[YouTube Lecture Summary] Andrej Karpathy - Deep Dive into LLMs like ChatGPT

Introduction

Pre-Training

Step 1: Download and preprocess the internet

Step 2: Tokenization

Step 3: Neural network training

Step 4: Inference

Base model

Post-Training: Supervised Finetuning

Conversations

Hallucinations

Knowledge of Self

Models need tokens to think

Things the model cannot do well

Post-Training: Reinforcement Learning

Reinforcement learning

DeepSeek-R1

AlphaGo

Reinforcement learning from human feedback (RLHF)

Preview of things to come

Keeping track of LLMs

Where to find LLMs

Where to find LLMs

🌍 Where to find LLM (Large Language Model)? 🔎🤖

If you want to leverage and experiment with LLM, you can find it on these platforms! 🚀


1️⃣ Commercial (Private) LLMs 🔐

Popular large-scale AI models can be accessed directly from each company's official website .

Representative LLM provider sites:

They are cloud-based and can be accessed directly from your website!


2️⃣ Open Source (Open Weights) LLMs 🌍🛠

🔓 Open weight models that are free to download and use can be found on multiple platforms.

Open source LLM provider sites:

  • Together AI → Various open source LLMs can be run 💡

  • Hugging Face → Provides numerous open source models

  • Hyperbolic → Llama 3.1 Base model provided

🛠 Inference(Inference) platform allows you to directly select and test multiple models!


3️⃣ Locally executable LLMs 💻

You can also run LLM directly on your computer!
Especially if you use the lightweight or low-precision models, it can even run on your personal PC 🎯

How to run local LLM:
1️⃣ LM StudioDownload link

  • 💻 Run AI models directly locally

  • 📌 Supports Mac and Windows

  • The UI/UX is a bit difficult, but once you get used to it, it's a powerful tool.

  • You can choose and run various models.

2️⃣ OllamaDownload link

  • 🎯 Run AI models locally with simple commands

  • Supports the latest models including Llama 3 and DeepSeek

  • Performance is especially good on Mac

💡 Local running tips:

  • Use small model (lightweight version) → avoid memory shortage

  • Low Precision settings (FP8, INT4, etc.) → Runs on smaller PCs too