metadata
dataset_info:
features:
- name: sample_id
dtype: string
- name: text
dtype: string
splits:
- name: train
num_bytes: 3406678
num_examples: 45814
- name: pretrain
num_bytes: 12176825
num_examples: 89439
- name: validation
num_bytes: 81227
num_examples: 1082
- name: test
num_bytes: 80738
num_examples: 1243
download_size: 7894841
dataset_size: 15745468
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: pretrain
path: data/pretrain-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*