Zoof Tokenizer
This is the custom Byte-Pair Encoding (BPE) tokenizer built for the Zoof-250M language model family. It was trained from scratch to efficiently handle English text and code.
Model Details
- Vocabulary Size: 49152
- Type: Byte-Pair Encoding (BPE)
- Language: English
- Intended Use: Tokenization for the Zoof model family.
Usage
You can load this tokenizer directly with the transformers library:
from transformers import PreTrainedTokenizerFast
tokenizer = PreTrainedTokenizerFast.from_pretrained("Jiraya/zoof-tokenizer")
text = "Hello, world!"
tokens = tokenizer.encode(text)
decoded = tokenizer.decode(tokens)
print(f"Tokens: {tokens}")
print(f"Decoded: {decoded}")