arcade100k / README.md
jon-tow's picture
Update README.md
be6493b
|
raw
history blame
258 Bytes

Arcade100kTokenizer

BPE tokenizer based on tiktoken.cl100k_base.

from transformers import AutoTokenizer
tokenizer = AutoTokenizer.from_pretrained("stabilityai/arcade100k", trust_remote_code=True)
tokenizer("hello, world!", return_tensors='pt')