metadata
dataset_info:
features:
- name: input_ids
sequence: int64
- name: labels
sequence: int64
- name: context_size
dtype: int64
- name: __index_level_0__
dtype: int64
splits:
- name: train
num_bytes: 459270352
num_examples: 6060736
- name: test
num_bytes: 114803248
num_examples: 1515184
download_size: 280079346
dataset_size: 574073600
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*