tokenizer model
#5
by
csabakecskemeti
- opened
possible to share the tokenizer model?
Gracias muy buen desarrollo y expandirlo a la comunidad..
Hi, we've released the tokenizer along with the model, which you can find under the Files and versions page on the Hugging Face Hub. Below is a helpful snippet to directly access the tokenizer.
>>> from transformers import AutoTokenizer
>>> tokenizer = AutoTokenizer.from_pretrained('JetBrains/Mellum-4b-base')
>>> tokenizer.encode('def main() -> None:')
[904, 2614, 403, 1295, 2374, 52]
>>> tokenizer.decode([904, 2614, 403, 1295, 2374, 52])
'def main() -> None:'
dustalov
changed discussion status to
closed