metadata
dataset_info:
features:
- name: k
dtype: string
- name: v
sequence: int64
splits:
- name: train
num_bytes: 26888890
num_examples: 100000
download_size: 585007
dataset_size: 26888890
configs:
- config_name: default
data_files:
- split: train
path: data/train-*