metadata
dataset_info:
features:
- name: ind
dtype: int64
- name: activity_label
dtype: string
- name: ctx_a
dtype: string
- name: ctx_b
dtype: string
- name: ctx
dtype: string
- name: split
dtype: string
- name: split_type
dtype: string
- name: label
dtype: int64
- name: endings
sequence: string
- name: source_id
dtype: string
splits:
- name: train
num_bytes: 12408061
num_examples: 10042
download_size: 6957034
dataset_size: 12408061
configs:
- config_name: default
data_files:
- split: train
path: data/train-*