mlabonne/guanaco-llama2-1k
收藏资源简介:
--- dataset_info: features: - name: text dtype: string splits: - name: train num_bytes: 1654448 num_examples: 1000 download_size: 966693 dataset_size: 1654448 configs: - config_name: default data_files: - split: train path: data/train-* --- # Guanaco-1k: Lazy Llama 2 Formatting This is a subset (1000 samples) of the excellent [`timdettmers/openassistant-guanaco`](https://huggingface.co/datasets/timdettmers/openassistant-guanaco) dataset, processed to match Llama 2's prompt format as described [in this article](https://huggingface.co/blog/llama2#how-to-prompt-llama-2). It was created using the following [colab notebook](https://colab.research.google.com/drive/1Ad7a9zMmkxuXTOh1Z7-rNSICA4dybpM2?usp=sharing). Useful if you don't want to reformat it by yourself (e.g., using a script). It was designed for [this article](https://mlabonne.github.io/blog/posts/Fine_Tune_Your_Own_Llama_2_Model_in_a_Colab_Notebook.html) about fine-tuning a Llama 2 (chat) model in a Google Colab.
数据集信息: 特征: - 名称:text 数据类型:字符串 划分集: - 名称:train(训练集) 字节数:1654448 样本数:1000 下载大小:966693 数据集总大小:1654448 配置项: - 配置名称:default(默认) 数据文件: - 划分集:train(训练集) 路径:data/train-* # Guanaco-1k:免手动Llama 2格式适配 本数据集为优质数据集[`timdettmers/openassistant-guanaco`](https://huggingface.co/datasets/timdettmers/openassistant-guanaco)的子集,共包含1000条样本,已按照[本文档](https://huggingface.co/blog/llama2#how-to-prompt-llama-2)中所述的Llama 2提示词格式完成适配处理。本数据集通过以下[Colab笔记本](https://colab.research.google.com/drive/1Ad7a9zMmkxuXTOh1Z7-rNSICA4dybpM2?usp=sharing)生成。 若您不愿手动(或通过脚本)自行完成格式重构,本数据集将十分实用。本数据集专为[这篇关于在Google Colab中微调Llama 2(对话)模型的教程](https://mlabonne.github.io/blog/posts/Fine_Tune_Your_Own_Llama_2_Model_in_a_Colab_Notebook.html)设计。
数据集概述
数据集名称
Guanaco-1k: Lazy Llama 2 Formatting
数据集特征
- text (字符串类型)
数据集分割
- 训练集 (train)
- 示例数量: 1000
- 数据大小: 1654448 字节
数据集大小
- 下载大小: 966693 字节
- 数据集总大小: 1654448 字节
配置信息
- 默认配置 (default)
- 数据文件路径:
data/train-*
- 数据文件路径:




