have any questions or need further assistance, please feel free to contact us. Thank you for your interest and support in the function call dataset and its applications!

原始信息汇总

数据集概述

数据集名称

• glaiveai/glaive-function-calling

数据集格式

• SFT 格式

数据集存储位置

• Datasets at HF Mirror