metadata
dataset_info:
features:
- name: input_ids
sequence: float32
- name: label
dtype: float64
splits:
- name: train
num_bytes: 666148
num_examples: 643
download_size: 34483
dataset_size: 666148
configs:
- config_name: default
data_files:
- split: train
path: data/train-*