[GRPO Explained] DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models

#deepseek #llm #grpo GRPO is one of the core advancements used in Deepseek-R1, but was introduced already last year in this paper that uses a combinat...

Byte Latent Transformer: Patches Scale Better Than Tokens (Paper Explained)

#tokenization #llm #meta This paper does away with tokenization and creates an LLM architecture that operates on dynamically sized "patches" instead o...

The most painful plant in New Zealand

What Programming Font Should You Use?

I like programming. Fonts tried in this video 1. Menlo 2. Comic Shanns 3. Fira Code 4. JetBrains Mono 5. MonoLisa #benawad ---- Follow me online: http...

Let's reproduce GPT-2 (124M)

We reproduce the GPT-2 (124M) from scratch. This video covers the whole process: First we build the GPT-2 network, then we optimize its training to be...

I Turned 1-Star Toys into Military Nightmares

Go to http://crunchlabs.com/williamosman and claim your free kit today! Get your Open Sauce Tickets before they sell out!! https://opensauce.com/ticke...

Could you Survive a Blast from the Worlds Biggest Vortex Cannon?

Go to http://crunchlabs.com/backyardscientist and claim your free kit today! Also, bring your project to Open Sauce! http://opensauce.com Allens Video...

Let's build the GPT Tokenizer

The Tokenizer is a necessary and pervasive component of Large Language Models (LLMs), where it translates between strings and tokens (text chunks). To...

[1hr Talk] Intro to Large Language Models

This is a 1 hour general-audience introduction to Large Language Models: the core technical component behind systems like ChatGPT, Claude, and Bard. W...

GPT Function Calling: 5 Underrated Use Cases

OpenAI’s backend converting messy unstructured data to structured data via functions OpenAI’s “Function Calling” might be the most groundbreaking yet...

Deploy CoreML Models on the Server with Vapor

Get the benefits of Apple’s ML tools server-side. SwiftUI client showing image classification results Recently, at Sovrn , we had an AI Hackathon wher...

Full Stack Engineers Don’t Exist!

There’s a problem with those thousands of jobs available for Full Stack Engineers or Developers on LinkedIn, like a unicorn, that person… Continue rea...

Someone Paid $10,000 to Patent This

Thanks to Google Meet for sponsoring a portion of this video! ► Try Google Meet on Android or install it on your iOS device! https://goo.gle/3Rl9Opw O...

Let's build GPT: from scratch, in code, spelled out.

We build a Generatively Pretrained Transformer (GPT), following the paper "Attention is All You Need" and OpenAI's GPT-2 / GPT-3. We talk about connec...

Building makemore Part 5: Building a WaveNet

We take the 2-layer MLP from previous video and make it deeper with a tree-like structure, arriving at a convolutional neural network architecture sim...

Building makemore Part 2: MLP

We implement a multilayer perceptron (MLP) character-level language model. In this video we also introduce many basics of machine learning (e.g. model...

The spelled-out intro to language modeling: building makemore

We implement a bigram character-level language model, which we will further complexify in followup videos into a modern Transformer language model, li...