metadata
dataset_info:
features:
- name: label
dtype: int64
- name: input_ids
sequence: int32
- name: attention_mask
sequence: int8
splits:
- name: train
num_bytes: 480848
num_examples: 1688
- name: test
num_bytes: 39679
num_examples: 139
- name: dev
num_bytes: 31269
num_examples: 114
download_size: 234953
dataset_size: 551796
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: dev
path: data/dev-*