Daniel Han

Sorry on the issue - I will take a look!

Our conversational notebooks add `eos_token`s to llama-3 for eg: https://colab.research.google.com/drive/1XamvWYinY6FOSX9GLvnqSjjsNflxdhNc?usp=sharing All our notebooks on our Github page here: https://github.com/unslothai/unsloth?tab=readme-ov-file#-finetune-for-free add eos tokens

oh wait llama-3 base right hmm where are you all doing inference - ollama? llama.cpp?

@KillerShoaib @gamercoder153 @mxtsai Apologies I just fixed it! No need to change code - I updated the tokenizer configs so all should be fine now!

Wait does Apple M1 Pro have a NVIDIA GPU?

Yes - I haven't announced it yet, but you can use: ```python from unsloth import add_new_tokens add_new_tokens(model, tokenizer, new_tokens = ["", "") ``` Do this before `get_peft_model`

oh theyre all special tokens! just use `add_new_tokens` for all of them

Hmmm sadly multi GPU issues are not a top priority, since Unsloth's mission is to be the best single GPU library - ill see what I can do, but can't...

Ok that is weird unsure why its not working hmmm

Oh thanks! Interesting I'll take a lot