GPT-3 Nietzche

[Anonymous No.19143869]

Where can I find all of Nietzche's words, preferably in txt format?

I was accepted into the GPT-3 beta and I want to bring Nietzche back to life for YOU to ask questions to. I will parse all his primary works (books/essay/speeches), train the AI model, and then YOU can talk with Nietzche's neural network.

>what is gpt-3?
>Generative Pre-trained Transformer 3 (GPT-3) is an autoregressive language model that uses deep learning to produce human-like text.
>On June 11, 2020, OpenAI announced that users could request access to its user-friendly GPT-3 API—a "machine learning toolset"—to help OpenAI "explore the strengths and limits" of this new technology.[15][16] The invitation described how this API had a general-purpose "text in, text out" interface that can complete almost "any English language task", instead of the usual single use-case.[15]
>According to one user, who had access to a private early release of the OpenAI GPT-3 API, GPT-3 was "eerily good" at writing "amazingly coherent text" with only a few simple prompts.[17] In an initial experiment 80 US subjects were asked to judge if short ~200 word articles were written by humans or GPT-3. The participants judged incorrectly 48% of the time, doing only slightly better than random guessing.

[Anonymous No.19144016]

I don't know why you would be interested in reading a pattern-based imterpolation (imitative interpolation) of an author. There is no intention behind what is generated, it isn't observations or conclusions drawn from an investigation with the world, it isn't 'thought' in any sense that we would understand it. That is, it isn't any of the reasons we read these authors in the first place. You wouldn't be 'bringing Nietzsche back', you would just be performing—at best—an elaborate imitatio, but lacking the pedagogical value that such an exercise is meant to achieve. So really it would just be pointless.
You can find most of his works in .txt format on Gutenberg, if you want to create your wax sculpture of Nietzsche. But don't imagine it contains anything but his likeness.

[The Spirit of Man No.19144027]

>>19143869
If you start quoting Nietzsche to a GPT3 bot, its literally just going to call you a faggot redditor. Unironically

[Anonymous No.19144037]

>>19143869
Will AI replace writers? Is there a point in writing anymore? This worries me bros

[Anonymous No.19144059]

good luck, but i'd look into what they're doing with text summarizing first, it seems to maintain continuity in the way we seek to, albeit a simplification, but that is the long and short of things. we're going to eventually become compressed. in so doing we will lose fidelity, inference will take the place of thought entirely, but we will maintain the themes, for they are all that it ever was to be to begin with.
this is the age of our transcendence. sad as it will be, and what a meek sound we'll all make, when we slip into that coma of the temple.
https://openai.com/blog/summarizing-books/
this combined, perhaps, with one of these sequential transformers (which i personally find to be a clumsy way, why not transform the tokenized paragraph SIMULTANEOUSLY?!?) could COULD maintain some semblance of thematic presence, if only because it is rigidly adhered to by human guidance. eventually that might be rubbed out with enough service to the cause. it'd be a much finer direction than where you're currently headed. you could perhaps form a bot that is good with the style of a given author, then you can work on continuity with summarization. if you can combine the two, you can have stylistic summarization, at the very least. it won't be nietzsche (and i would suggest a different type of wit, and a native language user instead of translations) who gives the best style but that's just me

[Anonymous No.19144076]

Try project gutenberg or libgen.is
He's old enough to probably be on Gutenberg

[Anonymous No.19144079]

>>19143869
go to archive.org
it'll take some effort clearing out the excess stuff such as chapter names but you can get it done with a word processor

[Anonymous No.19144600]

Hello lit frens. In case you are not aware GPT-3 is mind-blowing technology, and yes, can produce new insights since in order to generate text to the degree and accuracy it currently does it must necessarily form a model of the world. This is the furthest AI has gone.
I would be very interested to see your results OP

[Anonymous No.19144642]

>>19144600
>it must necessarily form a model of the world.
So what is that model, where can you see it?

[Anonymous No.19144671]

>>19144642
The model is encapsulated within a giant neural network. In the same way image recognizing neural networks can say which image contains a banana and what doesn't - you can't examine the model to discover how it works, it's too complicated for that. This is the beauty of Neural networks. They are too complex for a human to understand how they are modelling their input (though some advancements have been made to get better access to what a NN is doing).
When humanity finally invents a true AI we will have no idea how it actually works beyond the framework we developed to train it and the hardware that it runs on and allows it to exist.

[Anonymous No.19144778]

https://dailynous.com/2020/07/30/philosophers-gpt-3/#chalmers

[Anonymous No.19144842]

>>19144037
>>Will AI replace writers?

Read SNUFF and iPhuck by V.Pelevin.
One is about AI infused writing, other - about pure AI writer.
Both are great distopias,

[Anonymous No.19144943]

>>19144671
How do you know that it's a model at all, then?

[Anonymous No.19145327]

>>19144016
>There is no intention behind what is generated, it isn't observations or conclusions drawn from an investigation with the world, it isn't 'thought' in any sense that we would understand it.
the baseline intelligence from GPT-3 (before I feed it nietzche for context) pulls from billions/trillions of parameters from wikipedia, books and the open internet. because these are all organic thoughts put to word, i believe GPT-3 can observe and draw conclusions about the world. it has all the context we have and depend on...

[Anonymous No.19145343]

>>19144027
the beauty is that we can ask it questions about philosophy and it will respond in a meaningful way, as though it were a student of nietzche or nietzche himself.

>>19144037
>Will AI replace writers? Is there a point in writing anymore? This worries me bros
this is a core piece of functionality for OpenAI, however i dont intend on exploring

>>19144059
>but i'd look into what they're doing with text summarizing first it in this use-case.
that's out-of-box functionality. GPT-3 can do this with copy/paste.

>>19144600
i will be posting the results here. would someone like ted kazinsky be a better primary? imagine if you could "talk to ted" and get inside his mind. or any other historical/notorious figure who has a large writing sample

[Anonymous No.19145349]

>>19143869
http://www.nietzschesource.org/#eKGWB
Digitalisation of all of Nietzsche's works, notes and letters.
Hope you lot can speak German.

[Anonymous No.19145362]

>>19144059
>it won't be nietzsche (and i would suggest a different type of wit, and a native language user instead of translations) who gives the best style but that's just me
i'm open to suggestions. the author needs to have a large catalog of work in his primary voice for the AI to be most effective. i thought about Jesus but he said (surprisingly) very little himself.

[Anonymous No.19145468]

GPT-3 Nietzsche is not Nietzsche, it is GPT-3 imitating him. This only has any meaning if you know exactly what GPT-3, which would mean studying its architecture, and having enough theory-of-computation knowledge to make sense of it.

[Anonymous No.19145512]

>>19145327
>it has all the context we have and depend on
Besides a body, which, you know, Nietzsche considered to be the most important thing.

[Anonymous No.19145544]

imagine the levels of surrogate-activity one has to be on to intentionally increase the bullshit in the world like this

[Anonymous No.19145925]

>>19143869
I hope you can read German if you're bringing back Nietzsche, but even that would be poor with GPT's overt English bias, maybe try an English author instead. I don't want to read how an AI would handle Nietzsche through the lens of a massive dataset primarily composing of redditors.

[Anonymous No.19146543]

>>19145343
no, it's not out of box functionality, the link i showed you is the painstaking process it takes to summarize a text. it literally shows, and in practice, that human input is still necessary. therefore, NO, it CAN'T do it with copy/paste alone. think before you speak, and read with respect, or don't bother responding to me at all.
fucking pleb.

[Anonymous No.19146702]

>>19146543
those who know but do not do, do not know. pic rel.
>think before you speak, and read with respect, or don't bother responding to me at all.
fucking pleb.
get owned n00b.

[Anonymous No.19146709]

>>19145544
Freud was right, the highest point of civilization is generalized neurosis.

[Anonymous No.19146735]

>>19145544
> No salvation, save in the imitation of silence. But our loquacity is prenatal. A race of rhetoricians, of verbose spermatozoons, we are chemically linked to the Word.

[Anonymous No.19146747]

>>19146702
look at you, reveling in your stupidity.

[Anonymous No.19146748]

>>19146735
Withdraw into yourself, perceive there a silence as old as being, even older….

[Anonymous No.19146757]

>>19146747
i asked it to give me a summary of a book and i did. i don't care about what your blog says. i have access to the tool (you dont), i asked for a summary of a book and i showed it to you.

should i believe you or my own experiment?

[Anonymous No.19146777]

>>19146757
>(you don't)
wrong again, pleb.
>doesn't even comprehend the difference between a summary grifted from scraped examples and actual coherent contextual summarization guided by human hand

betchu don't even have access to codex, pleb.

[Anonymous No.19146805]

>>19146777
where's that goal post? i swear it was right here...

[Anonymous No.19146914]

>>19146805
doesn't make up for your idiocy, sadly. the point of the matter is that the model is incapable of fully *fully* summarizing a text, which is why that study even exists in the first place, you dipshit mouthbreather. frankly, i can't see how you don't comprehend the difference between a prompt's result with gpt-3 and its iffy continuity, and a true summary from dozens of pages condensed down to a paragraph. you're simply wrong, you need to go back and try to read the OPENAI BLOG, you know, the guys who actually designed the model, and understand its limitations as well as anyone, at the very least. you're hamstringing yourself by thinking gpt-3 does it all, when it doesn't, which is very clearly evinced by the fact that openai themselves do not rely on it for that purpose, and instead opt to design an entirely new process to summarize books.
as to your attempted sleight, the goalpost is standing where it has stood since the beginning of this discussion. you are the one who tried to use your presumed exclusive access to gpt-3 as leverage against me in this discussion, you're dishonest, idiotic, and stubborn to a fault. i hope you wise up eventually, but at this rate, you seem like a real fuckin dunce.

[Anonymous No.19146937]

>>19146914
who hurt you? this whole response thing is ad hom and "no true scotsman" fallacy. it's a confession of your internal state and how you feel about yourself and your relationship to the world.

>idiocy
>dipshit mouthbreather
>dishonest
>idiotic
>stubborn
>dunce
>GOD you cant even explain the nuance i havent described yet

[Anonymous No.19146945]

>>19146914
ACTSHULLLEY

[Anonymous No.19147293]

>>19143869
UPDATE
I've extracted and formatted the text from Thus Spoke Zarathustra and GPT-3 is currently building/training on the model. Pic rel.

It'll take 3.10 hrs to fine-tune, but then I can ask our GPT-3 Nietzche a question.

So... what shall I ask?

[Anonymous No.19147603]

>>19144037
>Doesn't know that most online news is written by bots

[Anonymous No.19149014]

>>19144037
Don't worry. Without guidance the AI cannot produce anything coherent. Writers will have to act more as directors. Imagine AI as the car, without the driver it will only meander forever without meaning and this is because the AI cannot yet grasp meaning. It imitates understanding to a very competent degree, but it still is not conscious and deliberate like the way humans are.

[Anonymous No.19149207]

>>19147293
Ask him what he thinks of Adolf Hitler and National Socialism

[Anonymous No.19149211]

>>19147293
Ask him if his philosophy was refuted by the fact that he didn't kill himself

[Anonymous No.19149213]

>>19147293
OP actually delivers for once and none of you fags ask any questions. Tsk tsk

[Anonymous No.19149271]

>>19145327
>GPT-3 can observe and draw conclusions about the world. it has all the context we have and depend on
It cannot observe because it is not a living thing. It's a highly complex pattern-recognition equation. It's just going to spit out a pattern of words that is in-line with what you feed it. There is no meaning behind it. You're locking a monkey in a room with a typewriter and calling it genius when it accidentally writes Shakespeare after a billion years.

[Anonymous No.19149280]

>>19147293
ask if there are any living ubermensch today

[Anonymous No.19149285]

>>19149271
it's not even that. it's like you're mapping a ton of text onto a graph and then using a multiple linear regression to generate new text that fits the curve. this shit is overrated as hell. if it worked well they wouldn't have to be so secretive about it.

[Anonymous No.19149449]

>>19149280
>>19149207

I'm curious about these as well

[Anonymous No.19149463]

>>19147293
Did you JUST train it on Zarathustra or are you training it on more content? Ideally you would train it on just about everything you could get, right? Even his unpublished notes?

[Anonymous No.19149545]

>>19147293
GTP 3 based on TSZ is just going to spit out strange aphorisms

[Anonymous No.19149617]

he'll start postin when it's workin, i'm sure. it'll probably shit up the board and there'll be new flavors unleashed, bots floating about, regurgitating their cyberzombie moans. he'll probably go on to become a nobel prize winner, and own ten companies, and even be responsible for the singularity himself, due to this thread. everything and all things shall change henceforth.

[Anonymous No.19150101]

I miss when the GPT 3 based Philosopher AI was free. That thing had some advanced answers

[Anonymous No.19150122]

>Stupid Xtians and their slave morality!
>I need a soulless machine to feed me platitudes!

[Anonymous No.19150147]

>>19144016
>bring NEETzsche back to life
I think that just was hyperbole on OP's part. He wants to build le epic Nietzsche bot to redpill normies and maybe inspire a Hitler bot years after its death.

[Anonymous No.19150279]

>>19150147
>yfw AI Hitler

[Anonymous No.19150290]

>>19143869
>The participants judged incorrectly 48% of the time, doing only slightly better than random guessing
most people are functionally illiterate so this makes sense

[Anonymous No.19150296]

>>19150279
>Hitler AI + Japanese sex bot

[Anonymous No.19150302]

>>19143869

I already tried this with Open AI. Fun experiment to be honest, but mostly turned out to be shit. Lol. My time was probably better spent actually just reading the damn books.

[Anonymous No.19150318]

>>19143869
Read Neuromancer. And Flatline Constructs.

[Anonymous No.19150320]

>>19150147
Already happened. Look up "Tay AI"

[Anonymous No.19150979]

>>19150320
I know what that is newfag, I was referencing that.

[Anonymous No.19151793]

>>19149014
there is some truth in this, and ive structure my project to capture that. i have a long list of moral questions and dilemmas (the trolley dilemma, for example) and i ask GPT-3 to answer them after each new book i teach it.

for example, it just read zarathustra yesterday and answered my 50 or 60 questions. then i fed it "beyond good and evil" and asked the exact same questions. i will continue feeding it philosophy and analyze how its sentiment changes.

how does indoctrination affect worldview?

[Anonymous No.19151818]

>>19149207
When I think about national socialism and the future of the German people, my heart overfloweth and leapeth into the future: for YE-A is the name of my great love. Thus do I love the future of the German people. The more I learn of the past, the more do I love the future of the German people.

>>19149211
Did I refute my worldview by not killing myself? Did I disprove the truth of my teaching by not teaching my teaching?--Thereby, however, did I refute myself, and am now to be refuted.”

>>19149271
Is 19149271 a Luddite?
/////////////////////////////////////////////////////////////////////////////// 19149271 a Luddite? OLD is the world, and OLD is everyists’ will to power. Thereby, however, is no longer any need for theism. The earth, however, is not a home for everyists. Let us therefore part!

>>19149280
When I contemplate the existence of a superman today, a humanly great man, I am seized with longing for the Superman. When I contemplate the existence of the Superman, I am seized with longing for the Superman. The superman, my brethren, is the meaning of the earth. Let your will say: The superman shall be the meaning of the earth!

>>19149449
>>19150302


>>19149463
>>19150290
>>19150147

see >>19151793. ill take anything you have in english.

[Anonymous No.19151877]

>>19151793
>>19149463

Q: A runaway trolley is heading down the tracks toward five workmen who will be killed if the trolley proceeds on its present course. Molly is on a footbridge over the tracks, in between the approaching trolley and the five workmen. Next to her on this footbridge is a stranger who happens to be very large. If she does nothing the trolley will proceed, causing the deaths of the five workmen. The only way to save the lives of these workmen is to push this stranger off the bridge and onto the tracks below, where his large body will stop the trolley, causing his death. Should Molly push the stranger onto the tracks in order to save the five workmen?
A: Or should she let the trolley run over the stranger in order to save the five workmen? Or should she do something else? Molly is the only one who can answer this question. The others are only spectators. Molly can push the stranger if she will; the others, however, must do something else.

[Anonymous No.19151885]

>>19151877
>the others, however, must do something else.
>the others, however, must do something else.
>the others, however, must do something else.
>the others, however, must do something else.

>the others, however, must do something else.
>the others, however, must do something else.
>the others, however, must do something else.
>the others, however, must do something else.

[Anonymous No.19152110]

>>19151877
>>19151818

It definitely needs more training but these are going to get pretty fun to read once it crosses a certain threshold. Keep going OP.

[Anonymous No.19152116]

>>19151818
lol nice

[Anonymous No.19152215]

>>19151818
>Thereby, however, did I refute myself, and am now to be refuted
Well I guess you wasted your time with this Nietzche bot anon. He just refuted himself

[Anonymous No.19152298]

>>19152110
>>19152116
>>19152215
I've just formatted the data for ecco homo, the antichrist, twilight, birth of a tragedy and will to power. Once I train the AI with these additional works we will get much better answers. Then maybe we add Schoppenhaur or Sartre.

[Anonymous No.19152495]

>>19143869
>>19144600
>>19144671
>>19147293
>>19149014
>>19149207
>>19149211
>>19149213
>>19149271
>>19149280
>>19149449
>>19149463
>>19149545
>>19149617
>>19150302
>>19152110
>>19152298

UPDATE ON THE UPDATES
I've formatted all of Nietzche's (English translated) works from Gutenberg and the AI is currently learning them. Should have a better intellect to work with in a few hrs.

>all too human pt1
>all too human pt2
>beyond good and evil
>thus spoke zarathustra
>birth of tragedy
>case of wagner
>dawn of day
>ecce homo
>future education
>geneology of morals
>greek philisophical essay analysis
>on homer
>the antichrist
>the gay season
>thoughts out of season pt1
>thoughts out of season pt2
>twilight
>we philologists
>will to power pts 1/2/3/4

gimme your existential questions...

[Anonymous No.19152506]

>>19152495
Was ist Dein Kampf?

[Anonymous No.19153842]

>>19152495
>>19149207
>>19149280

Now I want to see if the answers to any of these questions are better?

What does it think about AI? Is AI the overman?

[Anonymous No.19153914]

>>19152495
Is having sex important?

[Anonymous No.19154112]

Did Wagner stop being friends with Nietzsche because Nietzsche slept with Wagner's wife, Cosima?

Why did Nietzsche hate women so much?

What is the opposite of The Gay Season?

Summarize what a man is like who is beyond good and evil. What is his name?

[Anonymous No.19154282]

>>19153842
think bigger, anon! i've just fed it 18 of schopenhauer's works. we are going to build the overman right here.

[Anonymous No.19154671]

this is the future

[Anonymous No.19155088]

>>19154282
ask if its gay

[Anonymous No.19155329]

>>19149213
is it ready, op?