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

documentaion on fastchat model support #200

Open
wants to merge 8 commits into
base: main
Choose a base branch
from
Open

documentaion on fastchat model support #200

wants to merge 8 commits into from

Conversation

bsabri
Copy link
Contributor

@bsabri bsabri commented Sep 5, 2023

This could be merged with tutorial.md

@bsabri bsabri requested a review from fdalvi September 5, 2023 08:13
@bsabri bsabri changed the title doc/fastchat documentaion on fastchat model support Sep 5, 2023
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@MaramHasanain @fdalvi,
I think we can add this under tutorial. What do you think?

Wait until the process finishes loading the model and you see "Uvicorn running on ...". The model worker will register itself to the controller .


#### Run the asset from LLMeBench
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@bsabri Can we add an asset specifically designed for gpt2? I assume the gpt2 adapter does not really care about the "system" role?

[
        {
            "role": "system",
            "content": "You are an expert translator specialized in translating texts from Arabic to English. You are concise as you only output the translation of the text without any illustrations or extra details",
        },
        {
            "role": "user",
            "content": f"Translate the following text to English.\nText: {input_sample}\nTranslation: ",
        },
    ]

We can either add the asset properly or just show the prompt function here in the doc (perhaps for a task that gpt2 is actually good/decent at)

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

Successfully merging this pull request may close these issues.

3 participants