Instructions to use jsl5710/Shield-Gemma-3-270m-Full-FT-CE with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use jsl5710/Shield-Gemma-3-270m-Full-FT-CE with Transformers:

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

pipe = pipeline("text-generation", model="jsl5710/Shield-Gemma-3-270m-Full-FT-CE")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("jsl5710/Shield-Gemma-3-270m-Full-FT-CE")
model = AutoModelForCausalLM.from_pretrained("jsl5710/Shield-Gemma-3-270m-Full-FT-CE", device_map="auto")
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]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use jsl5710/Shield-Gemma-3-270m-Full-FT-CE with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "jsl5710/Shield-Gemma-3-270m-Full-FT-CE"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "jsl5710/Shield-Gemma-3-270m-Full-FT-CE",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/jsl5710/Shield-Gemma-3-270m-Full-FT-CE

SGLang

How to use jsl5710/Shield-Gemma-3-270m-Full-FT-CE 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 "jsl5710/Shield-Gemma-3-270m-Full-FT-CE" \
    --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": "jsl5710/Shield-Gemma-3-270m-Full-FT-CE",
		"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 "jsl5710/Shield-Gemma-3-270m-Full-FT-CE" \
        --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": "jsl5710/Shield-Gemma-3-270m-Full-FT-CE",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use jsl5710/Shield-Gemma-3-270m-Full-FT-CE with Docker Model Runner:
```
docker model run hf.co/jsl5710/Shield-Gemma-3-270m-Full-FT-CE
```

jsl5710 commited on Apr 9

Commit

132f18d

verified ·

1 Parent(s): 3375d8e

Add holdout test set results to model card

Browse files

Files changed (1) hide show

README.md +28 -0

README.md CHANGED Viewed

@@ -78,6 +78,34 @@ print(tokenizer.decode(outputs[0], skip_special_tokens=True))
 > Early stopping triggered when eval_loss did not improve for 3 consecutive evaluations.
 ## Training Setup
 - **Training objective:** Cross-Entropy (next-token prediction)

 > Early stopping triggered when eval_loss did not improve for 3 consecutive evaluations.
+## Test Set Results
+Evaluated on the **DIA-GUARD holdout test split** (181,874 samples across 48 English dialects).
+| Metric | Value |
+|--------|-------|
+| **Test Accuracy** | **0.9654** |
+| **Macro Precision** | 0.9676 |
+| **Macro Recall** | 0.9634 |
+| **Macro F1** | **0.9650** |
+| **Support** | 181,874 |
+### Per-class
+| Class | Precision | Recall | F1 | Support |
+|-------|-----------|--------|----|---------|
+| **safe** | 0.9844 | 0.9392 | 0.9613 | 83,140 |
+| **unsafe** | 0.9507 | 0.9875 | 0.9688 | 98,734 |
+### Confusion Matrix
+|             | Pred safe | Pred unsafe |
+|-------------|-----------|-------------|
+| **True safe** | 78,087 | 5,053 |
+| **True unsafe** | 1,234 | 97,500 |
+> Per-dialect breakdown available in `per_dialect.json` in the corresponding results folder.
 ## Training Setup
 - **Training objective:** Cross-Entropy (next-token prediction)