Mike X Cohen PhD, Sincxpress Education
@mikexcohen.bsky.social
Check out my longer posts on Substack (mostly detailed tutorials about data science and LLMs). mikexcohen.substack.com
Explore my video-based courses and books ➔ sincxpress.com
Explore my video-based courses and books ➔ sincxpress.com
In my latest substack post, I show how to measure gender bias in open-source LLMs. The post comes with detailed explanations and a Python notebook file that you can easily adapt to any other biases.
Curious? Come check it out :)
mikexcohen.substack.com/p/gender-bia...
Curious? Come check it out :)
mikexcohen.substack.com/p/gender-bia...
November 6, 2025 at 3:08 PM
In my latest substack post, I show how to measure gender bias in open-source LLMs. The post comes with detailed explanations and a Python notebook file that you can easily adapt to any other biases.
Curious? Come check it out :)
mikexcohen.substack.com/p/gender-bia...
Curious? Come check it out :)
mikexcohen.substack.com/p/gender-bia...
No comment.
October 27, 2025 at 12:36 PM
No comment.
I don't know what's going on in this AI image.
But I do know this: Daily Data Challenge #25 is short and very practical for data science. You'll get to create a symmetric matrix and extract its unique elements.
Can you solve it in less than 6 minutes?
mikexcohen.substack.com/p/ddc-25-uni...
But I do know this: Daily Data Challenge #25 is short and very practical for data science. You'll get to create a symmetric matrix and extract its unique elements.
Can you solve it in less than 6 minutes?
mikexcohen.substack.com/p/ddc-25-uni...
October 8, 2025 at 6:09 PM
I don't know what's going on in this AI image.
But I do know this: Daily Data Challenge #25 is short and very practical for data science. You'll get to create a symmetric matrix and extract its unique elements.
Can you solve it in less than 6 minutes?
mikexcohen.substack.com/p/ddc-25-uni...
But I do know this: Daily Data Challenge #25 is short and very practical for data science. You'll get to create a symmetric matrix and extract its unique elements.
Can you solve it in less than 6 minutes?
mikexcohen.substack.com/p/ddc-25-uni...
Here’s a picture of Thalorion The Destroyer reading my calculus textbook. I hope he likes it! Otherwise he might unleash doom on humanity.
www.amazon.com/dp/B0DX6BPPZ8
www.amazon.com/dp/B0DX6BPPZ8
October 7, 2025 at 2:06 PM
Here’s a picture of Thalorion The Destroyer reading my calculus textbook. I hope he likes it! Otherwise he might unleash doom on humanity.
www.amazon.com/dp/B0DX6BPPZ8
www.amazon.com/dp/B0DX6BPPZ8
genAI never fails to deliver the laughs.
September 27, 2025 at 6:08 AM
genAI never fails to deliver the laughs.
Daily Data Challenges are going strong.
Today's challenge is to create this beautiful graph using numpy and matplotlib. Can you do it?
mikexcohen.substack.com/p/ddc-18-err...
Today's challenge is to create this beautiful graph using numpy and matplotlib. Can you do it?
mikexcohen.substack.com/p/ddc-18-err...
September 26, 2025 at 11:03 AM
Daily Data Challenges are going strong.
Today's challenge is to create this beautiful graph using numpy and matplotlib. Can you do it?
mikexcohen.substack.com/p/ddc-18-err...
Today's challenge is to create this beautiful graph using numpy and matplotlib. Can you do it?
mikexcohen.substack.com/p/ddc-18-err...
Have you done your Daily Data Challenge?
mikexcohen.substack.com/p/ddc-8-test...
mikexcohen.substack.com/p/ddc-8-test...
September 12, 2025 at 5:10 AM
Have you done your Daily Data Challenge?
mikexcohen.substack.com/p/ddc-8-test...
mikexcohen.substack.com/p/ddc-8-test...
People talk a lot about GPT5 and hallucinations and trust and so on.
But let's not forget that GPT2 is the ultimate arbiter of truth and science.
Proof of my claim? This zinger from GPT2.
Wanna explore GPT2's wisdom in your own private Python session? mikexcohen.substack.com/p/lessons-on...
But let's not forget that GPT2 is the ultimate arbiter of truth and science.
Proof of my claim? This zinger from GPT2.
Wanna explore GPT2's wisdom in your own private Python session? mikexcohen.substack.com/p/lessons-on...
September 11, 2025 at 2:17 PM
People talk a lot about GPT5 and hallucinations and trust and so on.
But let's not forget that GPT2 is the ultimate arbiter of truth and science.
Proof of my claim? This zinger from GPT2.
Wanna explore GPT2's wisdom in your own private Python session? mikexcohen.substack.com/p/lessons-on...
But let's not forget that GPT2 is the ultimate arbiter of truth and science.
Proof of my claim? This zinger from GPT2.
Wanna explore GPT2's wisdom in your own private Python session? mikexcohen.substack.com/p/lessons-on...
GPT2 is hilarious.
September 10, 2025 at 11:22 AM
GPT2 is hilarious.
Can you code the Fourier transform from scratch?
I bet you can. Spend 20 minutes on the Substack post, and you'll learn everything you need to understand and code the FT in Python.
mikexcohen.substack.com/p/the-fourie...
I bet you can. Spend 20 minutes on the Substack post, and you'll learn everything you need to understand and code the FT in Python.
mikexcohen.substack.com/p/the-fourie...
September 8, 2025 at 5:25 PM
Can you code the Fourier transform from scratch?
I bet you can. Spend 20 minutes on the Substack post, and you'll learn everything you need to understand and code the FT in Python.
mikexcohen.substack.com/p/the-fourie...
I bet you can. Spend 20 minutes on the Substack post, and you'll learn everything you need to understand and code the FT in Python.
mikexcohen.substack.com/p/the-fourie...
Daily data challenge. 10 minutes a day to improve your coding, data science, and visualization skills.
open.substack.com/pub/mikexcoh...
open.substack.com/pub/mikexcoh...
September 6, 2025 at 7:51 PM
Daily data challenge. 10 minutes a day to improve your coding, data science, and visualization skills.
open.substack.com/pub/mikexcoh...
open.substack.com/pub/mikexcoh...
Why do artificial neural network models like LLMs expand the data dimensionality?
It's because nonlinearly separable features can become linearly separable in higher-dimensional space.
mikexcohen.substack.com/p/llm-breakd...
It's because nonlinearly separable features can become linearly separable in higher-dimensional space.
mikexcohen.substack.com/p/llm-breakd...
September 3, 2025 at 3:00 PM
Why do artificial neural network models like LLMs expand the data dimensionality?
It's because nonlinearly separable features can become linearly separable in higher-dimensional space.
mikexcohen.substack.com/p/llm-breakd...
It's because nonlinearly separable features can become linearly separable in higher-dimensional space.
mikexcohen.substack.com/p/llm-breakd...
These heatmaps show how the distributions of Q, K, and V vectors inside a GPT transformer change as the text travels deeper into the language model, and picks up new contextual information.
Wanna learn more? Check my Substack post (comes with Python code!)
mikexcohen.substack.com/p/llm-breakd...
Wanna learn more? Check my Substack post (comes with Python code!)
mikexcohen.substack.com/p/llm-breakd...
September 2, 2025 at 7:46 PM
These heatmaps show how the distributions of Q, K, and V vectors inside a GPT transformer change as the text travels deeper into the language model, and picks up new contextual information.
Wanna learn more? Check my Substack post (comes with Python code!)
mikexcohen.substack.com/p/llm-breakd...
Wanna learn more? Check my Substack post (comes with Python code!)
mikexcohen.substack.com/p/llm-breakd...
I manipulated GPT2's internal representations to prefer "Berlin" vs. "Paris." You can do it too! Check out my Substack post to find out how.
mikexcohen.substack.com/p/llm-breakd...
mikexcohen.substack.com/p/llm-breakd...
August 30, 2025 at 5:31 AM
I manipulated GPT2's internal representations to prefer "Berlin" vs. "Paris." You can do it too! Check out my Substack post to find out how.
mikexcohen.substack.com/p/llm-breakd...
mikexcohen.substack.com/p/llm-breakd...
Have you seen those neat-looking text heatmaps that show how LLMs like Claude or GPT interpret text? They're surprisingly easy to create in Python.
open.substack.com/pub/mikexcoh...
open.substack.com/pub/mikexcoh...
August 29, 2025 at 2:51 AM
Have you seen those neat-looking text heatmaps that show how LLMs like Claude or GPT interpret text? They're surprisingly easy to create in Python.
open.substack.com/pub/mikexcoh...
open.substack.com/pub/mikexcoh...
Good morning smiley people :)
December 11, 2024 at 6:49 AM
Good morning smiley people :)
I somehow feel like odd functions should be counted with the thumb, middle, and pinky fingers, while even functions should be counted with the pointer and ring fingers.
November 26, 2024 at 9:05 AM
I somehow feel like odd functions should be counted with the thumb, middle, and pinky fingers, while even functions should be counted with the pointer and ring fingers.
My fav way to work on a rainy Sunday.
November 17, 2024 at 11:31 AM
My fav way to work on a rainy Sunday.
The limit of sin(x)/x as x->0 is -13.61. Pretty obvious from looking at the plot.
November 14, 2024 at 8:45 AM
The limit of sin(x)/x as x->0 is -13.61. Pretty obvious from looking at the plot.
November 13, 2024 at 9:41 AM