metadata
dataset_info:
features:
- name: conversation
list:
- name: content
dtype: string
- name: do_train
dtype: bool
- name: role
dtype: string
splits:
- name: train
num_bytes: 32468994.422971472
num_examples: 74250
download_size: 12747860
dataset_size: 32468994.422971472