metadata
dataset_info:
config_name: line
features:
- name: title
dtype: string
- name: genre
dtype: string
- name: mungchi
sequence: int64
- name: output
dtype: string
- name: guidance
struct:
- name: former
sequence: string
- name: latter
sequence: string
- name: lyrics
dtype: string
- name: instruction
dtype: string
- name: __index_level_0__
dtype: int64
splits:
- name: train
num_bytes: 1066493721.3160003
num_examples: 429100
- name: test
num_bytes: 118499854.68399969
num_examples: 47678
download_size: 515356250
dataset_size: 1184993576
configs:
- config_name: line
data_files:
- split: train
path: line/train-*
- split: test
path: line/test-*