🤗 LLM Parameter Visualization App

This interactive application helps users visualize and understand the effects of key parameters (temperature, top_p, and top_k) on text generation using language models from Hugging Face.

🌟 Features

• Interactive Parameter Control: Adjust temperature, top_p, and top_k in real-time
• Model Selection: Use any text generation model from Hugging Face
• Real-Time Text Generation: See how parameter changes affect generated text
• Save & Load Settings: Store your preferred configurations for later use
• Detailed Parameter Explanations: Learn what each parameter does and how it affects generation

📋 Requirements

• Python 3.10+
• PyTorch
• Transformers
• Gradio
• Requests

🚀 Installation

1. Clone this repository:

   git clone https://github.com/krishna-f22/llm-parameter-visualization.git
   cd llm-parameter-visualization

2. Create a virtual environment using Python 3.10:

   python3.10 -m venv .venv

3. Activate the virtual environment:

   • On macOS/Linux:

     source .venv/bin/activate

   • On Windows:

     .venv\Scripts\activate

4. Install the required dependencies:

   pip install -r requirements.txt

💻 Usage

1. Run the application:

   python app.py

2. Open your web browser and navigate to http://127.0.0.1:7860

3. Enter a model name from Hugging Face, adjust the parameters, and see the generated text change in real-time.

🔍 Understanding the Parameters

Temperature (0.1 - 2.0)

Temperature controls the randomness of the model's predictions.

• Low temperature (e.g., 0.1-0.3): More deterministic and focused outputs. The model will more consistently choose the highest probability tokens.
• Medium temperature (e.g., 0.7-1.0): Balanced between deterministic and creative outputs.
• High temperature (e.g., 1.5-2.0): More random and diverse outputs. The model is more likely to choose lower probability tokens.

Top-p / Nucleus Sampling (0.0 - 1.0)

Top-p sampling (also called nucleus sampling) sets a probability threshold and only considers tokens whose cumulative probability exceeds this threshold.

• Low top-p (e.g., 0.1-0.3): Only the most likely tokens are considered, resulting in more focused outputs.
• Medium top-p (e.g., 0.5-0.7): Balanced between focused and diverse outputs.
• High top-p (e.g., 0.9-1.0): Most tokens are considered, allowing for more diverse outputs.

Top-k (1 - 100)

Top-k limits the set of tokens the model considers to the k highest probability tokens.

• Low top-k (e.g., 1-10): Very restrictive, considering only the most likely tokens, which increases coherence but limits creativity.
• Medium top-k (e.g., 20-50): Balanced between coherence and creativity.
• High top-k (e.g., 80-100): More tokens are considered, allowing for greater diversity in the output.

💾 Saving and Loading Settings

Save Settings

1. Adjust parameters to your liking
2. Click "Save Current Settings" to store them locally
3. Use "Export Settings" to download a configuration file

Load Settings

1. Click "Browse" in the "Import Settings" section
2. Select a previously exported settings file
3. Click "Load Imported Settings"

📁 Project Structure

llm-parameter-visualization/
├── app.py             # Main application file
├── README.md          # This documentation
└── locally_stored_model/  # Directory for saved settings and models
    ├── settings.json  # Saved parameter settings
    └── selected_model.txt  # Last used model name

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

1. Fork the repository
2. Create a feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add some amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgements

• Hugging Face for their amazing transformers library
• Gradio for the interactive UI framework
• F22Labs for their hardware and immense opportunities.