Has anyone tried the OpenAI ChatGPT?

[heymike3]

Can you write a song like Bush’s Comedown with philosophy from Nietzsche?

Here’s a song that incorporates philosophy from Friedrich Nietzsche into the style of Bush’s “Comedown”:

Verse 1:
In the shadows, I see it clear
The eternal recurrence, my fear
To live this life again, what a curse
But to will it all, that’s the verse

Chorus:
Comedown, from the mountain top
Where I see the truth, I cannot stop
The will to power, it consumes
A life that’s never-ending, my muse

Verse 2:
The weight of the world, it’s on my back
The struggle, the pain, I will attack
The Ubermensch, I strive to be
A master of my fate, I am free

Chorus:
Comedown, from the mountain top
Where I see the truth, I cannot stop
The will to power, it consumes
A life that’s never-ending, my muse

Bridge:
Am I the one, who dares to be
The one who says, “No!” to eternity
Do I have the strength, to break the chains
And live this life, without the pains

Chorus:
Comedown, from the mountain top
Where I see the truth, I cannot stop
The will to power, it consumes
A life that’s never-ending, my muse

Outro:
In this chaos, I find my peace
The will to power, it will never cease
For I am the master, of my fate
And that, my friend, is my philosophy’s gate.

[heymike3]

@Dale this confirms what you said earlier about it not accessing the internet to make its responses

How much data do use for a knowledge base?

OpenAI’s GPT-3, one of the largest language models, was trained on a diverse range of internet text, including web pages, books, and scientific articles. The exact amount of data used for training is not publicly disclosed, but it is estimated to be in the range of hundreds of billions or even trillions of words. It’s important to note that having a large amount of data does not necessarily guarantee a high-quality knowledge base, as the quality and diversity of the data used for training also play a significant role.

Do you access the internet for answering questions?

No, I do not have direct access to the internet. I was trained on a diverse range of text data, including web pages, books, and scientific articles, but I don’t have the ability to search the web or access updated information in real-time. My responses are generated based on the patterns and associations I learned from the training data, and I do my best to provide accurate and relevant information based on that training. However, it’s possible that my knowledge may be out of date or incomplete, so it’s always a good idea to verify information with other sources.

[heymike3]

How many words are contained in the average university library?

It is not possible to determine the number of words contained in an average university library as it can vary greatly depending on the size and focus of the library and its collections, as well as factors such as whether the library is primarily print or digital. Some university libraries contain millions of volumes and billions of words, while others may be much smaller.

What about the Library of Congress?

The Library of Congress is one of the largest libraries in the world and is estimated to have approximately 167 million items in its collections, including books, manuscripts, photographs, maps, sound recordings, and much more. It is difficult to estimate the exact number of words contained in these items, but it is likely to be in the billions.

[Terry_Sampson]

• I’m surprised that you haven’t queried your latest love regarding it’s ability to “snap it’s fingers” an infinite number of times, or in AI-speak, repeat the same reply an infinite number of times.

[heymike3]

I did and the results were fascinating. Let me see if I can find the post on this.

[heymike3]

Here it is:

What’s the main reason, not reasons, on why you are a Christian?

Woah!!! I think I gave the machine (ChatGPT) a run for its money: Generic question about an infinite series through successive addition, and a follow up comment confirming a program can complete an infinite number of tasks, gave me this response: “In theory, a program can complete an infinite number of tasks, as long as the tasks can be described algorithmically and the program has access to an infinite amount of resources (such as memory, processing power, etc.). In practice, ho…

[heymike3]

Does ChatGPT recognize errors in its responses and learn in real-time?

No, ChatGPT is a pre-trained language model that does not recognize errors in its responses or learn in real-time. It is trained on a large dataset and provides response based on the context and information present in that dataset. To improve its responses, it needs to be retrained on updated data.

[T_aquaticus]

@heymike3

The writer of the article confused the time of the decision and the time of consciously declaring the decision.

But that’s exactly what you have been arguing all along. You state that you can detect when you consciously decide to do something, and then do it. If this isn’t how it works, then your argument is shot.

What the authors are describing is a decision that is being made outside of conscious intent.

@heymike3

I’m confused about this experiment, is it saying someone stared at two images for 20 seconds, then pushed a button to indicate they made a decision, and then they visualize the image for 10 seconds.

Is the experiment finding the person’s choice 11 seconds before they pushed the button to indicate they made a decision?

They were already shown the images. They were asked to clear their minds, and when prompted they would mentally choose an image. Once they had consciously chosen an image they would hit a button indicated that they had made the choice. They would then concentrate on the image and report how intense the image was in their imagination.

What they were able to determine is which of the two images they would choose, and they were able to do so up to 11 seconds before they hit the button indicating that they had made a conscious choice.

[heymike3]

@T_aquaticus

What the authors are describing is a decision that is being made outside of conscious intent.

The first article confused me with the statement about consciously declaring. In hindsight I can see how that can mean pushing the button, or the choice being communicated after the 10 second visualization period.

@T_aquaticus

What they were able to determine is which of the two images they would choose, and they were able to do so up to 11 seconds before they hit the button indicating that they had made a conscious choice.

Can you decipher the second article to tell what the data set was? If the average time to indicate a choice was 5.48 sec, I’m expecting a wide range of results, as well as conflicting results like subjects whose decisions could not be predicted.

[heymike3]

Also interested to see if results correspond to IQ levels in some way.

Edit: Developmental factors may also contribute to the disparity of results, or something as simple as whether the person has given much thought to consciously acting.

Another factor which could affect experimental results is whether the person can imagine an empty space and whether it appears from their point of view in their imagination.

[heymike3]

Can you design an experimental p-zombie test?

An experimental p-zombie test aims to determine whether an entity is conscious or not. One way to design such a test is as follows:

Select a sample of entities, including both human participants and artificially intelligent systems, to be tested.

Create a set of tasks that measure the entities’ responses to stimuli that are known to evoke conscious experiences. For example, tasks could include responding to painful stimuli, recognizing emotional expressions, or making moral judgments.

Administer the tasks to each entity and measure its responses.

Analyze the responses to determine whether they are consistent with conscious experience. For example, if an entity responds to a painful stimulus in the same way that a conscious human would, it would be considered as having conscious experience.

Repeat the process with a large enough sample size to increase the reliability and validity of the results.

It is important to note that while such a test may provide some evidence of conscious experience, it cannot definitively prove that an entity is not a p-zombie, as it is a philosophical concept and not a scientific one.

[heymike3]

@T_aquaticus

What the authors are describing is a decision that is being made outside of conscious intent.

For the record, I went through Susan Blackmore’s short but dense book Consciousness: A Very Short Introduction, and am well aware of the odd things our minds do. I have no doubt that at certain times for different people, our minds can fool us. I had that experience of driving the car while my mind wasn’t present. And no it was not a Tesla on a California freeway. But my experience of it was not an illusion, at least not a total illusion, because we do agree the observable phenomena generally follows the intention. Now there may be a real issue in the synchronization of the intention and experience that I’m sure will correct itself very quickly with more testing and observing or use at this particular level.

Plus, it may be as the scientists probe deeper into this unconscious decision making process that presents itself at times, something which they are only scratching the surface of, they may find teasers for a real possibility of quantum interaction. Now that would be something!

[Dale (evolutionary providentialist)]

@heymike3

For example, tasks could include responding to painful stimuli…

@heymike3

Administer the tasks to each entity…

What stimuli could cause pain to a bot entity? Too many facetious or stupid questions?

[heymike3]

A got the impression from the program’s response that a p-zombie test among humans would be controversial. If determinists keep trying to determine consciousness is an illusion, I don’t think they are prepared to consider that some people may be more conscious than others. I have no interest in going there, but the results of this experiment, if confirmed, will inevitably raise the question.

[Terry_Sampson]

• Silly me. Either I didn’t notice that post or forgot it in the appropriate move of several items to their own, new thread, i.e. “Religious Neutrality and Philosophical/Scientific Theories.”
• ChatGPT’s response to your query was noteworthy, IMO, to wit: "“The transition from a finite number of tasks to an infinite number of tasks is not a well-defined concept, as it depends on the definition of “finite” and “infinite.” So, the transition from a finite set of numbers to an infinite set of can’t be done.

[heymike3]

What if the ability to conceptualize this impossibility were tied to experimental “tasks that measure the entities’ responses to stimuli that are known to evoke conscious experiences”?

What a revival this could be for Scientology!

[Terry_Sampson]

@heymike3

What if the ability to conceptualize this impossibility were tied to experimental “tasks that measure the entities’ responses to stimuli that are known to evoke conscious experiences”?

• You’re funny!
• Speaking of “experimental tasks”, back in the now-defunct “Jordan Peterson book 12 rules for life”, I proposed taking the Paramecium as a model for living in a chaotic world:

• Lo and behold, Frenchman Romain Brette is “a research director in computational and theoretical neuroscience in the Vision Institute, where [he leads] the Computational neuroscience of sensory systems group (CV). These days, I am working on 1) biophysics of action potentials, 2) epistemology of neuroscience and 3) integrative modeling of Paramecium, the “swimming neuron”.”
• “Integrative modeling of Paramecium”? LOL! Right up your “search for consciousness” alley, no?
• By the way, I think it’s marvelous that someone as “popular” as David Bentley Har,t, “a strong theist” speaks of the bond between “consciousness” and the divine [shades of Roy Clouser] in David Bentley Hart - The nature of consciousness [9:00] and, in another DBH and Closer to Truth video that I can’t find at the moment, the teleological “purposeful” behavior of Paramecia.
• Just think of it: a single cell protist like the paramecium, which can be called “a living thing” that has purposeful behavior and … consciousness, albeit rudimentary/basic as it is.

[heymike3]

@Terry_Sampson

“Integrative modeling of Paramecium”

The wonder never ceases… which kind of ties into something I stumbled upon yesterday when looking up the phrase “world without end” - “in saecula saeculorum” and found David Bentley Hart has written about the ambiguity in the Greek of Philippians 4:20 which may mean world or age without end:

“Much depends, naturally, on how content one is to see the Greek adjective αιωνιον, aionios, rendered simply and flatly as “eternal” or “everlasting.” It is, after all, a word whose ambiguity has been noted since the earliest centuries of the church.”

But as is so often the case, human beings have a hard time distinguishing between the possibility of a sequence proceeding to infinity with the impossibility of it becoming actually infinite.

[Terry_Sampson]

Curiosity led me to order his version of the New Testament. When it comes, I intend to order DBH’s That All Shall Be Saved: Heaven, Hell, and Universal Salvation and Atheist Delusions: The Christian Revolution and Its Fashionable Enemies.

• Regrettably, DBH is too “dense” for me to follow in Youtube, without downloading a transcript or slowly working my way through a written version of what he’s saying (or reading aloud, which–annoyingly–he does from time to time (in his prepared lectures to an audience.
• Remarkable connection, IMO, between what he says about his version of consciousness and some of what Clouser says. Remind me to ask Roy if he’s familiar with DBH.
• Unfortunately, DBH, despite his erudition, is one of those who speaks of God as if He can be “outside of Time and Space”, which I have historically rejected.

[heymike3]

That will be a good question to ask Clouser