|
|
--- |
|
|
license: other |
|
|
license_name: upstage-solar-license |
|
|
--- |
|
|
|
|
|
Upstage `solar-pro3` tokenizer |
|
|
- Vocab size: 196,608 |
|
|
- Language support: English, Korean, Japanese and more |
|
|
|
|
|
Please use this tokenizer for tokenizing inputs for the Upstage [solar-pro3](https://developers.upstage.ai/docs/apis/chat) model. |
|
|
|
|
|
You can load it with the transformers library like this: |
|
|
|
|
|
```python |
|
|
from transformers import AutoTokenizer |
|
|
|
|
|
tokenizer = AutoTokenizer.from_pretrained("upstage/solar-pro3-tokenizer") |
|
|
text = "Hi, how are you?" |
|
|
enc = tokenizer.encode(text) |
|
|
print("Encoded input:") |
|
|
print(enc) |
|
|
|
|
|
inv_vocab = {v: k for k, v in tokenizer.get_vocab().items()} |
|
|
tokens = [inv_vocab[token_id] for token_id in enc] |
|
|
print("Tokens:") |
|
|
print(tokens) |
|
|
|
|
|
number_of_tokens = len(enc) |
|
|
print("Number of tokens:", number_of_tokens) |
|
|
``` |
|
|
|
