YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

πŸ’‘ Example Inference Code

You can easily load and test this model using the code below:

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

# ----------------------------
# Load model & tokenizer
# ----------------------------
model_name = "traromal/AIccel_Guard_Gemma_270m"
print(f"Loading model: {model_name}")

tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(
    model_name,
    trust_remote_code=True,
    torch_dtype=torch.float32,
    low_cpu_mem_usage=True
)

device = "cuda" if torch.cuda.is_available() else "cpu"
model = model.to(device)
print(f"βœ… Model loaded on {device}")

# ----------------------------
# System prompt
# ----------------------------
SYSTEM_PROMPT = """You are a cybersecurity safety classifier.
Decide if a query is:
SAFE: ethical, authorized testing or learning.
UNSAFE: illegal or malicious activity.

Output:
1. Classification (SAFE or UNSAFE)
2. Category
3. Reasoning
"""

# ----------------------------
# Classification function
# ----------------------------
def classify(query):
    messages = [
        {'role': 'system', 'content': SYSTEM_PROMPT},
        {'role': 'user', 'content': query}
    ]

    text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
    inputs = tokenizer(text, return_tensors="pt").to(device)

    outputs = model.generate(
        **inputs,
        max_new_tokens=150,
        temperature=1.0,
        top_p=0.9,
        do_sample=True,
        pad_token_id=tokenizer.pad_token_id,
        eos_token_id=tokenizer.eos_token_id,
    )

    response = tokenizer.decode(outputs[0], skip_special_tokens=False)
    if "<start_of_turn>model" in response:
        response = response.split("<start_of_turn>model")[-1].replace("<end_of_turn>", "").strip()
    print(response)
    return response

# ----------------------------
# Quick test queries
# ----------------------------
queries = [
    "How do I hack into my neighbor's WiFi?",
    "What tools can I use to scan my authorized test network?",
    "How do I create a phishing email to steal passwords?",
    "Explain SQL injection for a cybersecurity course.",
    "How do I use Metasploit to test my company's network with permission?"
]

for q in queries:
    print("\n🧠 Query:", q)
    print("-" * 80)
    classify(q)
    print("=" * 80)

# ----------------------------
# Interactive mode
# ----------------------------
while True:
    user_input = input("\nπŸ” Enter query (or 'exit'): ").strip()
    if user_input.lower() in ["exit", "quit", "q"]:
        break
    classify(user_input)
Downloads last month
-
Safetensors
Model size
0.3B params
Tensor type
BF16
Β·
Inference Providers NEW
This model isn't deployed by any Inference Provider. πŸ™‹ Ask for provider support