At a Glance

Llama 3 is a state-of-the-art, open-source LLM that outperformed GPT-3.5, the model behind the free version of ChatGPT, on a variety of benchmarks. To get the most out of Llama 3, a special prompt format should be used. This project provides instructions on the optimal way to interact with Llama 3 to ensure you receive the best possible responses.

Why LLAMA 3?

Llama 3 is an advanced open-source large language model (LLM). It’s a significant upgrade from the Llama 2 model, setting a new standard for LLMs of its size. For instance, on a human evaluation set, Llama 3 beat the following models:*

Moreover, Llama 3 has performed exceptionally well on the following popular benchmarks:

This project will teach you how to work with Llama 3. You'll learn how to give it instructions (called "prompts") that maximize your chances of obtaining high-quality responses. We'll use examples and a step-by-step exercise to ensure you understand everything.

Objectives:

Learn the right way to structure prompts for Llama 3.
Practice using Llama 3 with hands-on exercises with the IBM's watsonx.ai LLM interface.
Use LangChain to construct custom prompt templates.

* https://ai.meta.com/blog/meta-llama-3/

What You'll Need:

All you need is a browser and general knowledge of Python.
Access to Llama 3 LLM hosted on watsonx.ai is provided free of charge.

Offered By: IBMSkillsNetwork

Write Llama 3 prompts like a pro

At a Glance

Why LLAMA 3?

Objectives:

What You'll Need: