LLM Workshop 2024

Important

🎯 DataHack Summit 2024 | 📆 August 10 2024 | 📍 Bengaluru, India

Explore this comprehensive repository on LLMs, covering everything from the basics of NLP to fine-tuning and even RLHF. If you find the resources helpful, consider giving it a star ⭐ to show your support and help others discover it.

---

Table of Contents

• Modules
• Prerequisites
• Environment Setup

---

Modules

Module 1: "Foundations of Generative AI and Language Models"

• Overview of Generative AI and the basics of language modeling.
• ⭐ Hands-On:
  • Getting Started: Text Representation
  • Language Modeling Basics and Text Generation using a basic LM.

Module 2: "Building Blocks of LLMs"

• Transformer Architectures: Detailed look into the Transformer architecture that powers modern LLMs.
• GPT Series of Models: Overview of the evolution of GPT models.
• Evaluation Metrics and Benchmarks: Methods to evaluate and benchmark LLM performance.
• ⭐ Hands-On: Training a mini Transformer model and experimenting with GPT-2 for text generation.

Module 3: "Advanced LLM Techniques"

• Training Process and Scaling Laws: Understand how LLMs are trained and the laws governing their scaling.
• PEFT: Learn Parameter-Efficient Fine-Tuning methods.
• LoRA: Introduction to Low-Rank Adaptation.
• Instruction Tuning: Techniques for fine-tuning models using instructions.
• RLHF: Reinforcement Learning from Human Feedback and its applications.
• ⭐ Hands-On:
  • Instruction Tuning: Text 2 SQL using LLaMA3.1
  • RLHF Hands-on: Sentiment aligment for generating movie reviews

Module 4: "Operationalizing LLMs"

• Prompt Engineering: Crafting effective prompts to get desired outputs.
• Prompt Hacking and Backdoors
• Vector Databases: Using vector databases for efficient data retrieval.
• RAGs: Techniques for retrieval-augmented generation.
• Beyond Prompting: Understanding Frameworks such as DSPY
• ⭐ Hands-On:
  • Implementing basic prompt engineering techniques and
  • Building a simple RAG system.
  • Handson with DSPY

Module 5: "The Future of LLMs and Next Steps"

• Next Steps: Speculative topics on future advancements.
• Beyond: Future possibilities and directions for LLM research.

---

Prerequisites

• Basics/hands-on experience of working with python
• Basic understanding of linear algebra and machine larning
• Basic understanding of Deep Neural Networks
• Basics/hands-on experience with pytorch
• Access to google-colab or similar python environment
• Access to chatGPT or Google-Bard (free access)

---

Environment Setup Instructions

Important

• Follow Step by Step for a quick setup. This should work as-is for Mac/Linux based systems.
• If you already have your own way of managing dependencies, checkout pyproject.toml for poetry or requirements.txt for pip based systems
• The requirements.txt file is generated using the command poetry export --without-hashes --format=requirements.txt > requirements.txt

• We will make use of :

  • pyenv for python version management
  • virtualenv for virtual environment management
  • poetry for dependency management

• Pyenv:

  • brew install pyenv or curl https://pyenv.run | bash

• VirtualEnv:

  • install:
    • brew install pyenv-virtualenv or
    • git clone https://github.com/pyenv/pyenv-virtualenv.git $(pyenv root)/plugins/pyenv-virtualenv
  • add this to your .rc file: eval "$(pyenv virtualenv-init -)"

• Poetry:

  • install:
    • curl -sSL https://install.python-poetry.org | python3 - or
    • or check here

• Setup:

  • Local Mac/Linux :If you have make available, simply execute: make setup otherwise:
  • RunPod or other Similar Providers: simply execute: make runpod_setup otherwise:
  • If you are using other ways of dependency management:
    • Python Environment: -pyenv install 3.11.9
      • pyenv virtualenv 3.11.9 datahack
      • cd <path to this repo clone>
      • pyenv activate datahack
      • poetry install <- Make sure pyproject.toml file is available in directory you execute this command OR
      • use the requirements.txt file for reference.
    • Setup nvm / node and install localtunnel