metadata
dataset_info:
features:
- name: _id
dtype: string
- name: id
dtype: string
- name: created_at
dtype: timestamp[ns, tz=UTC]
- name: downloads_all_time
dtype: int64
- name: downloads
dtype: int64
- name: likes
dtype: int64
- name: trending_score
dtype: float64
- name: tokenizer_str
dtype: string
splits:
- name: train
num_bytes: 154927808562
num_examples: 150000
download_size: 85406627457
dataset_size: 154927808562
configs:
- config_name: default
data_files:
- split: train
path: data/train-*