Instructions to use kyujinpy/Sakura-SOLAR-Instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use kyujinpy/Sakura-SOLAR-Instruct with Transformers:

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

pipe = pipeline("text-generation", model="kyujinpy/Sakura-SOLAR-Instruct")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("kyujinpy/Sakura-SOLAR-Instruct")
model = AutoModelForCausalLM.from_pretrained("kyujinpy/Sakura-SOLAR-Instruct")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use kyujinpy/Sakura-SOLAR-Instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "kyujinpy/Sakura-SOLAR-Instruct"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "kyujinpy/Sakura-SOLAR-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/kyujinpy/Sakura-SOLAR-Instruct

SGLang

How to use kyujinpy/Sakura-SOLAR-Instruct 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 "kyujinpy/Sakura-SOLAR-Instruct" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "kyujinpy/Sakura-SOLAR-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "kyujinpy/Sakura-SOLAR-Instruct" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "kyujinpy/Sakura-SOLAR-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use kyujinpy/Sakura-SOLAR-Instruct with Docker Model Runner:
```
docker model run hf.co/kyujinpy/Sakura-SOLAR-Instruct
```

Data Contamination

by devasheeshG - opened Dec 28, 2023

Discussion

devasheeshG

Dec 28, 2023

bro, are you sure there is not any data leak from the dataset's test set. Have you checked for contamination?

devasheeshG changed discussion title from Data Leak to Data Contamination Dec 28, 2023

kyujinpy

Owner Dec 29, 2023

•

edited Dec 29, 2023

Hello! Thank you for your comment.

I just applied mergekit method(slerp) for my Sakura-SOLAR-Instruct model.
(-> It means, I don't use any dataset about Sakura-SOLAR-Instruct!!)

So if my model is contaminated, the contamination is in the base models that I combined.

And also, you check all of information through Sakura-SOLAR-github.

Is it clear? Thanks!

kyujinpy changed discussion status to closed Dec 29, 2023

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