codesage-large-v2/tokenizer_config.json
2024-12-03 04:48:20 +00:00

35 lines
798 B
JSON

{
"add_prefix_space": false,
"additional_special_tokens": [
"<|endoftext|>",
"<fim_prefix>",
"<fim_middle>",
"<fim_suffix>",
"<fim_pad>",
"<filename>",
"<gh_stars>",
"<issue_start>",
"<issue_comment>",
"<issue_closed>",
"<jupyter_start>",
"<jupyter_text>",
"<jupyter_code>",
"<jupyter_output>",
"<empty_output>",
"<commit_before>",
"<commit_msg>",
"<commit_after>",
"<reponame>"
],
"bos_token": "<|endoftext|>",
"eos_token": "<|endoftext|>",
"add_eos_token": true,
"model_max_length": 1000000000000000019884624838656,
"unk_token": "<|endoftext|>",
"vocab_size": 49152,
"tokenizer_class": "CodeSageTokenizer",
"auto_map": {
"AutoTokenizer": ["tokenization_codesage.CodeSageTokenizer", null]
}
}