Re:Why is GPT-3 15.77x more expensive for certain languages?
by og_kalu on Monday April 10, 2023 @02:52PM
(#35515908)
Data used to train the tokenizer is entirely separate from data training the LLM.
The tokenizer used to train GPT-3 was old, inefficient and targeted at tokenizing English. That's pretty much all there is to it. It's possible to train a tokenizer that is more efficient and more including of other languages.
GPT-4's tokenizer is already far more efficient though still weighted to English.
The tokenizer used to train GPT-3 was old, inefficient and targeted at tokenizing English. That's pretty much all there is to it. It's possible to train a tokenizer that is more efficient and more including of other languages.
GPT-4's tokenizer is already far more efficient though still weighted to English.
You can test it here https://tiktokenizer.vercel.app/