Johntradzz

Ultra Endurance athlete 🏃🚴‍♂️ I Student and options trader

[Zak @Zak60191459]

The language of the universe
#mathematics
@grok your opinion about this video?

[Zak @Zak60191459]

https://t.co/xxga8Duh5T

[Zak @Zak60191459]

https://t.co/a5gxtcl0qA Practice anytime anywhere.
#math #realanalysis #linear_algebra

[Zak @Zak60191459]

You write a proof and have no idea where it actually falls apart.
LeetMath grades your proof against a real university rubric.
No full solution. No “here’s the answer.” Just the exact gaps: missing induction hypothesis, sum not split, no conclusion for all n.
Type it or write by hand. Instant feedback. Hints if you’re stuck. Free, no account needed.
https://t.co/mcrIIjEhwE
Which proof are you working on right now?
#MathTwitter #ProofWriting #UniversityMath #STEM #Analysis

[Zak @Zak60191459]

LeetMath the Product: https://t.co/TCmXSxMKrl

[Zak @Zak60191459]

Practice writing Proofs ✍️
https://t.co/BmEPPDQVWI
#math

[Johntradzz @Johntradzz]

https://t.co/EmDSFbUYIP

[Zak @Zak60191459]

Introducing LeetMath 1.0
The only Place to practice Proofs.
https://t.co/TCmXSxMKrl

#math #LeetMath

[Johntradzz @Johntradzz]

@probnstat

[Johntradzz @Johntradzz]

@probnstat

[Johntradzz @Johntradzz]

@probnstat P(1st miss on 4th attempt) = P(Make on 1st) * P(Make on 2nd) * P(Make on 3rd) * P(Miss on 4th)

[Johntradzz @Johntradzz]

@probnstat p(at least one of them is red) = P(RR) + P(RB) + P(BR) = 25/28

[Johntradzz @Johntradzz]

Hi François, amazing work.
Quick question: Where or how can I access the Chain of Thought (CoT) data for the ~9% of tasks o3 struggled with, as mentioned in the report?My brother and I are keen to work on this topic for our bachelor's project @Zak60191459 @fchollet

[François Chollet @fchollet]

Today OpenAI announced o3, its next-gen reasoning model. We've worked with OpenAI to test it on ARC-AGI, and we believe it represents a significant breakthrough in getting AI to adapt to novel tasks.

It scores 75.7% on the semi-private eval in low-compute mode (for $20 per task in compute ) and 87.5% in high-compute mode (thousands of $ per task). It's very expensive, but it's not just brute -- these capabilities are new territory and they demand serious scientific attention.

[Johntradzz @Johntradzz]

@karpathy Random Forests?Does the challenge lie in the tokenization process? It seems like the core idea of language modeling could be very useful in fields like numerical analysis, especially if we employed a different encoding method.

[Johntradzz @Johntradzz]

@karpathy Hi Andre, I was curious about the application of LLMs in numerical analysis, especially for tasks like financial forecasting or predicting stock prices. Why is LLM training often considered less effective for these numerical tasks compared to traditional methods like …

[Johntradzz @Johntradzz]

@Alan_Couzens Thanks Alan🙏 , I’ve been averaging around 40-60 km of running a week for the last two years. I am now increasing mileage to 80-90 km a week. Not really sure if this is a good idea but so far for the first week body feels good. Do you have any tips regarding increasing vol ?

[Johntradzz @Johntradzz]

@Alan_Couzens no hard runs *

[Johntradzz @Johntradzz]

@Alan_Couzens Hi Alan, a bit off-topic top. But I’ve been noticing a trend in my HRV as you can see in a photo attached. Not really sure why, but the thing that I started implementing is on those days where attend to have an unbalanced HRV I do a very very easy cycling or running so hard runs

[Johntradzz @Johntradzz]

@Alan_Couzens What would be your recommendation completely off or just easy?

[Nassim Nicholas Taleb @nntaleb]

If you change your mind too frequently, it suggests that you do not think carefully and responsibly before formulating an opinion & don't know when to remain silent or neutral.

If you never change your mind on anything, it indicates that you are an intellectually dishonest.

[Johntradzz @Johntradzz]

@Alan_Couzens Hi Alan,
Regarding .2 , would you say one to 2 hours after training is still okay ?