embeddinggemma-asm-16384 / tokenizer_config.json
de-francophones's picture
Trimmed EmbeddingGemma 300M for Assamese
b517f0c verified
Raw
History Blame Contribute Delete
323 Bytes
{
"backend": "tokenizers",
"bos_token": "<bos>",
"clean_up_tokenization_spaces": false,
"eos_token": "<eos>",
"is_local": true,
"mask_token": "<mask>",
"model_max_length": 2048,
"pad_token": "<pad>",
"padding_side": "right",
"tokenizer_class": "TokenizersBackend",
"unk_token": "<unk>"
}