metadata
dataset_info:
features:
- name: prompt_id
dtype: string
- name: url
dtype: string
- name: subhealine
dtype: string
- name: body
dtype: string
- name: category
dtype: string
- name: id
dtype: string
- name: original_headline
dtype: string
- name: eyecatcher_id
dtype: string
- name: test_week
dtype: int64
- name: total_impressions
dtype: int64
- name: total_clicks
dtype: int64
- name: average_significance
dtype: float64
splits:
- name: train
num_bytes: 29661348
num_examples: 21674
download_size: 7280174
dataset_size: 29661348
configs:
- config_name: default
data_files:
- split: train
path: data/train-*