Picture by Creator
Giant language fashions (LLMs) like GPT-4 are quickly reworking the world and the sector of information science. In simply the previous few years, capabilities that after appeared like science fiction are actually turning into a actuality by LLMs.
The Generative AI with Large Language Models: Hands-On Training will introduce you to the deep studying breakthroughs powering this revolution, with a deal with transformer architectures. Extra importantly, you’ll straight expertise the unbelievable breadth of capabilities the newest LLMs like GPT-4 can ship.
You’ll learn the way LLMs are essentially altering the sport for growing machine studying fashions and commercially profitable information merchandise. You will note firsthand how they will speed up the artistic capacities of information scientists whereas propelling them towards turning into refined information product managers.
By hands-on code demonstrations leveraging Hugging Face and PyTorch Lightning, this coaching will cowl the total lifecycle of working with LLMs. From environment friendly coaching strategies to optimized deployment in manufacturing, you’ll study straight relevant expertise for unlocking the facility of LLMs.
By the top of this action-packed session, you should have each a foundational understanding of LLMs and sensible expertise leveraging GPT-4.
Picture from Coaching
The coaching has 4 brief modules that introduce you to Giant Language Fashions and train you to coach your personal giant language mannequin and deploy it to the server. Other than that, you’ll study concerning the industrial worth that comes with LLMs.
1. Introduction to Giant Language Fashions (LLMs)
- A Temporary Historical past of Pure Language Processing
- Transformers
- Subword Tokenization
- Autoregressive vs. Autoencoding Fashions
- ELMo, BERT, and T5
- The GPT (Generative Pre-trained Transformer) Household
- LLM Utility Areas
2. The Breadth of LLM Capabilities
- LLM Playgrounds
- Staggering GPT-Household progress
- Key Updates with GPT-4
- Calling OpenAI APIs, together with GPT-4
3. Coaching and Deploying LLMs
- {Hardware} Acceleration (CPU, GPU, TPU, IPU, AWS chips)
- The Hugging Face Transformers Library
- Finest Practices for Environment friendly LLM Coaching
- Parameter-efficient fine-tuning (PEFT) with low-rank adaptation (LoRA)
- Open-Supply Pre-Educated LLMs
- LLM Coaching with PyTorch Lightning
- Multi-GPU Coaching
- LLM Deployment Concerns
- Monitoring LLMs in Manufacturing
4. Getting Business Worth from LLMs
- Supporting ML with LLMs
- Duties that may be Automated
- Duties that may be Augmented
- Finest Practices for Profitable A.I. Groups and Initiatives
- What’s Subsequent for A.I.
The coaching consists of hyperlinks to exterior sources corresponding to supply code, presentation slides, and a Google Colab pocket book. These sources make it interactive and helpful for engineers and information scientists who’re implementing Generative AI into their workspace.
Picture from Coaching
Here’s a listing of important sources wanted to construct and deploy your personal LLM mannequin utilizing Huggingface and Pytorch Lighting:
Uncover the key to success in simply 2 hours! Do not wait any longer!
Abid Ali Awan (@1abidaliawan) is a licensed information scientist skilled who loves constructing machine studying fashions. At the moment, he’s specializing in content material creation and writing technical blogs on machine studying and information science applied sciences. Abid holds a Grasp’s diploma in Expertise Administration and a bachelor’s diploma in Telecommunication Engineering. His imaginative and prescient is to construct an AI product utilizing a graph neural community for college students scuffling with psychological sickness.
