Loading video...

Video Failed to Load

Go Home

Been tracking how LLMs do on IQ tests, especially on TrackingAI. Regardless of what you think about the test, LLMs have struggled to exceed 130, especially on questions not included in its training. Recently, models have breached this and now test better than 99% of humans

TeksEdge's profile picture

David Hendrickson

10,133 subscribers

84,101 views • 15 days ago •via X (Twitter)

Science & Technology
LIVE

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

There have been some questions about what LCR is and how it’s used in AI models. In this clip, Andrej breaks down how LLMs leverage LCR to pull real-time, up-to-date data from the web. To learn more, check out his tweet on how LCR will change AI forever.
0:30

There have been some questions about what LCR is and how it’s used in AI models. In this clip, Andrej breaks down how LLMs leverage LCR to pull real-time, up-to-date data from the web. To learn more, check out his tweet on how LCR will change AI forever.

Grass

269,160 views • 1 year ago

A 4-year-old child has seen 50x more information than the biggest LLMs. Yann LeCun is the Chief AI Scientist at Meta. He recently spoke on “The Expanding Universe of Generative Models” panel at the World Economic Forum in Davos. Yann highlighted the idea that a 4-year-old child is way smarter than current cutting-edge large language models (LLMs). “Think about what a child sees through vision. Put a number on how much information a 4-year-old child has seen during their life. It’s 20 Mbps going through the optical nerve for 16,000 wake hours in the first 4 years of life. 3,600 seconds per hour is 10^15 bytes. This is 50x more information than the biggest LLMs we have. A 4-year-old child is way smarter than these models having acquired an enormous amount of knowledge about how the world works.” The real constraint right now is the ability of LLMs to think. Today, LLMs are only capable of System 1 thinking. System 1 vs System 2 thinking was popularised in the book 'Thinking, Fast and Slow' by Daniel Kahneman. System 1 tasks involve quick, instinctive, automatic responses. LLMs struggle with discontinuous tasks that require a creative leap in progress as they imitate human responses. It's hard to go above human response accuracy if LLMs are only trained on humans. Models are building the track in front of them with each word being generated. What could it mean to give language models System 2 thinking? This remains a future development I'm excited about.
1:25

A 4-year-old child has seen 50x more information than the biggest LLMs. Yann LeCun is the Chief AI Scientist at Meta. He recently spoke on “The Expanding Universe of Generative Models” panel at the World Economic Forum in Davos. Yann highlighted the idea that a 4-year-old child is way smarter than current cutting-edge large language models (LLMs). “Think about what a child sees through vision. Put a number on how much information a 4-year-old child has seen during their life. It’s 20 Mbps going through the optical nerve for 16,000 wake hours in the first 4 years of life. 3,600 seconds per hour is 10^15 bytes. This is 50x more information than the biggest LLMs we have. A 4-year-old child is way smarter than these models having acquired an enormous amount of knowledge about how the world works.” The real constraint right now is the ability of LLMs to think. Today, LLMs are only capable of System 1 thinking. System 1 vs System 2 thinking was popularised in the book 'Thinking, Fast and Slow' by Daniel Kahneman. System 1 tasks involve quick, instinctive, automatic responses. LLMs struggle with discontinuous tasks that require a creative leap in progress as they imitate human responses. It's hard to go above human response accuracy if LLMs are only trained on humans. Models are building the track in front of them with each word being generated. What could it mean to give language models System 2 thinking? This remains a future development I'm excited about.

Alex Banks

22,985 views • 2 years ago

This is how you handle getting pulled over, especially when you have guns in the car and on you. Any questions?
2:11

Sensitive content

This is how you handle getting pulled over, especially when you have guns in the car and on you. Any questions?

I Meme Therefore I Am 🇺🇸

68,117 views • 3 days ago

This is how large language models turn objects to vector representations. In this video, we explore how large language models (LLMs) convert objects into internal representations, especially when translating between languages like English and Hindi. Using real-world examples, we highlight the challenges of gender inference, grammatical structure, and why direct word-to-word translations often fail. If you're curious about how LLMs deal with multilingual contexts and what it takes to improve translation quality across languages, this video is for you. #LLMs #Vectors #LCM
6:57

This is how large language models turn objects to vector representations. In this video, we explore how large language models (LLMs) convert objects into internal representations, especially when translating between languages like English and Hindi. Using real-world examples, we highlight the challenges of gender inference, grammatical structure, and why direct word-to-word translations often fail. If you're curious about how LLMs deal with multilingual contexts and what it takes to improve translation quality across languages, this video is for you. #LLMs #Vectors #LCM

Gaurav Sen

27,368 views • 1 year ago

Here's how I do tests in math class: For the first 5 minutes of the session, students put their writing utensils on the ground, I hand out the tests, and students can chat with each other about the test. So many students have mentioned how much this lowers their test anxiety.
0:45

Here's how I do tests in math class: For the first 5 minutes of the session, students put their writing utensils on the ground, I hand out the tests, and students can chat with each other about the test. So many students have mentioned how much this lowers their test anxiety.

Howie Hua

226,395 views • 2 years ago

If you feel like you still don’t have a handle on the fundamentals of AI/LLMs, you might be interested in a talk Gustav Söderström recently gave to our team. He dives into that how these systems work is actually much simpler than many of us think Full link:
1:02

If you feel like you still don’t have a handle on the fundamentals of AI/LLMs, you might be interested in a talk Gustav Söderström recently gave to our team. He dives into that how these systems work is actually much simpler than many of us think Full link:

Daniel Ek

67,813 views • 3 years ago

Breathalyzer tests are claimed to be alcohol specific yet many substances have been proved to test positive in court that do not contain Alcohol. This is not a "false positive" this is a falsification of the specificity of the test and is the same as PCR.
0:59

Breathalyzer tests are claimed to be alcohol specific yet many substances have been proved to test positive in court that do not contain Alcohol. This is not a "false positive" this is a falsification of the specificity of the test and is the same as PCR.

Jamie Andrews

16,929 views • 21 days ago

Breathalyzer tests are claimed to be alcohol specific yet many substances have been proved to test positive in court that do not contain Alcohol. This is not a "false positive" this is a falsification of the specificity of the test and is the same as PCR.
0:59

Breathalyzer tests are claimed to be alcohol specific yet many substances have been proved to test positive in court that do not contain Alcohol. This is not a "false positive" this is a falsification of the specificity of the test and is the same as PCR.

Jamie Andrews

50,425 views • 6 months ago

Richard Sutton says that the current dominance of LLMs is a "momentary fixation" the real breakthroughs will come from scaling computation, not building AI systems based on how humans think they work. LLMs will not be the leading edge of AI for more than another decade, perhaps even half a decade.
1:16

Richard Sutton says that the current dominance of LLMs is a "momentary fixation" the real breakthroughs will come from scaling computation, not building AI systems based on how humans think they work. LLMs will not be the leading edge of AI for more than another decade, perhaps even half a decade.

Haider.

137,573 views • 1 year ago

Bloomberg's Joel Weber puts @ScarletFu, Katie Greifeld and Eric Balchunas to the test on "Bloomberg ETF IQ" in this week's edition of "IQ Test"
2:38

Bloomberg's Joel Weber puts @ScarletFu, Katie Greifeld and Eric Balchunas to the test on "Bloomberg ETF IQ" in this week's edition of "IQ Test"

Bloomberg TV

32,610 views • 3 months ago

Bloomberg's Joel Weber puts @ScarletFu, Katie Greifeld, and Eric Balchunas to the test on "Bloomberg ETF IQ" in this week's edition of "IQ Test"
2:55

Bloomberg's Joel Weber puts @ScarletFu, Katie Greifeld, and Eric Balchunas to the test on "Bloomberg ETF IQ" in this week's edition of "IQ Test"

Bloomberg TV

17,594 views • 2 months ago

Bloomberg's Joel Weber puts Katie Greifeld, Scarlet Fu and Eric Balchunas to the test on "Bloomberg ETF IQ" in this week's edition of "IQ Test"
3:19

Bloomberg's Joel Weber puts Katie Greifeld, Scarlet Fu and Eric Balchunas to the test on "Bloomberg ETF IQ" in this week's edition of "IQ Test"

Bloomberg TV

17,227 views • 1 month ago

LLMs have complex joint beliefs about all sorts of quantities. And my postdoc James Requeima visualized them! In this thread we show LLM predictive distributions conditioned on data and free-form text. LLMs pick up on all kinds of subtle and unusual structure: 🧵
0:26

LLMs have complex joint beliefs about all sorts of quantities. And my postdoc James Requeima visualized them! In this thread we show LLM predictive distributions conditioned on data and free-form text. LLMs pick up on all kinds of subtle and unusual structure: 🧵

David Duvenaud

193,732 views • 1 year ago

Bloomberg's Joel Weber puts Scarlet Fu, Katie Greifeld and Eric Balchunas to the test on "Bloomberg ETF IQ" in this week's edition of "IQ Test"
2:32

Bloomberg's Joel Weber puts Scarlet Fu, Katie Greifeld and Eric Balchunas to the test on "Bloomberg ETF IQ" in this week's edition of "IQ Test"

Bloomberg TV

15,691 views • 1 month ago

Bloomberg's Joel Weber puts Scarlet Fu, Katie Greifeld, and Eric Balchunas to the test on "Bloomberg ETF IQ" in this week's edition of "IQ Test"
2:42

Bloomberg's Joel Weber puts Scarlet Fu, Katie Greifeld, and Eric Balchunas to the test on "Bloomberg ETF IQ" in this week's edition of "IQ Test"

Bloomberg TV

18,395 views • 1 month ago

Bloomberg's Joel Weber puts Katie Greifeld, Scarlet Fu, and Eric Balchunas to the test on "Bloomberg ETF IQ" in this week's edition of "IQ Test"
3:07

Bloomberg's Joel Weber puts Katie Greifeld, Scarlet Fu, and Eric Balchunas to the test on "Bloomberg ETF IQ" in this week's edition of "IQ Test"

Bloomberg TV

10,826 views • 4 months ago

We pre-train LLMs on the whole of the internet. You might think this explains how they learn so many emergent capabilities: the knowledge is implicit in the training data. But in fact models can do things that were never demonstrated anywhere in training! Sergey Levine argues that the real source of emergent capabilities is compositionality:
2:02

We pre-train LLMs on the whole of the internet. You might think this explains how they learn so many emergent capabilities: the knowledge is implicit in the training data. But in fact models can do things that were never demonstrated anywhere in training! Sergey Levine argues that the real source of emergent capabilities is compositionality:

Dwarkesh Patel

151,718 views • 2 months ago

Stanford just dropped a 1 hour course on how it teaches its AI engineers to train LLMs from scratch. This is the exact way LLMs like ChatGPT & Claude are trained. • 00:00 - LLM Policy Gradient basics • 23:24 - PPO for LLM training • 56:20 - how LLMs learn Chain of Thought • 1:03:50- LLM architecture explained Anthropic pays $750,000/year to engineers who understand this exact knowledge of LLMs. Bookmark this & give 1 hours today, no matter what. It'll be the most productive thing you do this week.
1:18:27

Stanford just dropped a 1 hour course on how it teaches its AI engineers to train LLMs from scratch. This is the exact way LLMs like ChatGPT & Claude are trained. • 00:00 - LLM Policy Gradient basics • 23:24 - PPO for LLM training • 56:20 - how LLMs learn Chain of Thought • 1:03:50- LLM architecture explained Anthropic pays $750,000/year to engineers who understand this exact knowledge of LLMs. Bookmark this & give 1 hours today, no matter what. It'll be the most productive thing you do this week.

Roan

37,436 views • 20 days ago

In this rare footage, an alternate shot of Mayor Vaughn and Chief Brody on the ferry. Do you think this shot should have been included in the final edit of #JAWS?
0:59

In this rare footage, an alternate shot of Mayor Vaughn and Chief Brody on the ferry. Do you think this shot should have been included in the final edit of #JAWS?

The Daily Jaws

23,919 views • 11 months ago