gemma-tokenizer / README.md
much1na's picture
Fix title typo
9134e8d verified
|
Raw History Blame Contribute Delete
436 Bytes
metadata
library_name: transformers
pipeline_tag: feature-extraction
tags:
  - tokenizer

gemma-tokenizer

The text tokenizer used in Google's Gemma models.

Usage

from transformers import AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained("much1na/gemma-tokenizer")

ids = tokenizer("hello world").input_ids
print(ids)  # [2, 23391, 1902, 1]
print(tokenizer.decode(ids, skip_special_tokens=True))  # hello world