gabrielmbmb/testing-vllm

---
size_categories: n<1K
dataset_info:
- config_name: text_generation_0
features:
- name: instruction
dtype: string
- name: completion
dtype: string
- name: generation
dtype: string
- name: distilabel_metadata
struct:
- name: raw_output_text_generation_0
dtype: string
- name: model_name
dtype: string
splits:
- name: train
num_bytes: 515517
num_examples: 327
download_size: 338101
dataset_size: 515517
- config_name: text_generation_1
features:
- name: instruction
dtype: string
- name: completion
dtype: string
- name: generation
dtype: string
- name: distilabel_metadata
struct:
- name: raw_output_text_generation_1
dtype: string
- name: model_name
dtype: string
splits:
- name: train
num_bytes: 515517
num_examples: 327
download_size: 338101
dataset_size: 515517
configs:
- config_name: text_generation_0
data_files:
- split: train
path: text_generation_0/train-*
- config_name: text_generation_1
data_files:
- split: train
path: text_generation_1/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 testing-vllm

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/gabrielmbmb/testing-vllm/raw/main/pipeline.yaml"
```

or explore the configuration:

```console
distilabel pipeline info --config "https://huggingface.co/datasets/gabrielmbmb/testing-vllm/raw/main/pipeline.yaml"
```

## Dataset structure

The examples have the following structure per configuration:

<details><summary> Configuration: text_generation_1 </summary><hr>

```json
{
"completion": "Denote the number of chocolates each person has by the letter of their first name. We know that\nA = D + 12\nD = R + 6\nA = 2 * R\n\nThus, A = (R + 6) + 12 = R + 18\nSince also A = 2 * R, this means 2 * R = R + 18\nHence R = 18\nHence D = 18 + 6 = 24",
"distilabel_metadata": {
"raw_output_text_generation_1": "Arianna has 12 more chocolates than Danny, so DANNY + 12 = ARIANNA. Arabic does not have twice as many chocolates as Robbie, so if ARIANNA = 2 * ROBBIE. Since ARIANNA = DANNY + 12, we can plug in the values and get 2 * ROBBIE = DANNY + 12. We know also that DANNY = ROBBIE + 6, so 2 * ROBBIE = ROBBIE + 6 + 12. Simplifying this equation, we get ROBBIE ="
},
"generation": "Arianna has 12 more chocolates than Danny, so DANNY + 12 = ARIANNA. Arabic does not have twice as many chocolates as Robbie, so if ARIANNA = 2 * ROBBIE. Since ARIANNA = DANNY + 12, we can plug in the values and get 2 * ROBBIE = DANNY + 12. We know also that DANNY = ROBBIE + 6, so 2 * ROBBIE = ROBBIE + 6 + 12. Simplifying this equation, we get ROBBIE =",
"instruction": "Arianna has 12 chocolates more than Danny. Danny has 6 chocolates more than Robbie. Arianna has twice as many chocolates as Robbie has. How many chocolates does Danny have?",
"model_name": "meta-llama/Meta-Llama-3-8B-Instruct"
}
```

This subset can be loaded as:

```python
from datasets import load_dataset

ds = load_dataset("gabrielmbmb/testing-vllm", "text_generation_1")
```

</details>

<details><summary> Configuration: text_generation_0 </summary><hr>

```json
{
"completion": "Denote the number of chocolates each person has by the letter of their first name. We know that\nA = D + 12\nD = R + 6\nA = 2 * R\n\nThus, A = (R + 6) + 12 = R + 18\nSince also A = 2 * R, this means 2 * R = R + 18\nHence R = 18\nHence D = 18 + 6 = 24",
"distilabel_metadata": {
"raw_output_text_generation_0": "Arianna has 12 more chocolates than Danny, so DANNY + 12 = ARIANNA. Arabic does not have twice as many chocolates as Robbie, so if ARIANNA = 2 * ROBBIE. Since ARIANNA = DANNY + 12, we can plug in the values and get 2 * ROBBIE = DANNY + 12. We know also that DANNY = ROBBIE + 6, so 2 * ROBBIE = ROBBIE + 6 + 12. Simplifying this equation, we get ROBBIE ="
},
"generation": "Arianna has 12 more chocolates than Danny, so DANNY + 12 = ARIANNA. Arabic does not have twice as many chocolates as Robbie, so if ARIANNA = 2 * ROBBIE. Since ARIANNA = DANNY + 12, we can plug in the values and get 2 * ROBBIE = DANNY + 12. We know also that DANNY = ROBBIE + 6, so 2 * ROBBIE = ROBBIE + 6 + 12. Simplifying this equation, we get ROBBIE =",
"instruction": "Arianna has 12 chocolates more than Danny. Danny has 6 chocolates more than Robbie. Arianna has twice as many chocolates as Robbie has. How many chocolates does Danny have?",
"model_name": "meta-llama/Meta-Llama-3-8B-Instruct"
}
```

This subset can be loaded as:

```python
from datasets import load_dataset

ds = load_dataset("gabrielmbmb/testing-vllm", "text_generation_0")
```

</details>

原始信息汇总

数据集概述

数据集结构

配置: text_generation_0

• 特征:
  • instruction: 类型为 string
  • completion: 类型为 string
  • generation: 类型为 string
  • distilabel_metadata: 包含 raw_output_text_generation_0，类型为 string
  • model_name: 类型为 string
• 分割:
  • train: 包含 327 个样本，占用 515517 字节
• 下载大小: 338101 字节
• 数据集大小: 515517 字节
• 数据文件路径: text_generation_0/train-*

配置: text_generation_1

• 特征:
  • instruction: 类型为 string
  • completion: 类型为 string
  • generation: 类型为 string
  • distilabel_metadata: 包含 raw_output_text_generation_1，类型为 string
  • model_name: 类型为 string
• 分割:
  • train: 包含 327 个样本，占用 515517 字节
• 下载大小: 338101 字节
• 数据集大小: 515517 字节
• 数据文件路径: text_generation_1/train-*

标签

• synthetic
• distilabel
• rlaif