distilabel-internal-testing/first-distiset-with-routed-batches

---
size_categories: n<1K
dataset_info:
features:
- name: prompt
dtype: string
- name: completion
dtype: string
- name: meta
struct:
- name: category
dtype: string
- name: completion
dtype: string
- name: id
dtype: int64
- name: input
dtype: string
- name: motivation_app
dtype: string
- name: prompt
dtype: string
- name: source
dtype: string
- name: subcategory
dtype: string
- name: merged_generation
sequence: string
splits:
- name: train
num_bytes: 458933
num_examples: 327
download_size: 292464
dataset_size: 458933
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
tags:
- synthetic
- distilabel
- rlaif
---

<p align="left">
<a href="https://github.com/argilla-io/distilabel">
<img src="https://raw.githubusercontent.com/argilla-io/distilabel/main/docs/assets/distilabel-badge-light.png" alt="Built with Distilabel" width="200" height="32"/>
</a>
</p>

# Dataset Card for first-distiset-with-routed-batches

This dataset has been created with [distilabel](https://distilabel.argilla.io/).

## Dataset Summary

This dataset contains a `pipeline.yaml` which can be used to reproduce the pipeline that generated it in distilabel using the `distilabel` CLI:

```console
distilabel pipeline run --config "https://huggingface.co/datasets/distilabel-internal-testing/first-distiset-with-routed-batches/raw/main/pipeline.yaml"
```

or explore the configuration:

```console
distilabel pipeline info --config "https://huggingface.co/datasets/distilabel-internal-testing/first-distiset-with-routed-batches/raw/main/pipeline.yaml"
```

## Dataset structure

The examples have the following structure per configuration:

<details><summary> Configuration: default </summary><hr>

```json
{
"completion": "h\u01ceo",
"merged_generation": [
"I slept for 1.2750293183691193 seconds",
"I slept for 1.2232107381488229 seconds"
],
"meta": {
"category": null,
"completion": "h\u01ceo",
"id": 125,
"input": "\u597d",
"motivation_app": "Doulingo",
"prompt": "What sound does this make?\n Input: \u597d",
"source": "self-instruct",
"subcategory": null
},
"prompt": "What sound does this make?\n Input: \u597d"
}
```

This subset can be loaded as:

```python
from datasets import load_dataset

ds = load_dataset("distilabel-internal-testing/first-distiset-with-routed-batches", "default")
```

Or simply as it follows, since there's only one configuration and is named `default`:

```python
from datasets import load_dataset

ds = load_dataset("distilabel-internal-testing/first-distiset-with-routed-batches")
```

</details>

size_categories: 样本规模：样本量少于1000
dataset_info:
数据集信息：
features:
- 字段：提示词（prompt），数据类型：字符串
- 字段：补全结果（completion），数据类型：字符串
- 字段：元数据（meta），结构体：
- 字段：类别（category），数据类型：字符串
- 字段：补全结果（completion），数据类型：字符串
- 字段：编号（id），数据类型：64位整数
- 字段：输入（input），数据类型：字符串
- 字段：应用动因（motivation_app），数据类型：字符串
- 字段：提示词（prompt），数据类型：字符串
- 字段：来源（source），数据类型：字符串
- 字段：子类别（subcategory），数据类型：字符串
- 字段：合并生成序列（merged_generation），数据类型：字符串序列
splits:
- 拆分名称：训练集（train），数据字节数：458933，样本数量：327
download_size: 292464
dataset_size: 458933
configs:
- 配置名称：默认配置（default），数据文件：
- split: train，文件路径：data/train-*
tags:
- 合成数据集（synthetic）
- Distilabel
- 人工智能反馈强化学习（RLAIF）

<p align="left">
<a href="https://github.com/argilla-io/distilabel">
<img src="https://raw.githubusercontent.com/argilla-io/distilabel/main/docs/assets/distilabel-badge-light.png" alt="基于Distilabel构建" width="200" height="32"/>
</a>
</p>

# 数据集卡片：首个带路由批次的数据集（first-distiset-with-routed-batches）

本数据集由[Distilabel](https://distilabel.argilla.io/)工具构建完成。

## 数据集概览

本数据集包含一个`pipeline.yaml`配置文件，可通过`distilabel`命令行工具复现生成该数据集的流水线：
console
distilabel pipeline run --config "https://huggingface.co/datasets/distilabel-internal-testing/first-distiset-with-routed-batches/raw/main/pipeline.yaml"

或查看该流水线的配置详情：
console
distilabel pipeline info --config "https://huggingface.co/datasets/distilabel-internal-testing/first-distiset-with-routed-batches/raw/main/pipeline.yaml"

## 数据集结构

每个配置下的样本结构如下：

<details><summary> 配置：默认配置（default） </summary><hr>

json
{
"completion": "hu01ceo",
"merged_generation": [
"I slept for 1.2750293183691193 seconds",
"I slept for 1.2232107381488229 seconds"
],
"meta": {
"category": null,
"completion": "hu01ceo",
"id": 125,
"input": "u597d",
"motivation_app": "Doulingo",
"prompt": "What sound does this make?
Input: u597d",
"source": "self-instruct",
"subcategory": null
},
"prompt": "What sound does this make?
Input: u597d"
}

可通过以下代码加载该子集：
python
from datasets import load_dataset

ds = load_dataset("distilabel-internal-testing/first-distiset-with-routed-batches", "default")

由于仅存在一个名为默认配置的配置项，也可通过以下极简方式加载：
python
from datasets import load_dataset

ds = load_dataset("distilabel-internal-testing/first-distiset-with-routed-batches")