metadata
dataset_info:
features:
- name: sentence
dtype: string
- name: label
dtype: string
- name: field
dtype: string
- name: source
dtype: string
splits:
- name: consensus
num_bytes: 14076979
num_examples: 65907
- name: train_balanced
num_bytes: 6409783
num_examples: 30000
download_size: 10510948
dataset_size: 20486762
configs:
- config_name: default
data_files:
- split: consensus
path: data/consensus-*
- split: train_balanced
path: data/train_balanced-*