Forwarded from Samri’s Log
I’ve written down all the little things I know about embeddings. If you have any suggestions or corrections, feel free to comment. https://medium.com/@samrawitasfaw8/why-and-how-embeddings-come-a38c106fbb37
Medium
Why and How Embeddings come?
Think of the last time you searched for something online. Even if you typed something unclear, the search engine still figured it out. How…
🔥2
In 2D or 3D if you pick two vectors at random, they could point in similar directions, opposite directions, or be perpendicular so there are only a few "directions" you can go.
but In very high dimensions if you pick two random vectors, they are almost guaranteed to be perpendicular (orthogonal) to each other. and there is so much "room" that every new random vector finds a completely new direction to point in that doesn't overlap with the others
but In very high dimensions if you pick two random vectors, they are almost guaranteed to be perpendicular (orthogonal) to each other. and there is so much "room" that every new random vector finds a completely new direction to point in that doesn't overlap with the others
so technically this is the secret sauce behind Embeddings (like those used in ChatGPT). Because high-dimensional space is so vast, we can shove millions of different concepts (words, images, ideas) into a vector space, and they won't "bump into" each other unless they are actually related.
⚡3❤1
Forwarded from Onyx Design ⚡️
My Dad Was Right
I just didn’t know it then.
Every evening, he came home carrying the tiredness of the day.
Before I could speak, before I could escape into play... he asked the same quiet question
“Did you study?”
Sometimes he’d sit for a moment and say things that felt ordinary, almost annoying:
“ጌዜህን ባግባቡ ተጠቀም”
“መፅሀፍትን አንብብ“
“ጓደኛ ምርጥ”
As a child, I took it personally.
I thought he didn’t see me.
Didn’t understand what I liked.
Didn’t care about my world.
In my mind, he wasn’t guiding me
he was controlling me.
A dictator, I thought.
Then I said "ምን አለበት አሁን ብጫወት በስምእም"
He was always looking at who I could become, not who I was at that moment.
Years passed.
Life began to ask its own hard questions.
Time slipped away faster than I expected.
Mistakes became teachers.
Loss became loud.
And slowly, painfully, his voice started to make sense.
Those simple words weren’t restrictions.
They were protection.
They weren’t pressure.
They were preparation.
I wish I had listened sooner.
Not because life would have been easy
but because it wouldn’t have been so hard.
Now I think about the things he said back then.
So simple.
So easy to ignore.
So expensive to learn later.
One day, I might come home tired too.
And a child might look at me the way I once looked at him.
And I wonder
will they understand me then?
Or will they understand me only years later,
when they whisper to themselves:
“My dad was right.”
"ሰነፍ የአባቱን ተግሣጽ ይንቃል፤ ዘለፋን የሚቀበል ግን አእምሮው የበዛ ነው።"
-ምሳሌ 15:5
@OnyxDesignx
I just didn’t know it then.
Every evening, he came home carrying the tiredness of the day.
Before I could speak, before I could escape into play... he asked the same quiet question
“Did you study?”
Sometimes he’d sit for a moment and say things that felt ordinary, almost annoying:
“ጌዜህን ባግባቡ ተጠቀም”
“መፅሀፍትን አንብብ“
“ጓደኛ ምርጥ”
As a child, I took it personally.
I thought he didn’t see me.
Didn’t understand what I liked.
Didn’t care about my world.
In my mind, he wasn’t guiding me
he was controlling me.
A dictator, I thought.
Then I said "ምን አለበት አሁን ብጫወት በስምእም"
He was always looking at who I could become, not who I was at that moment.
Years passed.
Life began to ask its own hard questions.
Time slipped away faster than I expected.
Mistakes became teachers.
Loss became loud.
And slowly, painfully, his voice started to make sense.
Those simple words weren’t restrictions.
They were protection.
They weren’t pressure.
They were preparation.
I wish I had listened sooner.
Not because life would have been easy
but because it wouldn’t have been so hard.
Now I think about the things he said back then.
So simple.
So easy to ignore.
So expensive to learn later.
One day, I might come home tired too.
And a child might look at me the way I once looked at him.
And I wonder
will they understand me then?
Or will they understand me only years later,
when they whisper to themselves:
“My dad was right.”
"ሰነፍ የአባቱን ተግሣጽ ይንቃል፤ ዘለፋን የሚቀበል ግን አእምሮው የበዛ ነው።"
-ምሳሌ 15:5
@OnyxDesignx
❤5
Forwarded from Austererie⚔
Me after solving 'Two Sums' for the 100th time just to keep my leetcode streak up
😁5
Forwarded from Solo codes (Brook Solomon)
Niggas telling themselves they got potential energy like we still in physics class lol, then do something about it
( i am niggas)
( i am niggas)
😭2
Mike's ML Forge
I'm better at modeling than looking at raw data besmab
I basically just plot everything and pray😭
😁3
23 hours no sleep and i’ve unlocked the secret level where everything is funny lmaooo
Forwarded from Chapi Dev Talks
🚀 Introducing Dataset.ET
The future of AI should speak Ethiopian languages.
Today we are launching Dataset.ET — an open community initiative to build the largest dataset for Ethiopian languages.
Why this matters:
Most AI systems today barely understand Amharic, Afaan Oromo, Tigrinya, Somali, and many other Ethiopian languages. Without datasets, AI will ignore our languages.
Dataset.ET is changing that.
What we are building:
• Speech datasets
• Text datasets
• Community-validated language corpora
• Open infrastructure for Ethiopian AI
How you can help:
1️⃣ Record sentences using our Telegram bot
2️⃣ Validate recordings from other contributors
3️⃣ Help expand datasets for Ethiopian languages
No technical skills needed — just your voice and your language.
🤖 Start contributing:
https://news.1rj.ru/str/dataset_et_bot
🌍 Learn more:
https://dataset.et
💬 Join the community:
https://news.1rj.ru/str/dataset_et
Together we can teach AI to understand Ethiopia.
📢 Community leaders & channel admins:
The future of AI should speak Ethiopian languages.
Today we are launching Dataset.ET — an open community initiative to build the largest dataset for Ethiopian languages.
Why this matters:
Most AI systems today barely understand Amharic, Afaan Oromo, Tigrinya, Somali, and many other Ethiopian languages. Without datasets, AI will ignore our languages.
Dataset.ET is changing that.
What we are building:
• Speech datasets
• Text datasets
• Community-validated language corpora
• Open infrastructure for Ethiopian AI
How you can help:
1️⃣ Record sentences using our Telegram bot
2️⃣ Validate recordings from other contributors
3️⃣ Help expand datasets for Ethiopian languages
No technical skills needed — just your voice and your language.
🤖 Start contributing:
https://news.1rj.ru/str/dataset_et_bot
🌍 Learn more:
https://dataset.et
💬 Join the community:
https://news.1rj.ru/str/dataset_et
Together we can teach AI to understand Ethiopia.
If you run a Telegram group or community and believe Ethiopian languages should be part of the future of AI, we would really appreciate it if you could share this post with your community.
Your support can help thousands of people contribute their language and voice to this initiative. Thank you!
Please open Telegram to view this post
VIEW IN TELEGRAM