metadata
dataset_info:
features:
- name: conversations
list:
- name: role
dtype: string
- name: content
dtype: string
splits:
- name: train
num_bytes: 163249406.4951144
num_examples: 168193
- name: validation
num_bytes: 20406054.48593749
num_examples: 21024
- name: test
num_bytes: 20407025.09355193
num_examples: 21025
download_size: 104683787
dataset_size: 204062486.0746038
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*