README.md

metadata

license: llama3.2
language:
  - en
tags:
  - cybersecurity
  - security
  - log-analysis
  - threat-detection
  - windows
  - gguf
  - llama
base_model: meta-llama/Llama-3.2-3B-Instruct

🛡️ WinSentinel AI - Security Log Analysis Model

A fine-tuned Llama 3.2 3B model specialized in Windows security event log analysis and threat detection.

Features

• Threat Classification — Categorizes events as normal, suspicious, or malicious
• Severity Assessment — Rates threats from Low to Critical
• Intelligent Explanations — Provides context for why an event matters
• Action Recommendations — Suggests specific remediation steps
• Conversational — Responds naturally to general questions about itself and cybersecurity

Quick Start with Ollama

# Download both files: winsentinel-llama3.2-3b-f16.gguf and Modelfile
ollama create winsentinel -f Modelfile
ollama run winsentinel

Quick Start with LM Studio

1. Download winsentinel-llama3.2-3b-f16.gguf
2. Open LM Studio → My Models → Import
3. Select the GGUF file

Example

Input:

Timestamp: 2026-02-15 03:22:41
Source IP: 45.33.12.88
User: admin
Event Type: login_failed
Message: Multiple failed login attempts from external IP

Output:

Threat Category: suspicious
Severity: Medium
Explanation: This event is significant because it matches known behavior patterns for credential access.
Recommended Action: Monitor the system and review logs for anomalies. Consider implementing MFA.

Training Details

• Base Model: unsloth/Llama-3.2-3B-Instruct
• Method: QLoRA (4-bit quantization + LoRA adapters)
• Dataset: 28,000+ security event logs + conversational data
• Hardware: NVIDIA RTX 3050 6GB