metadata
size_categories: n<1K
dataset_info:
features:
- name: _id
dtype: string
- name: url
dtype: string
- name: title
dtype: string
- name: text
dtype: string
- name: score
dtype: float64
- name: views
dtype: float64
- name: model_name
dtype: string
- name: query
dtype: string
splits:
- name: train
num_bytes: 820541
num_examples: 1500
download_size: 534948
dataset_size: 820541
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
tags:
- synthetic
- distilabel
- rlaif
Dataset Card for cohere-wikipedia-2023-11-no-queries
This dataset has been created with distilabel.
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:
distilabel pipeline run --config "https://huggingface.co/datasets/rasdani/cohere-wikipedia-2023-11-no-queries/raw/main/pipeline.yaml"
or explore the configuration:
distilabel pipeline info --config "https://huggingface.co/datasets/rasdani/cohere-wikipedia-2023-11-no-queries/raw/main/pipeline.yaml"
Dataset structure
The examples have the following structure per configuration:
Configuration: default
{
"_id": "20231101.no_262035_4",
"model_name": "gpt-4o",
"query": "Hva er eksekutivfunksjoner i nevropsykologi?",
"score": 1.0,
"text": "Evne til \u00e5 strukturere, planlegge, igangsette og kontrollere sammensatte handlinger (eksekutivfunksjoner).",
"title": "Nevropsykologi",
"url": "https://no.wikipedia.org/wiki/Nevropsykologi",
"views": 728.6967671316081
}
This subset can be loaded as:
from datasets import load_dataset
ds = load_dataset("rasdani/cohere-wikipedia-2023-11-no-queries", "default")
Or simply as it follows, since there's only one configuration and is named default
:
from datasets import load_dataset
ds = load_dataset("rasdani/cohere-wikipedia-2023-11-no-queries")