metadata
dataset_info:
features:
- name: inputs
dtype: string
- name: targets
dtype: string
- name: language
dtype: string
- name: language_code
dtype: string
- name: annotation_type
dtype: string
- name: user_id
dtype: string
splits:
- name: train
num_bytes: 4836140.556250679
num_examples: 3844
- name: test
num_bytes: 125810.10812306659
num_examples: 100
download_size: 1606520
dataset_size: 4961950.664373746
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*