metadata
dataset_info:
features:
- name: script
dtype: string
- name: summary
dtype: string
- name: emb
sequence: float32
- name: emb2
sequence: float64
splits:
- name: train
num_bytes: 22578544
num_examples: 1370
download_size: 18677123
dataset_size: 22578544
configs:
- config_name: default
data_files:
- split: train
path: data/train-*