metadata
dataset_info:
features:
- name: id
dtype: int64
- name: input_ids
sequence: int32
- name: attention_mask
sequence: int8
- name: uncertainty_labels
sequence: int64
splits:
- name: train
num_bytes: 40091700
num_examples: 3300
- name: eval
num_bytes: 2235416
num_examples: 184
- name: test
num_bytes: 1214900
num_examples: 100
download_size: 3372483
dataset_size: 43542016
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: eval
path: data/eval-*
- split: test
path: data/test-*