metadata
dataset_info:
features:
- name: SELFIES
dtype: string
- name: input_ids
sequence: int32
- name: token_type_ids
sequence: int8
- name: special_tokens_mask
sequence: int8
- name: attention_mask
sequence: int8
- name: length
dtype: int64
splits:
- name: train
num_bytes: 70860000120.62552
num_examples: 170768422
- name: test
num_bytes: 12504312569.237696
num_examples: 30135604
download_size: 24105151007
dataset_size: 83364312689.86322
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*