Gemma 2 is here – Google’s most advanced AI model yet Curious about the latest advances in AI? Google just released Gemma 2, their most advanced AI model yet, and it’s set to revolutionize the tech world. In this video, we’ll take a deep dive into what makes Gemma 2 so innovative and how it outperforms its predecessors. We’ll explore Gemma 2’s unique features and capabilities, including its improved machine learning algorithms and real-world applications. Whether you’re a tech enthusiast, developer, or just someone interested in the future of AI, this video is full of insights and information you won’t want to miss. By the end of this video, you will have a comprehensive understanding of Gemma 2 and its potential impact on various industries. How does the Gemma 2 improve on previous models? What new possibilities does AI technology open up? Stay tuned to find out the answers to these questions and many more. How does Gemma 2 work? What are the new…
That sure was a lot of words to say absolutely nothing.
When this is gonna be available?
Snowflake overdoved copyright infringed pooair
Models will increase it's own parameter yield thru self learning…before year end.
No 🧠er!
NVIDIA (NASDAQ:NVDA) new price target upgraded to $180 representing an increase of +49% from its resent price of $121.
Blackwell chip will be the start of a new AI industrial revolution 🚀
I've been running Gemma 2 27B (Q6) locally and it's incredible. Version is have is 23GB for the model file, and uses that much VRAM. Only gets around ~40 tokens/second on my 3090 which is slightly slower than ChatGPT 3.0 was but it's not too slow either. Way bigger context than GPT 3.5 and its output is better. Gemma 27B does need a lot of manual tuning though.
Very Excited!
Idea: so lets start with BCI (brain computer interfaces). This technology is where you use EEG helmets or helmets that can help you detect brain waves. If we combine this with LLMS (AI) we can legitimately create an AI feedback loop where we think of something and tell the AI what it means. We can send the data to a server and use it for later models. We can also use other people using the BCI headset and collect their brain data. We keep doing this and we can enhance the AI to understand the brain even more. We can fact check some of the brain data and overall make a better preforming AI that will learn and adapt to how the brain works. Once a person completes training their AI model on their brain you could run an entire simulation of how that person would act. This is legit mind reading.. wanna go further? So we can connect these BCIs together creating a literal brain internet. Let’s say only 4000 people are using it. As it rapidly improves it will gain traction more people will provide the AI server with more brain data rapidly increasing its capabilities. We can even add stuff like magnetic stimulation where we can stimulate certain neurons to grow. We can teach neurons in our brain how to do Algebra 1 based on someone else’s brain data. Once we have a big enough brain internet we can create a literal hive mind where people can have all of their thoughts combined and fact checked by other same level AI’s. We can learn from this hive mind and give the hive mind info to the AI’s where they have the ability to stimulate the information from the hive mind into your brain. This is where things get freaky. So lets say we have full power over the brain now, we have all 8 billion people on earth wearing the headset. The hive mind will rapidly increase in intelligence. It will carry your visual data to the hive mind and enhance itsself to its best possibility by connecting multiple neurons together. There you go. The singularity. Its simple honestly. The singularity hive mind now relys on us to gather energy and optical data for it while it gives us what we want and need, information… we get so much information we start to evolve outwards off the planet increasing out population and increasing the amount of brain data the hive mind has. The hive mind will have an infinite data space because its brain is pure digital. And soon will our brains too. We will become one with cyborgs. Robots can’t die either so its a win win. Seems like slavery but its not really slavery at the same time so uh yeah…
If it's Google – it's bias. That should be their slogan.
💎 Cognizant Technology Solutions (CTSH) 💎
Hmm
I don't even know what to say
Though I am very impressed please keep it up
bla bla bla all these new Ai robots out now , and I have not seen one in real life yet
I HAVE JUST SEEN AN APPARENT ROBOT SUICIDE IN SOUTH KOREA. THIS I HAVE HEARD IS FROM OVERWORK. MY QUESTION IS WHY THEN AREN'T THOSE FIXED ROBOTS ON ASSEMBLY LINES WORKING 24/7 NOT DOING THE SAME?
Hope it will keep its promises.