metadata
dataset_info:
features:
- name: dataset_name
dtype: string
- name: subset_name
dtype: string
- name: prompt_id
dtype: string
- name: template_name
dtype: string
- name: dataset_key
dtype: string
- name: output
dtype: string
- name: instruction
dtype: string
- name: context
dtype: string
splits:
- name: train
num_bytes: 93813824
num_examples: 100000
download_size: 43554485
dataset_size: 93813824
configs:
- config_name: default
data_files:
- split: train
path: data/train-*