metadata
license: odc-by
dataset_info:
features:
- name: url
dtype: string
- name: tag
dtype: string
- name: text
dtype: string
- name: file_path
dtype: string
- name: dump
dtype: string
- name: file_size_in_byte
dtype: int64
- name: line_count
dtype: int64
splits:
- name: train
num_bytes: 18159796472
num_examples: 5241900
download_size: 9949701917
dataset_size: 18159796472
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
This data corpus is a math corpus recalled from Fineweb using fastText.