metadata
license: apache-2.0
dataset_info:
features:
- name: arSubject
dtype: string
- name: arPredicate
dtype: string
- name: arSentence
dtype: string
- name: oldArObject
dtype: string
- name: arObject
dtype: string
- name: maskedArSentence
dtype: string
- name: enSentence
dtype: string
- name: enSubject
dtype: string
- name: enPredicate
dtype: string
- name: enObject
dtype: string
- name: non_commonsense_sentence
dtype: string
- name: label
dtype: string
- name: non_commonsense_label
dtype: string
splits:
- name: train
num_bytes: 6254795
num_examples: 14811
download_size: 2594182
dataset_size: 6254795
configs:
- config_name: default
data_files:
- split: train
path: data/train-*