metadata
dataset_info:
features:
- name: input_ids
sequence: int32
- name: attention_mask
sequence: int8
- name: labels
sequence: int64
splits:
- name: train
num_bytes: 146421104
num_examples: 37516
- name: validation
num_bytes: 18453961
num_examples: 4689
- name: test
num_bytes: 18468141
num_examples: 4689
download_size: 59047247
dataset_size: 183343206
configs:
- config_name: default
data_files:
- split: train
path: mt5base-encoded/train-*
- split: validation
path: mt5base-encoded/validation-*
- split: test
path: mt5base-encoded/test-*