HaiVE’s Post

Transcript

Welcome everybody. So this is Rag dr.com. We have given a beautiful project categorization window where you can. Going to multiple projects and have. A lot of. Chats within each project so you know exactly where to look for when you're working on something. So we can also switch providers anytime you want. So we're going to grow, go drop now. A model we've chosen. Comma 38B because that has a lot more rate limit. If you don't have one, you can always Click to go sign up. You can choose whichever model of text to speech you want. I'm speech to text. And yeah. All good to go. No, we go ahead and upload a document. I'm going to upload this OG. Attention is all you need paper. It's all good. No, we can enable RAG. Let's see. And as soon as you say. The fact that you're going to use RAG, it automatically populates. A more optimized system prompt. And this can all be moved out of the way. And you can just start talking to it and use Grock. Can you summarize this document in 100 words? This is a MacBook Air and it clearly shows. The document presents the Transformer, a novel neural network architecture for sequence to sequence tasks such as machine translation. It replaces traditional recurrent neural networks aren't and ends with self attention mechanisms, allowing for parallelization and faster training. The Transformer achieves state-of-the-art results are. Now here's the best part you can change. Your AI server. Whenever you want. I'm going to choose Lemma 3.1. Let's see what. What is meant by attention in this paper? You can see the RAM usage. Creeping up. Because it's running locally now. One Lemma 3B to another. There we go. That's. Slightly more detailed than I expected. In this paper, attention refers to a mechanism used in the Transformer model for neural machine translation in traditional sequence. Right. So that does it.