tokenizer_config.json

{
  "add_prefix_space": false,
  "audio_bos_token": "<|audio_bos|>",
  "audio_eos_token": "<|audio_eos|>",
  "audio_token": "<|AUDIO|>",
  "auto_map": {
    "AutoProcessor": "processing_midashenglm.MiDashengLMProcessor"
  },
  "backend": "tokenizers",
  "bos_token": null,
  "clean_up_tokenization_spaces": false,
  "eos_token": "<|im_end|>",
  "errors": "replace",
  "extra_special_tokens": [
    "<|en|>",
    "<|kr|>",
    "<|de|>",
    "<|es|>",
    "<|fr|>",
    "<|hi|>",
    "<|uk|>",
    "<|th|>",
    "<|vi|>",
    "<|nl|>",
    "<|pt|>",
    "<|id|>",
    "<|ru|>",
    "<|it|>",
    "<|ar|>",
    "<|jp|>",
    "<|unknown|>"
  ],
  "image_token": "<|IMAGE|>",
  "is_local": false,
  "model_max_length": 32768,
  "model_specific_special_tokens": {
    "audio_bos_token": "<|audio_bos|>",
    "audio_eos_token": "<|audio_eos|>",
    "audio_token": "<|AUDIO|>",
    "image_token": "<|IMAGE|>",
    "video_token": "<|VIDEO|>",
    "vision_bos_token": "<|vision_bos|>",
    "vision_eos_token": "<|vision_eos|>"
  },
  "pad_token": "<|endoftext|>",
  "processor_class": "MiDashengLMProcessor",
  "split_special_tokens": false,
  "tokenizer_class": "TokenizersBackend",
  "unk_token": null,
  "video_token": "<|VIDEO|>",
  "vision_bos_token": "<|vision_bos|>",
  "vision_eos_token": "<|vision_eos|>"
}