metadata
dataset_info:
features:
- name: text
struct:
- name: attention_mask
sequence: int64
- name: input_ids
sequence: int64
- name: labels
sequence: int64
splits:
- name: train
num_bytes: 194052036
num_examples: 133317
download_size: 17104280
dataset_size: 194052036