---
license: mit
language:
- en
tags:
- IFT
---
# **Introduction**

This model originate from "LLaMA 2-7b" we trained only response part with the "Alpaca-GPT-4" dataset, utilizing LoRA (Low-Rank Adaptation) training. The weights from LoRA are merged into the model.

## Details

### Used Datasets
- vicgalle/alpaca-gpt4