metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: first_mask_count
dtype: int64
- name: first_start_index
dtype: int64
- name: first_mask_gen_count
dtype: int64
- name: first_mask_tokens
dtype: string
- name: first_mask_gen_count_real
dtype: int64
- name: generated_smiles
dtype: string
- name: original_smiles
dtype: string
- name: generated_tokens
dtype: string
splits:
- name: train
num_bytes: 233321100
num_examples: 1000000
download_size: 52920692
dataset_size: 233321100
configs:
- config_name: default
data_files:
- split: train
path: data/train-*