README.md

---
title: SFT Phi2 QLoRA
emoji: 🦀
colorFrom: gray
colorTo: purple
sdk: gradio
sdk_version: 5.20.0
app_file: app.py
pinned: false
license: mit
short_description: Supervised Fine Tuned Phi model with QLoRA
---

In this project, open source Phi 2 model (base model) is trained on OpenAssistant dataset using QLoRA.
This has given the base model, ability to chat instead of randomly generating next token.

Project Repository: https://github.com/Shilpaj1994/SFT-with-QLoRA