ClipPhrase
RU

neural network

76
clips found
5
videos

Examples in video

YouTubeLex Fridman758K views · 2026-01-31
State of AI in 2026: LLMs, Coding, Scaling Laws, China, Agents, GPUs, AGI | Lex Fridman Podcast #490
neural network and it has this attention mechanism”
just focusing in on the decoder part. It is essentially still ainside. And you predict one token at a time. You
Play at 40:34
YouTubeThe Diary Of A CEO4.4M views · 2024-08-29
Andrew Huberman: You Must Control Your Dopamine! The Shocking Truth Behind Cold Showers!
neural network now the brain loves”
different questions into the existingquestions like the the brain since we're
Play at 56:10
YouTubeThe Diary Of A CEO1.8M views · 2025-12-18
Godfather of AI: We Have 2 Years Before Everything Changes!
neural networks to be much much much”
realized that we could train thesebetter than other existing methods that
Play at 84:45
YouTubeThe Diary Of A CEO1.2M views · 2026-01-05
Dopamine Expert: How TikTok Is Physically Rewiring Your Brain (Permanent Damage?)
“new neural networks that route around”
is characterized by the development ofthose injured areas. But the bottom line
Play at 51:26
YouTubeLex Fridman758K views · 2026-01-31
State of AI in 2026: LLMs, Coding, Scaling Laws, China, Agents, GPUs, AGI | Lex Fridman Podcast #490
“fully connected neural network layer inside the transformer. And it's very”
perceptrons, you can think of a mini multi-layer perceptron, aexpensive, because it's fully connected. If you have a thousand inputs and a thousand outputs,
Play at 41:22
YouTubeLex Fridman758K views · 2026-01-31
State of AI in 2026: LLMs, Coding, Scaling Laws, China, Agents, GPUs, AGI | Lex Fridman Podcast #490
“try to train a neural network on that." It happened to work really well, and”
Krizhevsky. So someone took these GPUs and said, "Hey, let'sI think it only happened because you could purchase those GPUs.
Play at 246:23
YouTubeLex Fridman758K views · 2026-01-31
State of AI in 2026: LLMs, Coding, Scaling Laws, China, Agents, GPUs, AGI | Lex Fridman Podcast #490
“people familiar with deep neural networks, I mean, it's the same as”
It's just like a tweak. The nonlinear activation function—changing sigmoid with ReLU. It's not changing the network fundamentally.
Play at 43:47
YouTubeLex Fridman758K views · 2026-01-31
State of AI in 2026: LLMs, Coding, Scaling Laws, China, Agents, GPUs, AGI | Lex Fridman Podcast #490
“- Anything about neural networks will be remembered? Like, do you think”
GPUs and the data centers need to talk to each other.there's something very specific and singular to the fact that it's neural
Play at 250:36
YouTubeLex Fridman4.5M views · 2024-08-02
Elon Musk: Neuralink and the Future of Humanity | Lex Fridman Podcast #438
“of these convolutional neural networks,”
And when you study the behaviorslike what the different layers of the network is detecting,
Play at 194:54
YouTubeLex Fridman758K views · 2026-01-31
State of AI in 2026: LLMs, Coding, Scaling Laws, China, Agents, GPUs, AGI | Lex Fridman Podcast #490
“still that it's a neural network. It has the problem of catastrophic”
context documents. We have a specific phase. And one problem of LLMs isforgetting. So, you teach it something, it forgets other things. And you
Play at 66:52