metadata
dataset_info:
features:
- name: id
dtype: string
- name: title
dtype: string
- name: selftext
dtype: string
- name: subreddit
dtype: string
- name: created_utc
dtype: string
- name: upvote_ratio
dtype: float64
- name: score
dtype: int64
- name: num_comments
dtype: int64
- name: author_id
dtype: float64
- name: author
dtype: string
- name: author_created_utc
dtype: string
- name: media_embed
struct:
- name: content
dtype: string
- name: height
dtype: float64
- name: media_domain_url
dtype: 'null'
- name: scrolling
dtype: bool
- name: width
dtype: float64
- name: __index_level_0__
dtype: int64
- name: stella_embeddings
sequence: float64
- name: is_coping
dtype: bool
- name: coping_confidence
dtype: float32
- name: is_healing
dtype: bool
- name: healing_confidence
dtype: float32
- name: is_political_anxiety
dtype: bool
- name: political_anxiety_confidence
dtype: float32
- name: coping_cosine
dtype: float32
- name: healing_cosine
dtype: float32
- name: political_anxiety_cosine
dtype: float32
splits:
- name: train
num_bytes: 398033478
num_examples: 47070
download_size: 307674605
dataset_size: 398033478
configs:
- config_name: default
data_files:
- split: train
path: data/train-*