Instructions to use princeton-nlp/Sheared-LLaMA-1.3B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use princeton-nlp/Sheared-LLaMA-1.3B with Transformers:

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

pipe = pipeline("text-generation", model="princeton-nlp/Sheared-LLaMA-1.3B")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("princeton-nlp/Sheared-LLaMA-1.3B")
model = AutoModelForCausalLM.from_pretrained("princeton-nlp/Sheared-LLaMA-1.3B", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use princeton-nlp/Sheared-LLaMA-1.3B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "princeton-nlp/Sheared-LLaMA-1.3B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "princeton-nlp/Sheared-LLaMA-1.3B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/princeton-nlp/Sheared-LLaMA-1.3B

SGLang

How to use princeton-nlp/Sheared-LLaMA-1.3B 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 "princeton-nlp/Sheared-LLaMA-1.3B" \
    --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": "princeton-nlp/Sheared-LLaMA-1.3B",
		"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 "princeton-nlp/Sheared-LLaMA-1.3B" \
        --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": "princeton-nlp/Sheared-LLaMA-1.3B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use princeton-nlp/Sheared-LLaMA-1.3B with Docker Model Runner:
```
docker model run hf.co/princeton-nlp/Sheared-LLaMA-1.3B
```

Great work + Excellent model

by doberst - opened Oct 23, 2023

Discussion

doberst

Oct 23, 2023

Very interested in your research on pruning and dynamic batch training and to see where it evolves. We wanted to share with you that we are seeing some of the best RAG instruct fine-tuning (for a small model) built on top of the Sheared-LLama-1.3B, in particular, and would welcome you to check it out (llmware/bling-sheared-llama-1.3-0.1) - we just posted the RAG finetuned model and will be publishing some benchmark "RAG-instruct" test evaluations in the next couple of weeks. Would look forward to chances to collaborate in the future.

doberst changed discussion status to closed Oct 23, 2023

doberst changed discussion status to open Oct 23, 2023

princeton-nlp

Owner Oct 26, 2023

Thanks for your interest in other work!!!!

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