tokenizer_config.json

{
    "add_bos_token": false,
    "add_eos_token": false,
    "additional_special_tokens": [
        "<|fim_begin|>",
        "<|fim_hole|>",
        "<|fim_end|>",
        "<|fim_pad|>",
        "<|repo_name|>",
        "<|file_sep|>"
    ],
    "bos_token": "<|startoftext|>",
    "clean_up_tokenization_spaces": false,
    "cls_token": "[CLS]",
    "eos_token": "<|endoftext|>",
    "gmask_token": "[gMASK]",
    "merges_file": null,
    "model_max_length": 1000000000000000019884624838656,
    "tokenizer_class": "PreTrainedTokenizerFast",
    "vocab_file": null,
    "pad_token": "<|endoftext|>",
    "fast_tokenizer": true
}