metadata
dataset_info:
features:
- name: target_smiles
dtype: string
- name: prompt
list:
- name: content
dtype: string
- name: role
dtype: string
splits:
- name: train
num_bytes: 71371756
num_examples: 20000
- name: test
num_bytes: 33129765
num_examples: 9269
download_size: 16004348
dataset_size: 104501521
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*