Mo'Khatla

Writer

[Matt Henderson @matthen2]

an exponential curve looks identical anywhere along the x axis- so long as you stretch the y axis accordingly

[unusual_whales @unusual_whales]

BREAKING: The average 30 year mortgage rate is now to 8.09%, per CNBC.

[Physics In History @PhysInHistory]

A Physics paper that has 5154 authors 📄

A record-breaking publication of a physics paper having the largest number of contributors ever- 5154. The article was published in Physical Review Letters and comprises a total of 33 pages. The actual research and references are only described in the first 9 pages. The remaining 24 pages are dedicated to listing the authors and their respective institutions.

The paper is titled,

“Combined Measurement of the Higgs Boson Mass in pp Collisions at √s=7 and 8 TeV with the ATLAS and CMS Experiments”

Full story: https://t.co/xZIT2EPN9M

Paper link: https://t.co/nBKFdV7qaO

[Mo'Khatla @Shauiib]

@WallStreetSilv What about South Africa? 🫣😂

[Santiago @svpino]

GPT-4 is getting worse over time, not better.

Many people have reported noticing a significant degradation in the quality of the model responses, but so far, it was all anecdotal.

But now we know.

At least one study shows how the June version of GPT-4 is objectively worse than the version released in March on a few tasks.

The team evaluated the models using a dataset of 500 problems where the models had to figure out whether a given integer was prime. In March, GPT-4 answered correctly 488 of these questions. In June, it only got 12 correct answers.

From 97.6% success rate down to 2.4%!

But it gets worse!

The team used Chain-of-Thought to help the model reason:

"Is 17077 a prime number? Think step by step."

Chain-of-Thought is a popular technique that significantly improves answers. Unfortunately, the latest version of GPT-4 did not generate intermediate steps and instead answered incorrectly with a simple "No."

Code generation has also gotten worse.

The team built a dataset with 50 easy problems from LeetCode and measured how many GPT-4 answers ran without any changes.

The March version succeeded in 52% of the problems, but this dropped to a pale 10% using the model from June.

Why is this happening?

We assume that OpenAI pushes changes continuously, but we don't know how the process works and how they evaluate whether the models are improving or regressing.

Rumors suggest they are using several smaller and specialized GPT-4 models that act similarly to a large model but are less expensive to run. When a user asks a question, the system decides which model to send the query to.

Cheaper and faster, but could this new approach be the problem behind the degradation in quality?

In my opinion, this is a red flag for anyone building applications that rely on GPT-4. Having the behavior of an LLM change over time is not acceptable.

Have you noticed any issues when using GPT-4 and ChatGPT lately? Do you think these problems are overblown?

[Benjamin De Kraker @BenjaminDEKR]

This is huge.

Recent academic paper shows that GPT-4 lost significant capability in code skills over a few months.

I love GPT-4 but it would be nice to have OpenAI address this.

They've been insisting that users are "imagining it," basically.

[Mo'Khatla @Shauiib]

@D_DeGea Thanks for the great service, David 🙏🏾🙏🏾🙏🏾 ¡Muchos gracias! #daveSaves

[Mo'Khatla @Shauiib]

@iamkoshiek Also: The mountain is a geographical constraint to property development, creating limited space. That plus limited building permits and zoning rights means finite supply, which naturally leads to inflation.

[Mo'Khatla @Shauiib]

@iamkoshiek @iamkoshiek Abu Dhabi is a decent option to consider

[Marcos López de Prado @lopezdeprado]

Gross performance of Factor Investing strategies over the past 10 years, before: (a) transaction costs; (b) market impact of order execution; (c) cost of borrowing stocks for shorting positions; (d) management and incentive fees.

To understand why, read: https://t.co/6TPFZG4Ell

[PyQuant News 🐍 @pyquantnews]

The BEST algo trading simulator:

Zipline.

Quantopian had $100,000,000 in algorithmic strategies using Zipline.

300,000 people used Zipline on their platform.

So I spent 100 hours studying the code so I could use it too.

The step-by-step guide to getting started in 5 minutes:

[Mo'Khatla @Shauiib]

@e2meleng_ Good times! Mweb was way too expensive though.

[Mo'Khatla @Shauiib]

@e2meleng_ Congratulations!

[Cliff Pickover @pickover]

This humor, circulating widely on the Web, appears to combine mathematics, social media, and history.

[Mo'Khatla @Shauiib]

@TAKEALOT I have not received an order that was due for delivery yesterday. I still have not received any response for support query? What kind of service is this!?

[Mo'Khatla @Shauiib]

@Boltapp_za I forgot my phone in a Bolt car. I cant contact support because i cant login - cant get an OTP without my phone...

[Mo'Khatla @Shauiib]

@randomguy_za I was late on the Griselda bus, but I am happy to finally be on board. FIRE!

[life & times of MAMBUSH ✊🏾 @neekstro_alx]

Drip season is finally here 🤞🏾💦
NEW MUSIC VIDEO for THE SPLAH(fresh2def)
from my project MAMBUSH2001🔥🔥

..."Dont worry bout it sweetheart"😅
FULL VIDEO LINK BELOW ⬇️

https://t.co/d7PyI92MBh

Please RT !!!🌍🌍🌍🌍🌍