which tokenizer.model should be used?

#4
by aotsukiqx - opened

Dont see tokenizer.model in the repo, should I use llama tokenizer.model for run?

This is a GGML model so the tokenizer is built into the GGML file itself.

And you can't use it from plain transformers, if you want to load it from Python you should use llama-cpp-python or ctransformers.

Thank you very much for the response. I will try it out. I'm a beginner with LLM and just starting to learn. :)

aotsukiqx changed discussion status to closed

Sign up or log in to comment