|
|
--- |
|
|
task_categories: |
|
|
- text-generation |
|
|
language: |
|
|
- en |
|
|
pretty_name: a |
|
|
size_categories: |
|
|
- 10M<n<100M |
|
|
--- |
|
|
|
|
|
# Dataset Card for Dataset |
|
|
|
|
|
This dataset has the fineweb dataset splitted in 1M rows in each files |
|
|
|
|
|
This dataset card aims to be a base template for new datasets. It has been generated using [this raw template](https://github.com/huggingface/huggingface_hub/blob/main/src/huggingface_hub/templates/datasetcard_template.md?plain=1). |
|
|
|
|
|
## Dataset Details |
|
|
|
|
|
### Dataset Description |
|
|
|
|
|
Each of the files ending with _p{number}.csv has 1M rows in it and they are in series |
|
|
|
|
|
### Dataset Sources |
|
|
|
|
|
This dataset was created from [HuggingFaceFW/fineweb](https://huggingface.co/datasets/HuggingFaceFW/fineweb) |
|
|
|
|
|
## Uses |
|
|
|
|
|
Has text generation data |
|
|
|
|
|
|
|
|
## Dataset Structure |
|
|
|
|
|
Has the columns answer and questions but only use answer |
|
|
|
|
|
### Source Data |
|
|
|
|
|
Web scrapes |
|
|
|
|
|
#### Data Collection and Processing |
|
|
|
|
|
``` |
|
|
from datasets import load_dataset |
|
|
dataset = load_dataset("WICKED4950/Raw-GPT-traindata", data_files="") |
|
|
``` |
