metadata
dataset_info:
features:
- name: text
dtype: string
- name: labels
sequence: int64
- name: input_ids
sequence: int32
- name: attention_mask
sequence: int8
splits:
- name: train
num_bytes: 96564675
num_examples: 133317
download_size: 26452902
dataset_size: 96564675
configs:
- config_name: default
data_files:
- split: train
path: data/train-*