Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Instruct Template #24

Open
riccardomusmeci opened this issue Apr 30, 2024 · 3 comments
Open

Instruct Template #24

riccardomusmeci opened this issue Apr 30, 2024 · 3 comments

Comments

@riccardomusmeci
Copy link

Hi there, looking at OpenELM instruct template what I understood is that the template is something like this:

<|system|>
You are a helpful AI assistant.
<|user|>
How to explain Internet for a medieval knight?
<|assistant|>

Is that right? With MLX the model is not properly working on instruct mode.

@louis845
Copy link

louis845 commented May 3, 2024

Yes, this would be really useful. I am trying to use the OpenELM instruct models but I can't find out what the instruct template is.

@purefall
Copy link

purefall commented Jul 9, 2024

Is there any update regarding the chat template and tokenizer used for instruction tuning?

Especially when using OpenELM 1.1B Instruct models from Huggingface, I only get "<<<<<<" as the output.

I used tokenizer of "meta-llama/Llama-2-7b-hf" and original template from Llama2, but the output is garbage.... It would be great if you can clarify

@SongTonyLi
Copy link

SongTonyLi commented Sep 20, 2024

I'm also confused here. If indeed we need to use meta-llama/Llama-2-7b-hf chat template, so why does the source code also include another different chat template?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants