LongCat-Next-4bit / tokenizer_config.json
kernelpool's picture
Add files using upload-large-folder tool
74da6da verified
{
"add_prefix_space": false,
"audio_end_token": "<longcat_audio_end>",
"audio_pad_token": "<longcat_audio_pad>",
"audio_start_token": "<longcat_audio_start>",
"backend": "tokenizers",
"bos_token": "<longcat_s>",
"clean_up_tokenization_spaces": false,
"eos_token": "</longcat_s>",
"extra_special_tokens": [
"<mask_131048>",
"<mask_131049>",
"<mask_131050>",
"<mask_131051>",
"<mask_131052>",
"<mask_131053>",
"<mask_131054>",
"<mask_131055>",
"<mask_131056>",
"<mask_131057>",
"<mask_131058>",
"<mask_131059>",
"<mask_131060>",
"<mask_131061>",
"<mask_131062>",
"<mask_131063>",
"<mask_131064>",
"<mask_131065>",
"<longcat_img_token_size>",
"</longcat_img_token_size>",
"<mask_131068>",
"<mask_131069>",
"<mask_131070>",
"<mask_131071>",
"<longcat_point_start>",
"<longcat_point_end>",
"<longcat_point_delim>",
"<longcat_polygon_start>",
"<longcat_polygon_end>",
"<mask_131077>",
"<mask_131078>",
"<longcat_audio_start>",
"<longcat_audio_end>",
"<longcat_audio_pad>",
"<longcat_img_start>",
"<longcat_img_end>",
"<longcat_img_pad>",
"<longcat_img_newline>",
"<longcat_box_start>",
"<longcat_box_end>",
"<longcat_box_delim>",
"<longcat_ref_start>",
"<longcat_ref_end>",
"<longcat_img_delim>",
"<longcat_audio_delim>",
"<longcat_video_palce>",
"<longcat_video_start>",
"<longcat_video_end>",
"<longcat_audiotext_start>",
"<longcat_audiotext_end>",
"<longcat_audiotext_pad>",
"<longcat_audiogen_start>",
"<longcat_audiogen_end>"
],
"image_end_token": "<longcat_img_end>",
"image_newline_token": "<longcat_img_newline>",
"image_pad_token": "<longcat_img_pad>",
"image_start_token": "<longcat_img_start>",
"is_local": true,
"model_max_length": 131072,
"model_specific_special_tokens": {
"audio_end_token": "<longcat_audio_end>",
"audio_pad_token": "<longcat_audio_pad>",
"audio_start_token": "<longcat_audio_start>",
"image_end_token": "<longcat_img_end>",
"image_newline_token": "<longcat_img_newline>",
"image_pad_token": "<longcat_img_pad>",
"image_start_token": "<longcat_img_start>"
},
"pad_token": "<longcat_pad>",
"sp_model_kwargs": {},
"tokenizer_class": "TokenizersBackend",
"tool_parser_type": "longcat",
"unk_token": "<longcat_unk>"
}