HuggingFace Token

[AfamO]

Describe the bug Hugging Face is asking for token to access and run Llam2 model

To Reproduce Steps to reproduce the behavior:

1. Import Client from llm_vm
2. Try performing completion request using llam2 as big_model.
3. Then run the python code . See the attached screen shot.
4. See error. See the attached screen shot

Expected behavior Before invoking the model, Hugging Face expects a token having permission to the repo to be passed either via huggingface-cli loginor by passingtoken=`.

Screenshots

Desktop (please complete the following information):

• OS: Windows
• Browser [e.g. chrome, safari]
• Version The latest build. Just cloned from the main repo about 3 hours ago.

Smartphone (please complete the following information):

• Device: [e.g. iPhone6]
• OS: [e.g. iOS8.1]
• Browser [e.g. stock browser, safari]
• Version [e.g. 22]

Additional context Add any other context about the problem here.

[wansatya]

You can manage the HF login using from huggingface_hub import login

import sys
import os
# Import our client
from llm_vm.client import Client
# Import huggingface hub login
from huggingface_hub import login

# Login with HF API token
login(
    os.getenv("LLM_VM_HUGGINGFACEHUB_API_TOKEN")
)

# Instantiate the client specifying which LLM you want to use
client = Client(
    big_model='llama2'
)

# Put in to your prompt and go!
response = client.complete(
    context='',
    prompt = 'What is anarchy?'
)
print(
    response, file=sys.stderr
) # Anarchy is a political system in which the state is abolished and the people are free...

[AfamO]

It's alright. The above step, along with extra efforts to request for repo access from both Meta and Hugging Face websites, worked