metadata
dataset_info:
features:
- name: text_a
dtype: string
- name: text_b
dtype: string
- name: RE
dtype: float64
- name: AR
dtype: float64
- name: BE
dtype: float64
- name: EE
dtype: float64
- name: num_score
dtype: float64
- name: labels
dtype: int64
- name: 'Unnamed: 0'
dtype: float64
- name: toxicity
dtype: float64
- name: sentiment
dtype: float64
- name: engagingness_score
dtype: float64
splits:
- name: train
num_bytes: 393519526
num_examples: 683573
download_size: 248101134
dataset_size: 393519526
configs:
- config_name: default
data_files:
- split: train
path: data/train-*