🧠 Fine-Tuned GPT-2 Chatbot with Streamlit

This project fine-tunes the GPT-2 language model on a short, custom instruction-response dataset using LoRA (Low-Rank Adaptation) and deploys it using Streamlit Cloud.

---

🚀 Features

• Fine-tunes GPT-2 with a small dataset using Hugging Face + LoRA
• Interactive chatbot UI built with Streamlit
• Easily deployable via Streamlit Cloud

---

📂 Project Structure

my-llm-app/

• ├── app.py # Streamlit app
• ├── requirements.txt # Required Python libraries
• └── llm-finetuned/ # Fine-tuned model directory
• ├── config.json
• ├── pytorch_model.bin
• ├── tokenizer_config.json
• ├── tokenizer.json
• └── vocab.json

---

🛠️ How to Use

1. Train model in Colab and save llm-finetuned/
2. Push all files to GitHub
3. Deploy on : https://myllmapp-gpt2.streamlit.app/:
   • Repo: https://github.com/ifrazaib/MyllmApp
   • File: app.py

---

📥 Install Locally (Optional)

pip install -r requirements.txt
streamlit run app.py