Skip to content
GPT-3 vs Human Brain
YouTube

High IQ Club

GPT-3 vs Human Brain

Episode notes from Lex Fridman

GPT-3 has 175 billion parameters/synapses. Human brain has 100 trillion synapses. How much will it cost to train a language model the size of the human brain?

REFERENCES:

[1] GPT-3 paper: Language Models are Few-Shot Learners
https://arxiv.org/abs/2005.14165

[2] OpenAI’s GPT-3 Language Model: A Technical Overview
https://lambdalabs.com/blog/demystifying-gpt-3/

[3] Measuring the Algorithmic Efficiency of Neural Networks
https://arxiv.org/abs/2005.04305

Help improve this page

Add chaptersNo chapter list yet. Timestamps make long episodes easier to jump around.
Start at 0:00, in time order, at least 3. Tip: open the episode on YouTube and note the time when the topic changes.
Credit (optional)
Shown on the page as a thank-you.
Never shown.

Everything is checked before it goes on the site.

Who's on this episode?Add the guest names so fans can find every appearance.
Separate names with commas. Up to 5.
The guest's own post about it, for example.
Credit (optional)
Shown on the page as a thank-you.
Never shown.

Everything is checked before it goes on the site.

Report a problemWrong title, wrong show, broken video.
A link that backs it up gets it through faster.
Credit (optional)
Shown on the page as a thank-you.
Never shown.

Everything is checked before it goes on the site.

Comments

No comments yet. Start the conversation.

Join the conversation

Your email is never shown. First comments are reviewed before they appear.