metadata
dataset_info:
features:
- name: context
dtype: string
- name: summary
dtype: string
- name: sentences
sequence: string
- name: instruction
dtype: string
splits:
- name: train
num_bytes: 14196998433
num_examples: 203037
- name: validation
num_bytes: 435239658
num_examples: 6436
- name: test
num_bytes: 436024954
num_examples: 6440
download_size: 7276254120
dataset_size: 15068263045
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*