metadata
dataset_info:
features:
- name: text
dtype: string
- name: instruction
dtype: string
- name: label
dtype: string
- name: type
dtype: string
splits:
- name: train
num_bytes: 1567310
num_examples: 6493
- name: valid
num_bytes: 224258
num_examples: 928
- name: test
num_bytes: 447925
num_examples: 1856
download_size: 304861
dataset_size: 2239493
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: valid
path: data/valid-*
- split: test
path: data/test-*