tokenizer_config.json

{
  "add_prefix_space": false,
  "additional_special_tokens": [
    "<fim_prefix>",
    "<fim_middle>",
    "<fim_suffix>",
    "<fim_eot>",
    "<fim_pad>",
    "[INST]",
    "[/INST]",
    "<extra_id_2>",
    "<extra_id_3>",
    "<extra_id_4>",
    "<extra_id_5>",
    "<extra_id_6>",
    "<extra_id_7>",
    "<extra_id_8>",
    "<extra_id_9>",
    "<extra_id_10>"
  ],
  "bos_token": "<|endoftext|>",
  "clean_up_tokenization_spaces": false,
  "eos_token": "<|endoftext|>",
  "model_max_length": 8192,
  "tokenizer_class": "GPTNeoXTokenizer",
  "unk_token": "<|endoftext|>"
}