metadata
dataset_info:
features:
- name: keyword
dtype: string
- name: text
dtype: string
- name: response
struct:
- name: content
dtype: string
- name: task
dtype: string
- name: task
dtype: string
- name: __index_level_0__
dtype: int64
splits:
- name: train
num_bytes: 9568752
num_examples: 2961
- name: test
num_bytes: 1217698
num_examples: 329
download_size: 3973534
dataset_size: 10786450
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*