metadata
dataset_info:
features:
- name: text
dtype: string
- name: test_text
dtype: string
- name: speaker
dtype: string
- name: llama_text
dtype: string
- name: llama_test_text
dtype: string
- name: qwen_text
dtype: string
- name: qwen_test_text
dtype: string
- name: book
dtype: string
splits:
- name: test
num_bytes: 18458538
num_examples: 1694
- name: train
num_bytes: 68638006
num_examples: 5694
download_size: 45010570
dataset_size: 87096544
configs:
- config_name: default
data_files:
- split: test
path: data/test-*
- split: train
path: data/train-*