Instructions to use microsoft/Llama2-7b-WhoIsHarryPotter with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use microsoft/Llama2-7b-WhoIsHarryPotter with Transformers:

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

pipe = pipeline("text-generation", model="microsoft/Llama2-7b-WhoIsHarryPotter")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("microsoft/Llama2-7b-WhoIsHarryPotter")
model = AutoModelForCausalLM.from_pretrained("microsoft/Llama2-7b-WhoIsHarryPotter")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use microsoft/Llama2-7b-WhoIsHarryPotter with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "microsoft/Llama2-7b-WhoIsHarryPotter"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "microsoft/Llama2-7b-WhoIsHarryPotter",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/microsoft/Llama2-7b-WhoIsHarryPotter

SGLang

How to use microsoft/Llama2-7b-WhoIsHarryPotter 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 "microsoft/Llama2-7b-WhoIsHarryPotter" \
    --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": "microsoft/Llama2-7b-WhoIsHarryPotter",
		"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 "microsoft/Llama2-7b-WhoIsHarryPotter" \
        --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": "microsoft/Llama2-7b-WhoIsHarryPotter",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use microsoft/Llama2-7b-WhoIsHarryPotter with Docker Model Runner:
```
docker model run hf.co/microsoft/Llama2-7b-WhoIsHarryPotter
```

Releasing Evaluation Prompts?

by kevinyaoresearch - opened Dec 20, 2023

Discussion

kevinyaoresearch

Dec 20, 2023

I can't find the Github link claimed in the paper. Do the authors plan to release the code?

In addition, is there any plan to release the (completion-based and token-probability-based) prompts used in the evaluation? The description in Appendix 6.2.1 isn't clear to me. How is the "300-word long chunk" used in Figure 11? What's meant by " followed by a list of hand-curated examples"? Do you need to manually write examples to show how to create prompts based on the 300-word original text chunk?

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