hyperakrep/Haru-4B-bucket / tokenizer_config.json
hyperakrep's picture
download
raw
496 Bytes
{
"add_prefix_space": false,
"backend": "tokenizers",
"bos_token": null,
"clean_up_tokenization_spaces": false,
"eos_token": "<|im_end|>",
"errors": "replace",
"model_max_length": 262144,
"pad_token": "<|endoftext|>",
"pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
"split_special_tokens": false,
"tokenizer_class": "Qwen2Tokenizer",
"unk_token": null
}

Xet Storage Details

Size:
496 Bytes
·
Xet hash:
d6566ed01ac9b2beaeb99c21de4c541f082adc5c828903e1482842fec195cd2c

Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.