metadata
dataset_info:
features:
- name: chat_history
list:
- name: content
dtype: string
- name: role
dtype: string
- name: formatted_chat_history
dtype: string
- name: character_prompt
dtype: string
- name: baked_memory
dtype: string
splits:
- name: train
num_bytes: 2080552
num_examples: 337
download_size: 1130206
dataset_size: 2080552
configs:
- config_name: default
data_files:
- split: train
path: data/train-*