1. Silicon Valley developer (Waddy): An evaluation harness for evaluating LLMs
Are the LLMs you use really working properly?? How can you measure whether they are performing well in response to changes in prompts and internal tools?? You’ll be able to explore that process together.
2. Silicon Valley Developer (Waddy): Building Embeddings and a Mini GPT from Scratch
The concept of embeddings is incredibly difficult, and how exactly is GPT made?? This is a process where you can train a model based on actual Python, even send it prompts, and build your own model.
Do you know about an in-memory database called Valkey?? It is an in-memory database that is being compared with Redis, and various cloud providers such as AWS and Google now provide Valkey by default.
In line with the latest trends, let’s take a look at how ultra-high performance was measured in the course on Valkey!!