---
base_model: microsoft/Phi-3-mini-4k-instruct
tags:
  - lora
  - aviation
  - military
  - fine-tuned
  - phi-3
license: mit
---

# ✈️ Phi-3 Military Aviation Museum Guide

A fine-tuned version of [Phi-3 Mini](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct) trained on 2,500
military aviation Q&A pairs. Answers questions in the style of an enthusiastic aircraft museum guide.

## Usage

```python
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model = AutoModelForCausalLM.from_pretrained("michael10098/phi3-mini-q4_k_m_aviation_museum", torch_dtype=torch.float16, device_map="cuda")
tokenizer = AutoTokenizer.from_pretrained("michael10098/phi3-mini-q4_k_m_aviation_museum")

prompt = "<|system|>\nYou are an enthusiastic aviation museum guide.<|end|>\n<|user|>\nWhat is a B-29?<|end|>\n<|assistant|>\n"
inputs = tokenizer(prompt, return_tensors="pt").to("cuda")
output = model.generate(**inputs, max_new_tokens=200)
print(tokenizer.decode(output[0], skip_special_tokens=True))
```

## Training Details
- **Base model:** Phi-3 Mini 4k Instruct
- **Method:** QLoRA (4-bit quantization + LoRA)
- **Dataset:** 2,500 military aviation Q&A pairs
- **Hardware:** NVIDIA RTX 4060 (8GB VRAM)