This week, we will delve into the fascinating world of computer science and programming using Scratch, a visual programming language designed for beginners. Through hands-on projects, you will grasp fundamental concepts such as variables, loops, and conditional statements, empowering you to create your own interactive applications.
1
The spelled-out intro to neural networks and backpropagation: building micrograd
Discover the fundamentals of neural networks and backpropagation as we build a simple yet effective neural network from scratch.
2
The spelled-out intro to language modeling: building makemore
Learn about language modeling and how to build a language model from scratch using simple yet powerful techniques.
3
Building makemore Part 2: MLP
Dive deeper into the makemore project by exploring the intricacies of Multi-Layer Perceptrons (MLPs) and their role in language modeling.
4
Building makemore Part 3: Activations & Gradients, BatchNorm
Explore advanced concepts in neural networks, including activation functions, gradients, and Batch Normalization in the makemore series.
5
Building makemore Part 4: Becoming a Backprop Ninja
Master the art of backpropagation in this detailed guide, enhancing your skills in training neural networks effectively.
6
Building makemore Part 5: Building a WaveNet
In this part, we construct a WaveNet model, diving into the complexities of temporal convolutions and their applications in sequence modeling.
7
Let's build GPT: from scratch, in code, spelled out.
A comprehensive guide to building GPT from the ground up, understanding each component and coding it step by step.
8
State of GPT | BRK216HFS
An overview of the current state of GPT technology, including recent advancements and future directions.
9
Let's build the GPT Tokenizer
Learn how to build a GPT tokenizer from scratch, a crucial component for processing and understanding text data in GPT models.
10
Let's reproduce GPT-2 (124M)
Step-by-step instructions to reproduce GPT-2 (124M) model, with insights into its architecture and training process.