metadata
dataset_info:
features:
- name: tokens
sequence: int64
- name: tragets
sequence: int64
splits:
- name: train
num_bytes: 49288800
num_examples: 3300
- name: eval
num_bytes: 2748224
num_examples: 184
- name: test
num_bytes: 1493600
num_examples: 100
download_size: 3341205
dataset_size: 53530624
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: eval
path: data/eval-*
- split: test
path: data/test-*