Optimizer-sealion2pandas / tokenizer_config.json
kkatiz's picture
Upload tokenizer
6250d67 verified
{
"add_bos_token": false,
"add_eos_token": false,
"added_tokens_decoder": {
"0": {
"content": "<unk>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"1": {
"content": "<|endoftext|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"2": {
"content": "<|endofline|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"3": {
"content": "<|padding|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
}
},
"auto_map": {
"AutoTokenizer": [
"aisingapore/sea-lion-7b-instruct--tokenization_SEA_BPE.SEABPETokenizer",
null
]
},
"bos_token": null,
"clean_up_tokenization_spaces": false,
"eos_token": "<|endoftext|>",
"legacy": true,
"model_max_length": 1000000000000000019884624838656,
"pad_token": "<|padding|>",
"sp_model_kwargs": {},
"tokenizer_class": "SEABPETokenizer",
"unk_token": "<unk>"
}