Andrej Karpathy|11 - Building makemore Part 3: Activations and Gradients, BatchNorm
Andrej Karpathy, a founding member of OpenAI and former director of AI at Tesla, presents in-depth lectures on neural networks and large language models, often building them from scratch.Links:Andrej X: https://x.com/karpathyAndrej Karpathy web: https://karpathy.aiDiscord: https://discord.gg/3zy8kqD9CpEureka Labs: https://eurekalabs.ai/
00:00
00:00
Opis epizode nije dostupan.
Objavljeno29 lis 2026
Trajanje2:03:42
Epizode
11-
Building makemore Part 3: Activations and Gradients, BatchNorm
Slušano
29 lis 2026
10-
Building makemore Part 4: Becoming a Backprop Ninja
Slušano
29 lis 2026
9-
Building makemore Part 2: MLP
Slušano
29 lis 2026
8-
Building makemore Part 5: Building a WaveNet
Slušano
29 lis 2026
7-
The spelled-out intro to language modeling: building makemore
Slušano
29 lis 2026
6-
Let's build GPT: from scratch, in code, spelled out.
Slušano
29 lis 2026
5-
Stable diffusion dreams of psychedelic faces
Slušano
29 lis 2026
4-
[1hr Talk] Intro to Large Language Models
Slušano
29 lis 2026
3-
Stable diffusion dreams of steampunk brains
Slušano
30 lis 2026
2-
Stable diffusion dreams of "blueberry spaghetti" for one night
Slušano
30 lis 2026
1-
Stable diffusion dreams of steam punk neural networks