alantandrea/wifi-threat-detection-assistant

WiFi Threat Detection Assistant

Fine-tuned model for WiFi threat detection system support.

Model Description

This model is fine-tuned from microsoft/Phi-3-mini-4k-instruct using QLoRA on documentation for a WiFi threat detection system. It can:

• Answer questions about system features and configuration
• Help distinguish between real attacks and false alarms
• Provide troubleshooting guidance
• Explain WiFi security concepts

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel

# Load base model
model = AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
model = PeftModel.from_pretrained(model, "alantandrea/wifi-threat-detection-assistant")
tokenizer = AutoTokenizer.from_pretrained("alantandrea/wifi-threat-detection-assistant")

# Generate response
prompt = "<|system|>\nYou are a WiFi security expert.<|end|>\n<|user|>\nWhat is a deauth attack?<|end|>\n<|assistant|>\n"
inputs = tokenizer(prompt, return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(outputs[0]))

Training Details

• Base model: microsoft/Phi-3-mini-4k-instruct
• Method: QLoRA (4-bit quantization + LoRA)
• Hardware: NVIDIA RTX 4090