metadata
dataset_info:
features:
- name: input_values
sequence:
sequence:
sequence: float32
- name: attention_mask
sequence:
sequence: int32
- name: labels
dtype: int64
splits:
- name: train
num_bytes: 1895911184
num_examples: 5213
download_size: 1773617134
dataset_size: 1895911184