Atom2.7m / tokenizer_config.json
ucr-max's picture
Update Atom2.7m submission
2fd4f23 verified
Raw
History Blame Contribute Delete
332 Bytes
{
"additional_special_tokens": [
"<|endoftext|>"
],
"auto_map": {
"AutoTokenizer": [
"tokenization_atom.AtomTokenizer",
null
]
},
"bos_token": "<|bos|>",
"eos_token": "<|eos|>",
"model_max_length": 548,
"pad_token": "<|pad|>",
"tokenizer_class": "AtomTokenizer",
"unk_token": "<|unk|>"
}