metadata
dataset_info:
features:
- name: answer
dtype: string
- name: text
dtype: string
- name: instruction
dtype: string
splits:
- name: train
num_bytes: 26792708
num_examples: 6251
- name: valid
num_bytes: 3730508
num_examples: 883
- name: test
num_bytes: 4798186
num_examples: 1147
download_size: 15696154
dataset_size: 35321402
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: valid
path: data/valid-*
- split: test
path: data/test-*