babylm-2024-strict / tokenizer_config.json
davda54's picture
initial upload
7f321a9 verified
raw
history blame
No virus
222 Bytes
{
"tokenizer_class": "PreTrainedTokenizerFast",
"bos_token": "␂",
"eos_token": "␃",
"unk_token": "␦",
"sep_token": "␃",
"pad_token": "␒",
"cls_token": "␂",
"mask_token": "β₯"
}