metadata
dataset_info:
features:
- name: text
dtype: string
- name: source
dtype: string
- name: input_ids
sequence: int32
- name: attention_mask
sequence: int8
splits:
- name: train
num_bytes: 7104120334.486074
num_examples: 14907475
- name: test
num_bytes: 1087947.6655964435
num_examples: 2303
- name: validation
num_bytes: 1140553.189182514
num_examples: 2351
download_size: 5337569020
dataset_size: 7106348835.340853
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: validation
path: data/validation-*