Files
smolgen-pubchem-46M-base/tokenizer_config.json
ModelHub XC 40bc8331f4 初始化项目,由ModelHub XC社区提供模型
Model: ddidacus/smolgen-pubchem-46M-base
Source: Original Platform
2026-04-17 05:58:14 +08:00

37 lines
809 B
JSON

{
"added_tokens_decoder": {
"33": {
"content": "[PAD]",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"34": {
"content": "[EOS]",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"35": {
"content": "[BOS]",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
}
},
"bos_token": "[BOS]",
"clean_up_tokenization_spaces": false,
"eos_token": "[EOS]",
"extra_special_tokens": {},
"model_max_length": 1000000000000000019884624838656,
"pad_token": "[PAD]",
"tokenizer_class": "PreTrainedTokenizerFast",
"unk_token": "[UNK]"
}