mistral_lora_clm_with_added_tokens / tokenizer_config.json
thomasgauthier's picture
Upload folder using huggingface_hub
0ff4f8a verified
{
"add_bos_token": true,
"add_eos_token": false,
"add_prefix_space": null,
"added_tokens_decoder": {
"0": {
"content": "<unk>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"1": {
"content": "<s>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"2": {
"content": "</s>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"32000": {
"content": "<|startoftext|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"32001": {
"content": "<|endtarget|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"32002": {
"content": "<|pad|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"32003": {
"content": "<|begintarget|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"32004": {
"content": "<|begincontext|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"32005": {
"content": "<|endcontext|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"32006": {
"content": "<|system|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"32007": {
"content": "<|user|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"32008": {
"content": "<|beginlastuserutterance|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"32009": {
"content": "<|endlastuserutterance|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"32010": {
"content": "<|begindsts|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"32011": {
"content": "<|enddsts|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"32012": {
"content": "<|begindst|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"32013": {
"content": "<|enddst|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"32014": {
"content": "<|beginbelief|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"32015": {
"content": "<|endbelief|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"32016": {
"content": "<|beginresponse|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"32017": {
"content": "<|endresponse|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"32018": {
"content": "<|beginaction|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"32019": {
"content": "<|endaction|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"32020": {
"content": "<|beginuseraction|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"32021": {
"content": "<|enduseraction|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"32022": {
"content": "<|sysactions|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"32023": {
"content": "<|beginintent|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"32024": {
"content": "<|endintent|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"32025": {
"content": "<|beginrequestedslots|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
},
"32026": {
"content": "<|endrequestedslots|>",
"lstrip": false,
"normalized": false,
"rstrip": false,
"single_word": false,
"special": true
}
},
"additional_special_tokens": [
"<|begintarget|>",
"<|endtarget|>",
"<|begincontext|>",
"<|endcontext|>",
"<|system|>",
"<|user|>",
"<|beginlastuserutterance|>",
"<|endlastuserutterance|>",
"<|begindsts|>",
"<|enddsts|>",
"<|begindst|>",
"<|enddst|>",
"<|beginbelief|>",
"<|endbelief|>",
"<|beginresponse|>",
"<|endresponse|>",
"<|beginaction|>",
"<|endaction|>",
"<|beginuseraction|>",
"<|enduseraction|>",
"<|sysactions|>",
"<|beginintent|>",
"<|endintent|>",
"<|beginrequestedslots|>",
"<|endrequestedslots|>",
"<|pad|>",
"<|startoftext|>"
],
"bos_token": "<|startoftext|>",
"clean_up_tokenization_spaces": false,
"eos_token": "<|endtarget|>",
"legacy": false,
"model_max_length": 1000000000000000019884624838656,
"pad_token": "<|pad|>",
"sp_model_kwargs": {},
"spaces_between_special_tokens": false,
"tokenizer_class": "LlamaTokenizer",
"unk_token": "<unk>",
"use_default_system_prompt": false
}