metadata
dataset_info:
features:
- name: instruction
dtype: string
- name: input
dtype: string
- name: id
dtype: string
- name: peak_text
dtype: string
- name: output
dtype: string
splits:
- name: train
num_bytes: 3523667
num_examples: 4572
- name: test
num_bytes: 387700
num_examples: 508
download_size: 1427338
dataset_size: 3911367
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*