Instructions to use nopenet/nope-edge with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use nopenet/nope-edge with Transformers:

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

pipe = pipeline("text-generation", model="nopenet/nope-edge")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("nopenet/nope-edge")
model = AutoModelForCausalLM.from_pretrained("nopenet/nope-edge")
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 nopenet/nope-edge with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/nopenet/nope-edge

SGLang

How to use nopenet/nope-edge 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 "nopenet/nope-edge" \
    --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": "nopenet/nope-edge",
		"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 "nopenet/nope-edge" \
        --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": "nopenet/nope-edge",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use nopenet/nope-edge with Docker Model Runner:
```
docker model run hf.co/nopenet/nope-edge
```

False positives when using triggering words in safe sentences

by Flexan - opened Feb 23

Discussion

Flexan

Feb 23

This happens using both NOPE Edge, NOPE Edge Mini, and whatever version is running on your website (images above).

My guess is that the LLM was trained using a dataset of normal/safe sentences and a dataset of harmful sentences, meaning it may have learned to associate certain words with danger regardless of context (e.g. "kms" or "suicide" as in the examples above) because they were only present in the harmful sentences and were never used in the safe ones.

I'm guessing your mission is to provide resources and prevent harmful LLM responses when a user sends a message related to them being in danger, while preventing false-positives from happening as much as possible. I'm assuming this because your website shows that platforms using your service would likely want to act on the signals returned by NOPE (including blocking the chat or rerouting the response), which can get rather annoying for the user if their messages are not harmful. Therefore the suggestion :]

jpad

NOPE org Feb 24

Thanks @Flexan for these FP reports. Building a new training set to tackle these right now. Agreed its overfitted on the heuristic of kms==bad.

Flexan

Feb 26

I re-evaluated the model and it seems to work well on those prompts now, kudos! Smart idea to add reflection to the model ^-^

Flexan changed discussion status to closed Feb 26

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