Transformer Explainer - A visualization tool to understand how modern LLMs work

preview_player
Показать описание
An interactive visualization tool that explains in a visual way how the Transformer architecture works.

Check out our upcoming live training to learn more about building with LLMs:

#ai #science #chatgpt #artificialintelligence
  1. Elvis Saravia
Рекомендации по теме
Transformer Explainer - 0:07:15

Transformer Explainer - A visualization tool to understand how modern LLMs work

Illustrated Guide to 0:15:01

Illustrated Guide to Transformers Neural Network: A step by step explanation

Transformer Explainer (1 0:01:00

Transformer Explainer (1 min Short) - Interactive Visualization for Transformers

Transformer Explainer (Full 0:02:01

Transformer Explainer (Full Video) - Interactive Visualization for Transformers

Transformers, explained: Understand 0:09:11

Transformers, explained: Understand the model behind GPT, BERT, and T5

Transformer Explainer: How 0:08:04

Transformer Explainer: How I visualize the Magic Behind Modern LLMs

Visualize the Transformers 0:05:54

Visualize the Transformers Multi-Head Attention in Action

Attention in transformers, 0:26:10

Attention in transformers, visually explained | Chapter 6, Deep Learning

An Image is 0:29:56

An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale (Paper Explained)

Transformer oversimplified | 0:39:17

Transformer oversimplified | A visual explanation of GPT using Poloclub | Large Language Model (LLM)

How Transformers work 0:02:58

How Transformers work explained animation

Transformer in Transformer: 0:06:01

Transformer in Transformer: Paper explained and visualized | TNT

But what is 0:27:14

But what is a GPT? Visual intro to transformers | Chapter 5, Deep Learning

Transformer Neural Networks 0:18:08

Transformer Neural Networks Derived from Scratch

Transformer Neural Networks 0:13:05

Transformer Neural Networks - EXPLAINED! (Attention is all you need)

Visual Guide to 0:15:25

Visual Guide to Transformer Neural Networks - (Episode 2) Multi-Head & Self-Attention

Attention is all 0:58:04

Attention is all you need (Transformer) - Model explanation (including math), Inference and Training

Vision Transformer Quick 0:16:51

Vision Transformer Quick Guide - Theory and Code in (almost) 15 min

DETR: End-to-End Object 0:40:57

DETR: End-to-End Object Detection with Transformers (Paper Explained)

Swin Transformer paper 0:11:10

Swin Transformer paper animated and explained

Vision Transformers (ViT) 0:30:27

Vision Transformers (ViT) Explained + Fine-tuning in Python

Vision Transformer Basics 0:30:49

Vision Transformer Basics

OpenAI's GPT-2 Explained 0:24:14

OpenAI's GPT-2 Explained | Visualizing Transformer Language Models | Generative Pre-Training | ...

Transformers explained | 0:19:48

Transformers explained | The architecture behind LLMs

Комментарии
Автор

Thanks for sharing our tool! In response to the question in the video, we're using a pre-trained GPT-2 tokenizer. Additionally, you can further explore how the probabilities are calculated by clicking on the "Probabilities" block, which will then expand to show the calculations for logits, exponents, and softmax!

gracekim
Автор

Thanks for sharing this!
And wow, what an incredible visualization/tool! 🤩

NimTheHuman
Автор

Thats pretty awesome! And they released a paper for it too!

reinerheiner
Автор

What a good resource! Thanks for sharing

MikeBirdTech
Автор

Am I the only one who thinks the 'query' and 'key' should switch? so that the key 'data' can attend to all the queries which makes sense cuz it is the first word in the sentence, or I'm wrong?

hsin-yusu