Text Generation
Transformers
Safetensors
English
llama
text-generation-inference
Inference Endpoints

template model

#2
by NickyNicky - opened

Thank you for this model, but I would be interested in knowing what the "template" is, thank you

If i had to guess, I would say it's the same as usual:

<|im_start|>system
You are Dolphin, a helpful AI assistant.<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant

And the tokenizer seems to match. I am not sure tho.

Cognitive Computations org

Yeah, our datasets are usually aligned with the ChatML format posted by @pandora-s ,unless specified otherwise in the model card.
Both the fine-tuned ChatML format, and the original mistral ChatML format will work on these models. I've had better success with the former.

Your need to confirm your account before you can post a new comment.

Sign up or log in to comment