metadata
language:
- hi
dataset_info:
features:
- name: doc_id
dtype: string
- name: type
dtype: string
- name: text
dtype: string
- name: input_ids
sequence: int32
- name: attention_mask
sequence: int8
splits:
- name: train
num_bytes: 2731009
num_examples: 62
download_size: 988704
dataset_size: 2731009
configs:
- config_name: default
data_files:
- split: train
path: data/train-*