metadata
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- config_name: mt0-labse
data_files:
- split: test
path: mt0-labse/test-*
- split: train
path: mt0-labse/train-*
dataset_info:
- config_name: default
features:
- name: lp
dtype: string
- name: src
dtype: string
- name: mt
dtype: string
- name: ref
dtype: string
- name: score
dtype: float64
- name: score_type
dtype: string
splits:
- name: train
num_bytes: 718521119
num_examples: 1527567
- name: test
num_bytes: 31104504
num_examples: 77575
download_size: 0
dataset_size: 749625623
- config_name: mt0-labse
features:
- name: labels
dtype: float64
- name: input_ids
sequence: int32
- name: attention_mask
sequence: int8
- name: labse
sequence: float32
splits:
- name: test
num_bytes: 751787050
num_examples: 77575
- name: train
num_bytes: 14966913687
num_examples: 1527567
download_size: 16284469907
dataset_size: 15718700737