metadata
dataset_info:
features:
- name: id
dtype: int64
- name: instruction
dtype: string
- name: output
dtype: int64
- name: text
dtype: string
splits:
- name: train
num_bytes: 20621918
num_examples: 10840
- name: validation
num_bytes: 5105812
num_examples: 2710
- name: test
num_bytes: 1395314
num_examples: 1000
download_size: 18003364
dataset_size: 27123044
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*