metadata
dataset_info:
features:
- name: input_ids
sequence: int32
- name: grouping_vector
sequence:
sequence: int64
- name: result_labels
sequence: int64
- name: attention_mask
sequence: int8
splits:
- name: train
num_bytes: 305693
num_examples: 18
download_size: 54541
dataset_size: 305693
configs:
- config_name: default
data_files:
- split: train
path: data/train-*