File size: 278 Bytes
19d30ed
242a90c
 
01abd33
19d30ed
 
d81eb24
 
 
1
2
3
4
5
6
7
8
9
---
language:
- ko
library_name: transformers
---

based on phi-3 tokenizer, expanded 17291 tokens

Following [The Optimal Vocabulary Size Predictor](https://huggingface.co/spaces/sail/scaling-with-vocab-demo), I recommend using this tokenizer with 3-4B model such as phi-3-mini