Joseph Magara 🇺🇬🇺🇬

6.7K posts

Joseph Magara 🇺🇬🇺🇬 banner
Joseph Magara 🇺🇬🇺🇬

Joseph Magara 🇺🇬🇺🇬

@joemagara

Life as seen through the prism of my glasses#

Kampala, Uganda Katılım Temmuz 2012
938 Takip Edilen374 Takipçiler
Joseph Magara 🇺🇬🇺🇬
I don’t disagree with anything you say there, but just a gentle reminder sister, remember the patience of our Lord with us. I have heard you share your beautiful and powerful testimony. Remember how He was patient with you. Remember how He was patient with Paul. Remember how He is still patient with us all even today. With that in mind, remember how He has called us to be patient with the outsider in the hope that they would repent. It’s the kindness of God that leads us to repentance and often times, God shows His kindness to us, through the words and actions of His people. ❤️ Don’t loose hope that God can still save Sneako from the snare of the evil one. Better he be in a land that is filled with many professing and true children of God who can continually tell and show Him the gospel in their words and deeds than in a land where the snare of the evil one has a strong hold.
Joseph Magara 🇺🇬🇺🇬 tweet media
English
0
0
1
18
🌷 LIZZIE🌷
🌷 LIZZIE🌷@farmingandJesus·
@joemagara @sneako But I believe he needs to go back honestly. He hates it here and he’s already rejected the gospel
English
1
0
4
67
SNEAKO
SNEAKO@sneako·
I need to get married
English
1.6K
206
5.2K
882.8K
Joseph Magara 🇺🇬🇺🇬 retweetledi
Lex Mercado
Lex Mercado@LexisMercado152·
@mancityhardcore How do you do a VAR review 100 yards away from goal?
English
3
2
5
358
Rory Talks Football
Rory Talks Football@Rory_Talks_Ball·
Reckon if that was an Argentina equaliser they’d have ruled it out?
English
109
209
6.5K
193.4K
Oladoja
Oladoja@_onlyscott·
🚨 Guess the player VERY HARD 🤯
Oladoja tweet media
English
4.3K
420
7.6K
1.9M
Analistas APP descárgala
Analistas APP descárgala@SomosAnalistas_·
Jordan Henderson se rompió el brazo saltando la publicidad en los festejos de Inglaterra. Se perderá lo que resta de la Copa del Mundo
Español
1.3K
6K
94.6K
6.5M
Professor Hamza Ballon Dor
Professor Hamza Ballon Dor@mancityhardcore·
Gabriel Magalhaes let Arsenal down and now he has let Brazil 🇧🇷 down
English
28
15
168
9.9K
WelBeast
WelBeast@WelBeast·
I just saw Morocco’s path to the final and I weeped for Africa.
English
183
547
14.7K
531.9K
Jay
Jay@Jay_marvins·
@WelBeast Also morroco to the rest of African countries
Jay tweet media
English
1
0
7
7.8K
Larry Madowo
Larry Madowo@LarryMadowo·
I will not be explaining any further
English
73
8
901
241.7K
Larry Madowo
Larry Madowo@LarryMadowo·
Ghana and Cabo Verde leaving hurts but Africa still has 3 teams left
Larry Madowo tweet media
Imizamo Yethu, South Africa 🇿🇦 English
718
386
9.9K
2.3M
Joseph Magara 🇺🇬🇺🇬
@theo It’s reached the point we’re is no longer the model, it’s your process. I’ve reached the point of assessing outcomes (essentially QA’ing) and leaving the code to the LLM. I can dig into it if needed but mostly, I trust the agent, the process and the coding patterns I’ve setup.
English
0
0
0
49
Theo - t3.gg
Theo - t3.gg@theo·
How much better do the models have to get before you'll stop reading the code?
English
1.1K
62
3.4K
2M
Joseph Magara 🇺🇬🇺🇬
I guess my point is that if you’re doing all this to get around gov restrictions, you would have to run your won inference (which requires the infrastructure and the costs associated with it). If you are using a stolen/restricted model (as the OP said), the people running inference for you would be easily found and shut down by the government (and prosecuted). With open source models, you have the weights and can reduce the size of the model to run on consumer hardware but yeah, if you don’t have the weights, that would not be possible AFIAK. Also, you’d have to live looking over your shoulder, so don’t steal, it’s not worth the trouble and it’s just not good - don’t do it.
English
0
0
0
14
John Manedrake (L/0)
John Manedrake (L/0)@Xrayhighs·
@joemagara @rationalaussie bro, its not different to any other open source model rn. even if there is elaborate architecture around it, required to run, it would work. there is a lot of inference waiting for these
English
1
0
1
28
Rational Aussie
Rational Aussie@rationalaussie·
The idea that a model can be kept behind closed doors at all over any sustained period of time is completely retarded anyway. It's not like nuclear materials. Someone can literally walk out of there with the weights on a hard drive and give it to the world. And that's probably what is going to end up happening eventually if the government's stance is only they and the companies they select have access to it. Just a massive own goal.
David Ondrej@DavidOndrej1

We must build our own, open-source, decentralized, distributed AGI there is no other way. right now politicians and big tech executives are holding us hostage they literally hold our future in their hands this HAS to change.

English
7
4
73
4.2K
Elon Musk
Elon Musk@elonmusk·
Grok 4.5, based on our 1.5T V9 foundation model, with Cursor data added in supplemental training, is now in private beta at SpaceX & Tesla. Early evals show performance close to, perhaps exceeding Opus. RL is continuing to significantly improve the model, and the Grok Build harness gets better every day. Nice work by all those involved! Completely trained from scratch new models will be released by @SpaceX every month this year.
English
5.4K
4.8K
36.9K
8.4M
Matthew Berman
Matthew Berman@MatthewBerman·
Ok, back to optimism ❤️
English
53
9
379
34.3K