metadata
dataset_info:
features:
- name: prompt
list:
- name: content
dtype: string
- name: role
dtype: string
- name: completion
dtype: string
- name: conv_id
dtype: string
- name: metadata
struct:
- name: article_id
dtype: string
- name: boosted_at
dtype: string
- name: claps
dtype: int64
- name: id
dtype: string
- name: image_url
dtype: string
- name: is_locked
dtype: bool
- name: is_series
dtype: bool
- name: is_shortform
dtype: bool
- name: lang
dtype: string
- name: last_modified_at
dtype: string
- name: model
dtype: string
- name: publication_id
dtype: string
- name: published_at
dtype: string
- name: reading_time
dtype: float64
- name: responses_count
dtype: int64
- name: subtitle
dtype: string
- name: tags
sequence: string
- name: top_highlight
dtype: string
- name: topics
sequence: string
- name: unique_slug
dtype: string
- name: url
dtype: string
- name: voters
dtype: int64
- name: word_count
dtype: int64
- name: turn_id
dtype: int64
splits:
- name: train
num_bytes: 886637
num_examples: 200
download_size: 516017
dataset_size: 886637
configs:
- config_name: default
data_files:
- split: train
path: data/train-*