metadata
dataset_info:
features:
- name: 'Unnamed: 0.1'
dtype: int64
- name: 'Unnamed: 0'
dtype: int64
- name: output
dtype: string
- name: input
dtype: string
- name: instruction
dtype: string
splits:
- name: train
num_bytes: 51191747
num_examples: 24061
download_size: 25324137
dataset_size: 51191747
configs:
- config_name: default
data_files:
- split: train
path: data/train-*