Instructions to use NousResearch/Llama-2-7b-chat-hf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use NousResearch/Llama-2-7b-chat-hf with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="NousResearch/Llama-2-7b-chat-hf")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("NousResearch/Llama-2-7b-chat-hf")
model = AutoModelForCausalLM.from_pretrained("NousResearch/Llama-2-7b-chat-hf", device_map="auto")

Inference
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use NousResearch/Llama-2-7b-chat-hf with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "NousResearch/Llama-2-7b-chat-hf"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "NousResearch/Llama-2-7b-chat-hf",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/NousResearch/Llama-2-7b-chat-hf

SGLang

How to use NousResearch/Llama-2-7b-chat-hf with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "NousResearch/Llama-2-7b-chat-hf" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "NousResearch/Llama-2-7b-chat-hf",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "NousResearch/Llama-2-7b-chat-hf" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "NousResearch/Llama-2-7b-chat-hf",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use NousResearch/Llama-2-7b-chat-hf with Docker Model Runner:
```
docker model run hf.co/NousResearch/Llama-2-7b-chat-hf
```

Need dataset format for fine-tune

by deepak1197 - opened Oct 11, 2023

Discussion

deepak1197

Oct 11, 2023

Can I get the dataset in CSV format to fine-tune the model?"

NPap

Oct 11, 2023

•

edited Oct 11, 2023

If you are asking for the format of the prompts for fine-tuning, then from:
https://huggingface.co/blog/llama2

<s>[INST] <<SYS>>
{{ system_prompt }}
<</SYS>>

{{ user_message }} [/INST]

I don't see anything for the assistant response though, so maybe something like this(?)

<s>[INST] <<SYS>>
{{ system_prompt }}
<</SYS>>

{{ user_message }} [/INST]{{ assistant_response }}

That or with a \n inbetween

teknium

NousResearch org Oct 11, 2023

Can I get the dataset in CSV format to fine-tune the model?"

The dataset was never released by Meta. So, unfortunately, no

strangeloopcanon

Nov 15, 2023

Getting a weird error: "The model did not return a loss from the inputs, only the following keys: logits. " ... Any ideas on what I can do to fix this?

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment