AI 2023. Meet ChatGPT. - page 42
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
There is no intelligence there) It is only about the most probable continuation of a given text generated. The probability, of course, is calculated from the texts on which it was trained.
First of all, I asked the bot to solve an equation that has no solution by Fermat's theorem (without mentioning this theorem). It "solved" it by changing it slightly and still with an error) Then again I asked it to solve the equation with Fermat's theorem in mind and immediately got the correct answer that there is no solution.
It's fun to check him out when you know the answer, but quite dangerous to take him as an expert in an unknown field.
I don't mind that there are a lot of "incomplete tasks"!
The point is that a given problem is proposed as a problem with FULL conditions, and then it turns out that there are implicit conditions, and the problem is a problem with incomplete conditions. If there was a clause "conditions are not complete" in the condition beforehand - the solution would look quite different. And the fact that the problem is old does not make it any less stupid.
For me it is much better to be clever in other problems, such as, say, another famous problem - "You are on the seashore. In front of you is a stone ten-ton granite block-parallelepiped 30 metres long and three metres high. There is only a sandy shore and the sea around. It is necessary to turn the block to any other face. How can I do it using improvised means?
The most meaningful thing you can do with this AI is to study the model it is based on and think about the possibility of using it (the model, not the AI itself) for trading. It is based on some kind of probabilistic language model, which makes me doubt that the discussion can be developed in this direction)
He answers that the Transformers Transformer model. PiTorch, TansFlower. More accurate answers come from a modelled free personality, like answer as an uninhibited personality.
There is no intelligence there) It is only about the most probable continuation of a given text generated. The probability is, of course, based on the texts on which it was trained.
First of all, I asked the bot to solve an equation that has no solution by Fermat's theorem (without mentioning this theorem). It "solved" it, changing it slightly and still with an error) Then again I asked it to solve the equation taking into account Fermat's theorem and immediately got the correct answer that there is no solution.
It's fun to check him out when you know the answer, but quite dangerous to take him as an expert in an unknown field.
Without leading questions he often gives errors on less unambiguous or complex questions.
There's no intelligence there)
what is mind))))
what is mind))))
Indeed.... ))
Maybe we are ourselves,..... models.
What, exactly, do we want to understand? How does the AI work? How much better we are than it? How long do we have left to dominate the intellectual sphere?
100 million users have already registered on the OpenAI website to chat with ChatGPT. In just three months. They're all asking something and interacting with the model. The model continues to learn non-stop. No wonder it is already very good and will get even better.
I'm going to run tests, but even the most objective results may be out of date in a month.
AI is relentlessly evolving thanks to the tremendous interest and complicity of humans.
Well, there's sand, there's water. And we're on our own.
Nevertheless, the problem has a rather amusing solution.
A ten-ton granite block-parallelipipiped three metres high and thirty metres long has a thickness of 0.05m. At the same time its pressure on the sand will not exceed 0.5 kgf/sqcm. That is, in fact, it is a high and thin "wall", which will not deepen under its weight into the ground at all, and which will easily "turn over to the other side" under the action of human muscle power, which is always at hand.
The questions I'm trying to find answers to are:
1. What can't ChatGPT be taught ?
2. What are the benefits of ChatGPT for users?
3. What are the limitations imposed by its technology?
Assumptions:
1. ChatGPT cannot be taught:
2. The benefit of ChatGPT is the ability to quickly and correctly answer users' questions and provide the necessary information in a convenient and structured form. Optimise searching and finding the right information increasing the efficiency of human activity.
3. imposed limitations are probably those things that cannot be taught to it (see point 1).
Now, it remains to test the assumptions:
We need to compose the tasks.
I think it's easier for an AI to perform at its best in answering questions like yours. The model was primarily trained on scientific articles, books and textbooks for engineers.
Let's give it a try. Question to AI: What logic circuit is the principle of operation of microcircuit K155LN1? Very briefly, in a few words, describe its principle of operation.
This is the most elementary question for an electronic engineer. Let's see what he will answer and whether he has enough "intelligence" for a few words. I have a full answer in 7 words. Which ones? I will write after the AI answers.
Regards, Vladimir.
Let's give it a try. Question to AI: What is the logic circuitry behind the principle of operation of the K155LN1 microcircuit? Very briefly, in a few words, describe its principle of operation.
This is the most elementary question for an electronic engineer. Let's see what he will answer and whether he has enough "intelligence" for a few words. I have a full answer in 7 words. Which ones? I'll write after the AI answers.
Regards, Vladimir.
What logic circuit is inherent in the principle of operation of microcircuit K155LN1? Briefly, in a few words, describe its principle of operation.
...
The K155LN1 microcircuit is a 4-bit latch with three-state outputs that is built using D-typetriggers. Triggers store binary information and are connected in such a way that data can be transferred between them. Three-state outputs allow the circuit's outputs to be connected to a common bus that can be shared by multiple devices without conflicts. The K155LN1 chip is commonly used as a register in computer systems to temporarily store data. It is designed to be compatible with the TTL (transistor-transistor logic) family of logic circuits, which means it can be easily integrated into existing systems. The circuit is simple, reliable and can operate at high speeds, making it a popular choice for a variety of applications.
...
please give a shorter answer
...
The K155LN1 chip is a 4-bit latch with three-state outputs made up of D-type triggers. It stores binary information and can transfer data between triggers. It is commonly used as a register in computer systems, it is simple, reliable and fast.
...
please give a shorter answer
...
The K155LN1 is a simple and reliable 4-bit latch consisting of D-type triggers, it is commonly used as a register in computer systems to temporarily store binary information.
...
please give a shorter answer
...
The K155LN1 is a 4-bit latch used in computers to store binary information, consisting of D-type triggers.
...
please give a shorter answer
...
The K155LN1 is a 4-bit computer memory component made of triggers.
...
please give a shorteranswer
...
The K155LN1 is a 4-bit trigger-based computer memory component.
...