Transformers
OLMO_Smiles_aware_tokenizer / tokenizer_config.json
Codemaster67's picture
Add SPE tokenizer (1125 SMILES subword tokens) + <|start_of_smiles|>/<|end_of_smiles|> special tokens. Trained on 2M ZINC20 + 2M ChEMBL canonical SMILES. SPE vocab_size=1000, min_freq=4000.
f6eb1d8 verified
{
"add_prefix_space": false,
"backend": "tokenizers",
"bos_token": null,
"clean_up_tokenization_spaces": true,
"eos_token": "<|endoftext|>",
"errors": "replace",
"extra_special_tokens": [
"<|start_of_smiles|>",
"<|end_of_smiles|>"
],
"is_local": false,
"local_files_only": false,
"model_max_length": 1000000000000000019884624838656,
"pad_token": "<|padding|>",
"tokenizer_class": "GPTNeoXTokenizer",
"trim_offsets": true,
"unk_token": null
}