datadreamer-dev/abstracts_and_tweets

---
dataset_info:
features:
- name: abstracts
dtype: string
- name: prompts
dtype: string
- name: tweets
dtype: string
splits:
- name: train
num_bytes: 3127163
num_examples: 900
- name: validation
num_bytes: 343839
num_examples: 100
download_size: 1765300
dataset_size: 3471002
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
library_name: datadreamer
size_categories:
- 1K<n<10K
tags:
- datadreamer
- datadreamer-0.1.0
- synthetic
- gpt-4
- gpt-4
---
# Dataset Card

This a synthetic dataset of arXiv-style research paper abstracts and tweets summarizing them used as a demonstration of the [DataDreamer 🤖💤 library](https://datadreamer.dev/docs/latest/). It was used to train an ["Abstract to Tweet" model](https://huggingface.co/datadreamer-dev/abstracts_to_tweet_model).

---
This dataset was produced with [DataDreamer 🤖💤](https://datadreamer.dev). The synthetic dataset card can be found [here](datadreamer.json).

原始信息汇总

数据集信息

特征

• abstracts: 数据类型为字符串
• prompts: 数据类型为字符串
• tweets: 数据类型为字符串

数据分割

• train:
  • 字节数: 3127163
  • 样本数: 900
• validation:
  • 字节数: 343839
  • 样本数: 100

数据大小

• 下载大小: 1765300 字节
• 数据集大小: 3471002 字节

配置

• config_name: default
  • data_files:
    • train: data/train-*
    • validation: data/validation-*

库名称

• library_name: datadreamer

大小分类

• 1K < n < 10K

标签

• datadreamer
• datadreamer-0.1.0
• synthetic
• gpt-4