distilabel-internal-testing/dpo-mix-4k-criticurus-temperature0-v0.0

---
size_categories: n<1K
dataset_info:
features:
- name: instruction
dtype: string
- name: response
dtype: string
- name: rating
dtype: float64
- name: dataset_name
dtype: string
- name: model_name
dtype: string
- name: score
dtype: string
- name: critique
dtype: string
- name: raw_output
dtype: string
splits:
- name: train
num_bytes: 9817993
num_examples: 3996
download_size: 4624133
dataset_size: 9817993
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 dpo-mix-4k-criticurus-temperature0-v0.0

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/dpo-mix-4k-criticurus-temperature0-v0.0/raw/main/pipeline.yaml"
```

or explore the configuration:

```console
distilabel pipeline info --config "https://huggingface.co/datasets/distilabel-internal-testing/dpo-mix-4k-criticurus-temperature0-v0.0/raw/main/pipeline.yaml"
```

## Dataset structure

The examples have the following structure per configuration:

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

```json
{
"critique": "You\u0027ve done a good job identifying the anagram of \" dirty room \" as \" Dormary \" which is a good start. However, the phrase \"dormary\" is not a commonly used term for a place where students live. It\u0027s important to ensure that the answer you provide is accurate and well-known. \n\nTo improve, you could have used a more common term like \"dormition\" which is a more common anagram of \" dirty room \" and is a place where students live. \n\nRemember, accuracy and understanding of the context are key in providing a helpful and truthful answer.",
"dataset_name": "argilla/distilabel-capybara-dpo-7k-binarized",
"instruction": "A phrase that\u0027s an anagram of \"dirty room\", it refers to a place where students live.",
"model_name": "distilabel-internal-testing/criticurus-v0.0",
"rating": 5.0,
"raw_output": null,
"response": "dormitory",
"score": "6\u003c|im_end|\u003e"
}
```

This subset can be loaded as:

```python
from datasets import load_dataset

ds = load_dataset("distilabel-internal-testing/dpo-mix-4k-criticurus-temperature0-v0.0", "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/dpo-mix-4k-criticurus-temperature0-v0.0")
```

</details>

原始信息汇总

数据集概述

数据集基本信息

• 数据集名称: dpo-mix-4k-criticurus-temperature0-v0.0
• 数据集大小:
  • 下载大小: 4624133字节
  • 数据集大小: 9817993字节
• 示例数量: 3996
• 分类: 小于1K

数据集特征

• 特征名称：instruction, response, rating, dataset_name, model_name, score, critique, raw_output
• 数据类型：
  • instruction: string
  • response: string
  • rating: float64
  • dataset_name: string
  • model_name: string
  • score: string
  • critique: string
  • raw_output: string

数据集结构

• 配置名称: default
• 数据文件路径: data/train-*
• 示例结构: json { "critique": ..., "dataset_name": "argilla/distilabel-capybara-dpo-7k-binarized", "instruction": ..., "model_name": "distilabel-internal-testing/criticurus-v0.0", "rating": 5.0, "raw_output": null, "response": "dormitory", "score": "6u003c|im_end|u003e" }

数据集加载

• 加载方式: python from datasets import load_dataset ds = load_dataset("distilabel-internal-testing/dpo-mix-4k-criticurus-temperature0-v0.0", "default")

  或 python from datasets import load_dataset ds = load_dataset("distilabel-internal-testing/dpo-mix-4k-criticurus-temperature0-v0.0")