metadata
dataset_info:
features:
- name: gold_label
dtype: string
- name: sentence1_tokens
dtype: string
- name: sentence2_tokens
dtype: string
- name: Flikr30kID
dtype: string
- name: sentence1
dtype: string
- name: sentence2
dtype: string
- name: pairID
dtype: string
- name: Explanation_1
dtype: string
- name: Explanation_1_tokens
dtype: string
splits:
- name: train
num_bytes: 7372371
num_examples: 17572
download_size: 4112339
dataset_size: 7372371
configs:
- config_name: default
data_files:
- split: train
path: data/train-*