metadata
dataset_info:
features:
- name: doc
dtype: string
- name: text
dtype: string
- name: label
dtype: int64
- name: prompt
dtype: string
splits:
- name: train
num_bytes: 883773207
num_examples: 814023
download_size: 202746898
dataset_size: 883773207
configs:
- config_name: default
data_files:
- split: train
path: data/train-*