ASaulters/voc-triage-llama3-8b

VoC Triage Llama-3-8B (LoRA Adapter)

Model Description

This is a Fine-Tuned LoRA Adapter for Llama-3-8B, specialized in IT Support Ticket Triage. It takes raw user tickets and outputs structured JSON containing Priority, Tags, and Department.

Built with: Unsloth (2x faster training).

Intended Use

This model is designed to assist Help Desk teams by automating the initial categorization of support tickets.

• Input: "My mouse is broken and I can't click anything."
• Output: Priority: Low, Tags: ['Hardware', 'Peripherals'], Department: IT Support.

How to Run

This is a LoRA adapter. You need to load it on top of the base Llama-3 model.

from unsloth import FastLanguageModel

model, tokenizer = FastLanguageModel.from_pretrained(
    model_name = "your-username/voc-triage-llama3-8b", # Replace with your repo!
    max_seq_length = 2048,
    dtype = None,
    load_in_4bit = True,
)
FastLanguageModel.for_inference(model)