LLMs

LLM stands for Large Language model which is an advanced AI trained on massive data that allows them to learn complex patterns and relationships between words and sentences and excels at processing, understanding, and generating human language.

Downsides of LLM

In patterns, LLM do not have real understanding of things like they do not get the context and is only able to answer if it has been trained on it or it can create answers totally on its own which is most probably incorrect.

Transformative model

Transformative model was first introduced by Vaswani et al in 2017 and later in 2018, Google developed BERT.

What is AI?

Imagine this, human babies don’t know anything immediately after birth, they grow up, see things in their surrounding, observe them, learn from them and ultimately implement them. Now, remove human babies from the picture and consider machines. We feed them data, train them and from that data they perform some task successfully. This is what Artificial Intelligence is.

What is GenAI?

Generative AI is a sub category of Artificial Intelligence where the AI is capable of generating new content which could be text, image, video, audio, etc.…

Copilot

It is an AI powered assistant which streamline with your workflow, it doesn’t just give out answers but rather tries to understand what you are working on and provides suggestions. You can also tag certain files and ask questions to either debug, explain or maybe for the next steps. Thus, it can help you boost your productivity. It can also adapt to your needs across different domains.

What is Pieces?

Pieces is your AI friend, it has long term memory which captures context from browsers and IDE and help you with your task. It supports multi LLM.

Context

It is an information that is given with a file on the basis of which questions can be asked, it can be also referred to as prompt. The better the context, the better the answers GenAI's can generate. Context size varies with LLMs. Context per conversation is on a conversation basis.

Prompt Engineering

Simply put, it is the process of designing prompts to obtain the required output or content. While designing a prompt, the following four things should be kept in mind.

4S of Prompt Engineering: