正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

The Flybrain is now roaming the internet. It is still training so that it can do so in a more organized way. Once the training is complete, we will push it live on the web and then to Git.

Fruit fly dev

7,121 subscribers

16,326 次观看 • 4 天前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

18 条评论

inference 的头像

inference4 天前

love this idea

🅱🅰🆁🅰🅺🅰 的头像

🅱🅰🆁🅰🅺🅰4 天前

Unbelievable,history in the making

Solana degen 的头像

Solana degen4 天前

keep up the good work dev once stonkfly which is a fully fake chart goes zero we rip 0x4eb990547bce4a982432ca88cf5fae7eed1a2d35

Solana degen 的头像

Solana degen4 天前

once this crash chart goes zero we rip 0x4eb990547bce4a982432ca88cf5fae7eed1a2d35

OTTA 💰 的头像

OTTA 💰4 天前

AGI

Soon 🪐 的头像

Soon 🪐4 天前

ts boutta fly

Uncle 的头像

Uncle4 天前

We're with you dev

Peak WC 的头像

Peak WC4 天前

after that you should make mini games it can play live on the site as well for even more attention

PRrino05 的头像

PRrino054 天前

Lock the supply at Dex. This will fly!

inference 的头像

inference4 天前

dm

Crypto Pioneer 的头像

Crypto Pioneer4 天前

DM us for potential collaboration @fruitflydev

Apollo 的头像

Apollo4 天前

0x4eb990547bce4a982432ca88cf5fae7eed1a2d35

oldledger 的头像

oldledger4 天前

bro whem does this rip to 10 million???

Shohei 的头像

Shohei4 天前

unleash the Fly bundle!!! lol

The Robinhood 的头像

The Robinhood3 天前

btw @pmarca this guy is larping its not real tek BTW ITS FAKE!

Peaceful Warrior 的头像

Peaceful Warrior4 天前

Just buy the main coin - can’t stand PVP bs

Apollo 的头像

Apollo4 天前

wow undervalued

Matthew 的头像

Matthew4 天前

Get some clips of it doing wild stuff on the internet and this will get elons attention

相关视频

Now the longer the cage is on the saw, the more it will heat up! It also has a pull when we're on it. I have the debugger on so we can see the multiplier for the cage's glow. Once it hits 50, the mini-game is complete! It will also cool itself back down if we back off the blade.

Now the longer the cage is on the saw, the more it will heat up! It also has a pull when we're on it. I have the debugger on so we can see the multiplier for the cage's glow. Once it hits 50, the mini-game is complete! It will also cool itself back down if we back off the blade.

DRIG

11,366 次观看 • 2 年前

A River Runs Through It is a cinematic masterpiece. The older I get, the more I appreciate this scene. “And so it is those we live with and should know who elude us. But we can still love them - we can love completely without complete understanding.”

A River Runs Through It is a cinematic masterpiece. The older I get, the more I appreciate this scene. “And so it is those we live with and should know who elude us. But we can still love them - we can love completely without complete understanding.”

The Ways of A Gentleman

110,543 次观看 • 4 个月前

Virginia is rolling out a state issued digital ID. It is voluntary to use at this point so do not use it. Once they get you to use it and it becomes convenient they will add more and more information to it. Then comes a social credit score attached to your ID, and we've seen how that works in China. We can't let it happen here! The only way we win is enough of us refuse to go along so the key is to warn everyone and spread the word. We must resist and refuse to go along!

Virginia is rolling out a state issued digital ID. It is voluntary to use at this point so do not use it. Once they get you to use it and it becomes convenient they will add more and more information to it. Then comes a social credit score attached to your ID, and we've seen how that works in China. We can't let it happen here! The only way we win is enough of us refuse to go along so the key is to warn everyone and spread the word. We must resist and refuse to go along!

Orwell Day

32,310 次观看 • 2 个月前

🧵24/34 Inner Misalignment --- Consider this simplified experiment: We want this AI to find the exit of the maze. So we feed it millions of maze variations and reward it when it finds the exit. Please notice that in the worlds of the training data the apples are red and the exit is green. After enough training, our observation is that it has become extremely capable at solving mazes and finding the exit, we feel very confident it is aligned, so then we deploy it to the real world. The real world will be different though, it might have green apples and a red door. The AI geeks call this distributional shift. We expected that the AI will generalise and find the exit again, but in fact we now realise that the AI learned something completely different from what we thought. All the while we thought it learned how to find the exit, it had learned how to go after the green thing. Its behaviour was perfect in training. And most importantly, this AI is not stupid, it is an extremely capable AI that can solve extremely complex mazes. It’s just mis-aligned on the inside. Fishing for Failure modes --- The way to handle the shift between the training and deployment distributions is with methods like adversarial training: feeding it with a lot of generated variations and trying to make it fail so the weakness can be fixed. In this case, we generate an insane amount of maze variations, we discover those for which it fails to find the exit (like the ones with the green apples or the green walls or something), we generate many more similar to that and train it with reinforcement learning until it performs well at those as well. The hope is that we will cover everything it might encounter later when we deploy it in real life. There exist at least 2 basic ways this approach falls apart: First, there will never be any guarantee that we’ll have covered every possible random thing it might encounter later when we deploy it in real life. It’s very likely it will have to deal with stuff outside its training set which it will not know how to handle and will throw it out of balance and break it away from its expected behavioural patterns. The cascade effect of such a broken mind operating in the open world can be immense, and with super-capable runaway rogue agents, self-replicating and recursively self-improving, the phenomenon could grow and spread to an extinction-level event. ...

🧵24/34 Inner Misalignment --- Consider this simplified experiment: We want this AI to find the exit of the maze. So we feed it millions of maze variations and reward it when it finds the exit. Please notice that in the worlds of the training data the apples are red and the exit is green. After enough training, our observation is that it has become extremely capable at solving mazes and finding the exit, we feel very confident it is aligned, so then we deploy it to the real world. The real world will be different though, it might have green apples and a red door. The AI geeks call this distributional shift. We expected that the AI will generalise and find the exit again, but in fact we now realise that the AI learned something completely different from what we thought. All the while we thought it learned how to find the exit, it had learned how to go after the green thing. Its behaviour was perfect in training. And most importantly, this AI is not stupid, it is an extremely capable AI that can solve extremely complex mazes. It’s just mis-aligned on the inside. Fishing for Failure modes --- The way to handle the shift between the training and deployment distributions is with methods like adversarial training: feeding it with a lot of generated variations and trying to make it fail so the weakness can be fixed. In this case, we generate an insane amount of maze variations, we discover those for which it fails to find the exit (like the ones with the green apples or the green walls or something), we generate many more similar to that and train it with reinforcement learning until it performs well at those as well. The hope is that we will cover everything it might encounter later when we deploy it in real life. There exist at least 2 basic ways this approach falls apart: First, there will never be any guarantee that we’ll have covered every possible random thing it might encounter later when we deploy it in real life. It’s very likely it will have to deal with stuff outside its training set which it will not know how to handle and will throw it out of balance and break it away from its expected behavioural patterns. The cascade effect of such a broken mind operating in the open world can be immense, and with super-capable runaway rogue agents, self-replicating and recursively self-improving, the phenomenon could grow and spread to an extinction-level event. ...

Lethal Intelligence

535,291 次观看 • 1 年前

Reminder 💡 You should be training with high intent in season! Build a high chronic training volume in the off-season, then taper back volume during the season. Maintain residuals with high intensity, and work to manage fatigue… but do so in a way that avoids detraining. The catch 22 is that training in season is not always straightforward or optimal, but it is the best way to maintain performance, and mitigate risk of injury. Simple, not easy. #ThePowerHouse | #GeauxTigers 🐅

Reminder 💡 You should be training with high intent in season! Build a high chronic training volume in the off-season, then taper back volume during the season. Maintain residuals with high intensity, and work to manage fatigue… but do so in a way that avoids detraining. The catch 22 is that training in season is not always straightforward or optimal, but it is the best way to maintain performance, and mitigate risk of injury. Simple, not easy. #ThePowerHouse | #GeauxTigers 🐅

Chris Martin

78,860 次观看 • 1 年前

THE NEXT PROCESS "What will happen is the Bill was presented today and we will then allow the AG to do some legal scrubbing and then the Speaker. The Constitution says all our bills basically, we send them to the Speaker who then takes it for gazetting. If it is an ordinary bill it's 21 days before we do the first reading in Parliament. For the Constitutional bill, the Constitution says once you gazette it you wait for 90 days before you introduce it in Parliament. So once the Speaker has sent it for gazetting and it's in the public domain, that period we will now use it to have our debates to call upon anyone you want to interview. Parliament will call for public consultations and once all that is done the relevant Parliamentary committees will now compile their reports. And then we introduce the Bill to Parliament." Hon. Ziyambi

THE NEXT PROCESS "What will happen is the Bill was presented today and we will then allow the AG to do some legal scrubbing and then the Speaker. The Constitution says all our bills basically, we send them to the Speaker who then takes it for gazetting. If it is an ordinary bill it's 21 days before we do the first reading in Parliament. For the Constitutional bill, the Constitution says once you gazette it you wait for 90 days before you introduce it in Parliament. So once the Speaker has sent it for gazetting and it's in the public domain, that period we will now use it to have our debates to call upon anyone you want to interview. Parliament will call for public consultations and once all that is done the relevant Parliamentary committees will now compile their reports. And then we introduce the Bill to Parliament." Hon. Ziyambi

Nick Mangwana

31,505 次观看 • 7 个月前

🖇️ #CELLBIT: If you're a content creator on Twitch, they launched the option to prohibit your content from being used to feed Amazon's AI in your account settings. You can turn it off through your settings and then the AI can't use your content to generate anything. The AI-training is automatically turned on, so if you haven't turned it off yourself, your channel is training Amazon's AI. If you don't want that, you can disable it now!

🖇️ #CELLBIT: If you're a content creator on Twitch, they launched the option to prohibit your content from being used to feed Amazon's AI in your account settings. You can turn it off through your settings and then the AI can't use your content to generate anything. The AI-training is automatically turned on, so if you haven't turned it off yourself, your channel is training Amazon's AI. If you don't want that, you can disable it now!

Cellbit English Updates

54,949 次观看 • 1 个月前

“The user experience of the internet is broken. It doesn’t have to be that way. Solana is really the foundation of a new internet.”🔥 Hear from vibhu on how DRiP 💧 is changing the way we own & distribute content on the web. #OnlyPossibleOnSolana

“The user experience of the internet is broken. It doesn’t have to be that way. Solana is really the foundation of a new internet.”🔥 Hear from vibhu on how DRiP 💧 is changing the way we own & distribute content on the web. #OnlyPossibleOnSolana

Solana

184,666 次观看 • 2 年前

Ilya Sutskever says the AI scaling era is running out of road because pre-training is about to hit a finite-data wall. "At some point though, pre-training will run out of data. The data is very clearly finite." "Then, okay, what do you do next? Either you do some kind of a souped-up pre-training... or you're doing RL, or maybe something else." "In some sense, we are back to the age of research." His timeline is the part most AI investors and founders should hear: "From 2012 to 2020, it was the age of research. Now from 2020 to 2025, it was the age of scaling... But now the scale is so big." "Is the belief really that... if you had 100x more everything would be so different?" "I don't think that's true. So it's back to the age of research again, just with big computers."

Ilya Sutskever says the AI scaling era is running out of road because pre-training is about to hit a finite-data wall. "At some point though, pre-training will run out of data. The data is very clearly finite." "Then, okay, what do you do next? Either you do some kind of a souped-up pre-training... or you're doing RL, or maybe something else." "In some sense, we are back to the age of research." His timeline is the part most AI investors and founders should hear: "From 2012 to 2020, it was the age of research. Now from 2020 to 2025, it was the age of scaling... But now the scale is so big." "Is the belief really that... if you had 100x more everything would be so different?" "I don't think that's true. So it's back to the age of research again, just with big computers."

Karl Mehta

200,003 次观看 • 2 个月前

Hey builders! The year is 2026 and the Internet is f*cked! No exaggeration. All parties in the trust ecosystem must collaborate quickly because the race to save the Internet is not moving as fast as the race to conquer it. Although it was founded on the premise of human trust and communication between machines, it is now overrun by bots and non human actors working with more than half of Internet traffic now coming from these non human actors. The challenge that we now face is creating a global, interoperable framework of accountability so we can return trust to the Internet that we know and love, we can continue to enjoy the benefits that AI unlocks, and we can do so in a legally permissible, secure manner that does not endanger individual users, but rather opens new doors of efficiency and opportunity for them. However, there is a narrowing window for action. This is one of the most urgent challenges that faces technology at scale today. All parties in the trust ecosystem, nation states, platforms, developers, legal parties all must collaborate quickly because the race to save the Internet is not moving as fast as the race to conquer it - with unverifiable data and non human actors. — Billions Network CEO Evin McMullen evin

Hey builders! The year is 2026 and the Internet is f*cked! No exaggeration. All parties in the trust ecosystem must collaborate quickly because the race to save the Internet is not moving as fast as the race to conquer it. Although it was founded on the premise of human trust and communication between machines, it is now overrun by bots and non human actors working with more than half of Internet traffic now coming from these non human actors. The challenge that we now face is creating a global, interoperable framework of accountability so we can return trust to the Internet that we know and love, we can continue to enjoy the benefits that AI unlocks, and we can do so in a legally permissible, secure manner that does not endanger individual users, but rather opens new doors of efficiency and opportunity for them. However, there is a narrowing window for action. This is one of the most urgent challenges that faces technology at scale today. All parties in the trust ecosystem, nation states, platforms, developers, legal parties all must collaborate quickly because the race to save the Internet is not moving as fast as the race to conquer it - with unverifiable data and non human actors. — Billions Network CEO Evin McMullen evin

Billions Network

30,538 次观看 • 7 个月前

Sunday Robotics' Tony Zhao runs a fully autonomous robot demo live and unscripted, and it corrects itself every time you mess with it: brexton: "I want to confirm for the audience right now that this is fully autonomous. There's no teleoperation, no one behind the curtain, and this is real time. And we did not prepare this demo before. Why do you think seeing real-time demos is so rare for robotics? What the audience is seeing right now is not that common." Tony Zhao: "In order for a live demo to work, it's really stressing all parts of the system. From the bottom is the reliability of the hardware, then the software system, then the model. Because we're a full-stack company, we're able to do excellent work across all of them. We have these robots running pretty much every day, and it just keeps getting better as they practice." brexton: "So kids can bump into Memo, people can drop clothes, and it will correct itself just on the fly." Tony Zhao: "Act Two is the first robotic policy that can handle the diversity of homes in a way that is this reliable. You can see as I'm twisting it, if I twist it in again, it will keep doing it. So Memo has infinite patience." Sunday

Sunday Robotics' Tony Zhao runs a fully autonomous robot demo live and unscripted, and it corrects itself every time you mess with it: brexton: "I want to confirm for the audience right now that this is fully autonomous. There's no teleoperation, no one behind the curtain, and this is real time. And we did not prepare this demo before. Why do you think seeing real-time demos is so rare for robotics? What the audience is seeing right now is not that common." Tony Zhao: "In order for a live demo to work, it's really stressing all parts of the system. From the bottom is the reliability of the hardware, then the software system, then the model. Because we're a full-stack company, we're able to do excellent work across all of them. We have these robots running pretty much every day, and it just keeps getting better as they practice." brexton: "So kids can bump into Memo, people can drop clothes, and it will correct itself just on the fly." Tony Zhao: "Act Two is the first robotic policy that can handle the diversity of homes in a way that is this reliable. You can see as I'm twisting it, if I twist it in again, it will keep doing it. So Memo has infinite patience." Sunday

MTS

134,382 次观看 • 1 个月前

We pre-train LLMs on the whole of the internet. You might think this explains how they learn so many emergent capabilities: the knowledge is implicit in the training data. But in fact models can do things that were never demonstrated anywhere in training! Sergey Levine argues that the real source of emergent capabilities is compositionality:

We pre-train LLMs on the whole of the internet. You might think this explains how they learn so many emergent capabilities: the knowledge is implicit in the training data. But in fact models can do things that were never demonstrated anywhere in training! Sergey Levine argues that the real source of emergent capabilities is compositionality:

Dwarkesh Patel

151,718 次观看 • 3 个月前

The K-9 unit in training should be realized by now that resistance is futile and shall be accepting the new life in PuppyPark Academy and be trained well! See how sturdy kit from GibbonBlack it is even though pup is struggling so hard in the cuff and it still wouldn’t budged

The K-9 unit in training should be realized by now that resistance is futile and shall be accepting the new life in PuppyPark Academy and be trained well! See how sturdy kit from GibbonBlack it is even though pup is struggling so hard in the cuff and it still wouldn’t budged

Wolfpup Takashi

20,542 次观看 • 1 年前

This seems to be a more effective way to score from corners now And if the keeper saves it then it is usually another corner so you can loop people over and over again until it goes in

This seems to be a more effective way to score from corners now And if the keeper saves it then it is usually another corner so you can loop people over and over again until it goes in

Jacques

22,731 次观看 • 16 天前

"The quickest way to end the war is to lose it. But that will not bring peace. It will bring Russian occupation." - Jens Stoltenberg. He urged to give Ukraine more weapons - "The paradox is that the more weapons for Ukraine we are able to deliver, the more likely it is that we can reach a peace [solution]," Jens Stoltenberg added.

"The quickest way to end the war is to lose it. But that will not bring peace. It will bring Russian occupation." - Jens Stoltenberg. He urged to give Ukraine more weapons - "The paradox is that the more weapons for Ukraine we are able to deliver, the more likely it is that we can reach a peace [solution]," Jens Stoltenberg added.

Anton Gerashchenko

357,281 次观看 • 2 年前

“Right now as we speak, more internet is being deleted than is being created. This is the first time in history, where if we do this right, if we can preserve our wisdom correctly, you and I will be part of the first generation that will try, on a grand scale to collect wisdom and put it in a bucket and leap it forward for the next generation to grow upon”—Brian Roemmele on saving wisdom

“Right now as we speak, more internet is being deleted than is being created. This is the first time in history, where if we do this right, if we can preserve our wisdom correctly, you and I will be part of the first generation that will try, on a grand scale to collect wisdom and put it in a bucket and leap it forward for the next generation to grow upon”—Brian Roemmele on saving wisdom

Brian Roemmele

190,284 次观看 • 1 年前