metadata
dataset_info:
features:
- name: text.en
dtype: string
- name: text.fr
dtype: string
- name: input_ids.en
sequence: int64
- name: input_ids.fr
sequence: int64
splits:
- name: train
num_bytes: 12161836
num_examples: 29000
- name: valid
num_bytes: 427078
num_examples: 1014
download_size: 4236558
dataset_size: 12588914
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: valid
path: data/valid-*