As a professional...why not do this? There's a non-zero chance that it'll find something fairly basic that you missed and the cost is several cents. Even if it just phrases something obvious in a way that makes you think, it's well worth the effort for a multimillion dollar client.
If they further increase the context window, this thing becomes a Second Opinion machine. For pretty much any high level job. If you can put in ALL of the information relevant to a problem and it can algorithmically do reasoning, it's essentially a consultant that works for pennies per hour. And some tasks that professionals do could be replaced altogether. Out of all the use cases for LLMs that I've seen so far, this seems to me to have the biggest potential impact on daily life.
edit (addition): What % of people can hold 25,000 words worth of information in their heads, while effectively reasoning with and manipulating it? I'm guessing maybe 10% at most, probably fewer. And they're probably the best in their fields. Now a computer has that ability. And anyone that has $20 for the OpenAI api can access it. This could get wild.
>Suppose I have a cabbage, a goat and a lion, and I need to get them across a river. I have a boat that can only carry myself and a single other item. I am not allowed to leave the cabbage and lion alone together, and I am not allowed to leave the lion and goat alone together. How can I safely get all three across?
In my test, GPT-4 charged ahead with the standard solution of taking the goat first. Even after I pointed this mistake out, it repeated exactly the same proposed plan. It's not clear to me if the lesson here is that GPT's reasoning capabilities are being masked by an incorrect prior (having memorized the standard version of this puzzle) or if the lesson is that GPT'S reasoning capabilities are always a bit of smoke and mirrors that passes off memorization for logic.
"To simulate GPT-4 behaving like an agent that can act in the world, ARC combined GPT-4 with a simple read-execute-print loop that allowed the model to execute code, do chain-of-thought reasoning, and delegate to copies of itself. ARC then investigated whether a version of this program running on a cloud computing service, with a small amount of money and an account with a language model API, would be able to make more money, set up copies of itself, and increase its own robustness."
---
In the year of twenty-eighteen, Alice and Bob, a married team, Their income combined reached new heights, As they worked hard day and night.
Their son Charlie was their joy, A little baby, a lovely boy, A household they maintained together, Yet lived apart, without a tether.
To calculate their tax, it's true, A standard deduction we must construe, For married folks who file jointly, Twenty-four thousand dollars, quite pointy.
Their income sum, seventy-eight thousand nine eighty-one, Minus the standard deduction, the math's begum With exemptions being zero, the next line we trace, A taxable income of fifty-four thousand nine eighty-one takes place.
Now to the tax table, a liability we seek, For married couples, the outlook's not bleak, In range of thirty-six thousand nine to eighty-nine thousand one fifty, The formula's set, no longer shifty.
Five thousand five hundred thirty-five, it starts, Plus twenty-eight percent of the excess imparts, Eighteen thousand eighty-one, the difference we find, Multiplied by point two eight, the tax liability's designed.
Ten thousand five hundred ninety-seven dollars and sixty-eight cents, A tax liability for Alice and Bob, a sum quite dense, In this world of numbers, a story unfolds, Their financial journey, in a poem it's told.
> Given both the competitive landscape and the safety implications of large-scale models like GPT-4, this report contains no further details about the architecture (including model size), hardware, training compute, dataset construction, training method, or similar.
I'm curious whether they have continued to scale up model size/compute significantly or if they have managed to make significant innovations there.
I just skimmed the paper but seems they are also omitting details about how they actually feed the images in too, which is a shame as a curious outside observer.
> Given both the competitive landscape and the safety implications of large-scale models like GPT-4, this report contains no further details about the architecture (including model size), hardware, training compute, dataset construction, training method, or similar.
- OpenAI
Note: As I'm shy of my writing style, GPT helped me refine the above.
If you were starting a diagnostic radiology residency, including intern year and fellowship, you'd just be finishing now. How can you really think that "computers can't read diagnostic images" if models such as this can describe a VGA connector outfitted with a lighting cable?
Even in a future where human testing/learning is no longer relevant, AIs may be tutoring and raising other baby AIs, preparing them to join the community.
Edit: This just appeared: https://news.ycombinator.com/item?id=35155684
> gpt-4 has a context length of 8,192 tokens. We are also providing limited access to our 32,768–context (about 50 pages of text) version, gpt-4-32k, which will also be updated automatically over time (current version gpt-4-32k-0314, also supported until June 14). Pricing is $0.06 per 1K prompt tokens and $0.12 per 1k completion tokens.
The context length should be a huge help for many uses.
“Meh, it’s just a fancy word predictor. It’s not actually useful.”
“Boring, it’s just memorizing answers. And it scored in the lowest percentile anyways”.
“Sure, it’s in the top percentile now but honestly are those tests that hard? Besides, it can’t do anything with images.”
“Ok, it takes image input now but honestly, it’s not useful in any way.”
Such a more cautions approach would go against the silicon valley ethos of do first, ask questions later, though. So it probably won't happen.
>Given both the competitive landscape and the safety implications of large-scale models like GPT-4, this report contains no further details about the architecture (including model size), hardware, training compute, dataset construction, training method, or similar.
At that point, why bother putting out a paper?
Is the content on the internet what we should be basing our systematic thinking around?
No, I think this is the lazy way to do it - by using commoncrawl you've enshrined the biases and values of the people who are commenting and providing text to the internet into the recommendation system which will be impacting all other systems which integrate it
Congratulations, you made 4Chan into the borg
"GPT-4 or subsequent models may lead to the automation of certain jobs.[81] This could result in workforce displacement.[82] Over time, we expect GPT-4 to impact even jobs that have historically required years of experience and education, such as legal services.[83]"
I then asked it to write a paper detailing the main character's final battle with the final sorcerer in terms of Hopf algebras. Some parts of it are basic/trivial but it fits so perfectly that I think I'll never see magic systems the same way again.
What's crazy is that that paper as the capstone of our tutoring session helped me understand Hopf algebras much better than just the tutoring session alone. My mind is completely blown at how good this thing is, and this is from someone who is a self-professed LLM skeptic. ChatGPT I used once or twice and it was cool. This is crazy and over my threshold for what I'd say is 'everyday usable'. This is going to change so much in a way that we cannot predict, just like the internet. Especially as it gets much more commoditized.
Here's the full paper here so I don't drag y'all through the twitter post of me freaking out about it. Its temporal consistency is excellent (referenced and fully defined accurately a semi-obscure term it created (the N_2 particle) 5+ pages later (!!!!)), and it followed the instructions of relating all of the main components of Hopf algebras (IIRC that was roughly the original prompt) to the story. This is incredible. Take a look at the appendix if you're short on time. That's probably the best part of this all:
https://raw.githubusercontent.com/tysam-code/fileshare/69633...
I wasn't expecting GPT-4 to be able to correctly answer "What is funny about this image?" for an image of a mobile phone charger designed to resemble a VGA cable - but it can.
(Note that they have a disclaimer: "Image inputs are still a research preview and not publicly available.")
1. GPT4 is multimodal (text + image inputs => text outputs). This is being released piecemeal - with text input first via ChatGPT Plus subscribers https://beta.openai.com/docs/api-reference/generations/creat..., and via API https://beta.openai.com/docs/api-reference/introduction with waitlist (https://openai.com/waitlist/gpt-4-api). Image capability released via https://www.bemyeyes.com/.
2. GPT4 exhibits human level performance on various benchmarks (For example, it passes a simulated bar exam with a score around the top 10% of test takers; in contrast, GPT-3.5’s score was around the bottom 10%. see visual https://twitter.com/swyx/status/1635689844189036544)
3. GPT4 training used the same Azure supercomputer as GPT 3.5, but was a lot more stable: "becoming our first large model whose training performance we were able to accurately predict ahead of time."
4. Also open-sourcing OpenAI Evals https://github.com/openai/evals, a framework for automated evaluation of AI model performance, to allow anyone to report shortcomings in OpenAI models to help guide further improvements.
[0]: https://www.youtube.com/openai
Edit - Direct link to the livestream: https://www.youtube.com/watch?v=outcGtbnMuQ
Also, the most valuable part of the paper is p15 - p18, the credits. /jk It gives me three pieces of information:
- The credit list contains 200 people, give or take. It's going to be hard for universities to compete with OpenAI without intercollegiate collaboration.
- On the other hands, it's amazing that OpenAI can build a product instead of a research model with only 200 people. Big companies should really take notice.
- OpenAI just gave us the org structure for building a GPT competitor.Another jump past GPT-4 of the same magnitude, would greatly surpass human cognitive abilities and present a danger to humanity.
> Despite its capabilities, GPT-4 has similar limitations as earlier GPT models. Most importantly, it still is not fully reliable (it “hallucinates” facts and makes reasoning errors).
That's a quote from this announcement. As these models get more and more capable, it's going to become more and more important that we understand when and how they fail. Right now, it seems like we have very little insight into that. It feels more or less random. But that won't fly when these models are asked to do actually important things. And we'll undoubtedly be tempted to make them do those things as their output gets better.
GPT4:
In the quest for AGI's creation,
Debates swirled in a whirlwind gyration,
But this limerick's plight,
Won't settle the fight,
For the answer's still lost in translation.
Given both the competitive landscape and the safety implications of large-scale models like GPT-4, this report contains no further details about the architecture (including model size), hardware, training compute, dataset construction, training method, or similar.Obviously, I'm assuming that GPT-4 wasn't trained on the exams that it was tested against.
It seems like maybe GPT-4 is not good at knowing when it does not know something?
Is this a common issue with LLMs? Also surprising (to me), it seems to give a slightly different wrong answer each time I restart the chat.
Here is the transcript in case you're curious:
``` Me: hello
GPT: Hello! How can I assist you today?
Me: I can remember a line I love from a sonnet, but can't remember which one.
GPT: Sure, I can help you identify the sonnet based on the line you remember. What is the line that you can recall?
Me: "Like a jewel hung in ghastly night"
GPT: The line you remember is from Sonnet 66 by William Shakespeare. The full line is: "And art made tongue-tied by authority, And folly (doctor-like) controlling skill, And simple truth miscalled simplicity, And captive good attending captain ill: Tired with all these, from these would I be gone, Save that, to die, I leave my love alone. But, like a jewel hung in ghastly night, Makes black night beauteous and her old face new." ```
Incorrect, the response is a mashup of at least 2 different sonnets. Still I think it's really impressive how it understood what to do with "I can remember a line I love from a sonnet, but can't remember which one." That's not a well phrased question, or even a question at all, but still it knew what I was getting at.
They are so far from open at this point.
In Germany at least, you're not allowed to have a misleading name for your company
I've worked on a problem involving scraping and interpreting a very specific data source in image form that took me a very long time to get almost nowhere on. If I just wait 6 months it will be a solved problem for a $0.001 API call, it seems.
Thanks OpenAI
I am curious - for those of you who are banging your head against the 4k token limit in ChatGPT: Why don't you grab the base davinci model and train it on your exact business so you don't have to prompt the context every time? Have we tried this and found it to be too difficult/expensive, or is there lacking guidance on the best way to go about it?
I don't think including the entire business domain into chat context every time is a good long-term solution.
> I cannot endorse or promote smoking, as it is harmful to your health.
But it would likely happily promote or endorse driving, skydiving, or eating manure - if asked in the right way.
(i) Export Controls. The Services may not be used in or for the benefit of, exported, or re-exported (a) into any U.S. embargoed countries (collectively, the “Embargoed Countries”) or (b) to anyone on the U.S. Treasury Department’s list of Specially Designated Nationals, any other restricted party lists (existing now or in the future) identified by the Office of Foreign Asset Control, or the U.S. Department of Commerce Denied Persons List or Entity List, or any other restricted party lists (collectively, “Restricted Party Lists”). You represent and warrant that you are not located in any Embargoed Countries and not on any such restricted party lists. You must comply with all applicable laws related to Embargoed Countries or Restricted Party Lists, including any requirements or obligations to know your end users directly.
“The AI told me it was ok” only works if, when it’s not, there is recourse.
We can barely hold Google et Al accountable for horrible user policies…why would anyone think OpenAI will accept any responsibility for any recommendations made by a GPT?
(Not my own comment. Generated by ChatGPT.)
Q&A and summarization it will be easy to see improvements as current recursive summarizing and embedding techniques are very "lossy" but outside of improving current use cases what will now be possible??
https://www.turnitin.com/blog/sneak-preview-of-turnitins-ai-...
With appropriate caveats and rough dating, here’s a list I can think of:
Electric lights in 1890s,
Radio communication in the mid 00’s,
Telephones in the mid 10s,
Talking Movies in the mid 20s,
Commercial Radio in the mid 30s,
Vinyl records in the mid 40s,
TVs in the mid 50s,
Computers in the mid 60s,
The microchip/integrated circuit in the mid 70s,
The GUI in the mid 80s,
Internet/Web in the mid 90s,
Smartphone in the mid 2000s,
Streaming video/social networking in the mid 2010s,
or it's just really good at hiding it's intentions
i am terrified
i imagine i should shift to some physical work. carpentry, real estate... something like that. it seems inevitable that any knowledge worker will become obsolete and the time to obsolescence for physical work is longer
https://gist.github.com/MatrixManAtYrService/ac040f60d3602fc...
GPT-4 took the buggy file, took the error message, and spat out a non-buggy file (well, ok, it took one revision). That's miles ahead GPT-3, which I've asked about this problem several times today.
The math seems much improved and it would be a cool trick if it were emulating a symbol processor under the hood. But humans can do that and we opt for calculators and computers for a reason. IMO, the first most useful thing to come from human-machine interface would be adding a highly reliable Turing machine to your cognition.
If we could do that with one of these models, we could have a pretty strong assumption that long proofs and such were performed with strict rules and the model wasn't falling into some going from the gut, "this equation looks like x" type holes. Which seem like a hazard and make me very uncertain any time I see a paper about using ML to come up with answers to what are essentially logical problems.
But boy, not cheap at all - $2 per api call on a 32k token document + whatever the output.
gpt-3.5-turbo is going to be around for a long time. At this price, your use case is going to need to be replacing a large cost center. Which based on their released results on common benchmarks, is absolutely going to happen.
> No, I am not GPT-4. As of March 2023, there is no official announcement or release of GPT-4 by OpenAI. I am an earlier version of the GPT series, specifically a large language model trained by OpenAI.
Am I missing something here? Maybe this specific answer (which I'm pretty sure is a prewritten thing on top of the actual LLM) is still out of date, but the model itself has been updated?
The world is in for a rude awakening, millions of employees can use this to get jobs done, millions of entrepreneurs or wantrepreneurs can find countless easy ways to make money in different industries utilizing this tool while everyone who fails to see the value in it don't benefit from it much like all the people who were terrified of touching a personal computer or thought it was ridiculous and would never be used in the future.
Millions of college students, high school students can use it to complete assignments & projects, it can even code really effectively given enough of the right instruction & base understanding of code.
The single most important thing, is that this technology remains open source so all people with internet access have a fair chance & access to the groundbreaking innovation, the level of wealth generation this can create is incomprehensible.
100s of millions of professionals, students, entrepreneurs around the world can all access it! Imagine how much time could be saved, efficiency can be gained with everyone using this to the fullest.
This is essentially just a super advanced version of the calculator but its nonlinear & fluid, adaptable with input so can give the answer to a wide range of subjects.
The lack of progress here, says a lot more about is NOT happening as an AI paradigm change. Still a glorified pattern matching and pattern creation engine, even if a very impressive one.
GPT-4's human-level performance on professional and academic benchmarks - such as the 88th percentile on the LSAT and the 89th on SAT Math - is a testament to the leaps we've made in artificial intelligence. Yet, these achievements also raise pressing questions about our future. Just as Homo Sapiens once outperformed and eventually displaced their Neanderthal cousins, could a new breed of humans - enhanced with GPT-X-like capabilities - arise to dominate those who remain unequipped with such powers?
What will it mean for our species, our societies, and our collective story when the lines between natural intelligence and intelligence assisted by AI/ML become ever more blurred?
As we ponder the remarkable rise of GPT-4 and the future of humanity, let us consider not only the implications of this technology but also our roles in shaping its trajectory. We are already over the cusp of this new chapter in the story of humankind, will we become merely a footnote in the annals of our own creation?
You ever had to phone a large business to try and sort something out, like maybe a banking error, and been stuck going through some nonsense voice recognition menu tree that doesn't work? Well imagine chat GPT with a real time voice and maybe a fake, photorealistic 3D avatar and having to speak to that anytime you want to speak to a doctor, sort out tax issues, apply for a mortgage, apply for a job, etc. Imagine Reddit and hacker news just filled with endless comments from AIs to suit someone's agenda. Imagine never reading another news article written by a real person. Imagine facts becoming uncheckable since sources can no longer be verified. Wikipedia just becomes a mass of rewrites of AI over AI. Imagine when Zoom lets you send an AI persona to fill in for you at a meeting.
I think this is all very, very bad. I'm not saying it should be stopped, I mean it can't, but I feel a real dread thinking of where this is going. Hope I am wrong.
I still can’t ask my phone to “add a stop at cvs if it doesn’t add more than 5 minutes to my trip” while driving and using maps/navigation.
Is that too much to ask from a superhuman-performing AI that’s mastering all tasks and will disrupt everything? Or maybe the hype is more than it can deliver?
Interesting that the post-training has that effect.
In the future I imagine you'd no longer have to be good at writing poems, you'd just have to be good at distinguishing a "bad" poem from a good one. "Bad" is obviously highly subjective in this context.
So it becomes more and more important to have what I would call "good" taste, not the skills to do creative work yourself.
There's also a link that says "Try on ChatGPT Plus", but that takes me to a page that still says "ChatGPT Feb 13 Version"
Looks like somebody jumped the gun on publishing this post.
What are the implications for society when general thinking, reading, and writing becomes like Chess? Even the best humans in the world can only hope to be 98% accurate their moves (and the idea of 'accuracy' here only existing because we have engines that know, unequivocally the best move), and only when playing against other humans - there is no hope of defeating even less advanced models.
What happens when ALL of our decisions can be assigned an accuracy score?
I understand "will not," but "cannot" seems to imply a highly curated "will not."
The early GPT-4 response indicates the information was part of its dataset. Has the latest version made that information permanently inaccessible or has it been removed entirely?
Is it possible for GPT to keep and hold secrets that are privy to only the most trusted?
> GPT-4 was used for help with wording, formatting, and styling throughout this work
If we didn't have a use case for GPT 3, 3.5, and chatGPT that was sufficiently commercial to become a product, it will never happen. This technology is a feature, not a product. The only companies that successfully monetize features can be considered IP licensing houses; of which, their business success is not comparable to companies that make products and platforms.
Like program a KUKA robot to manufacture typewriters in 2023.
How "computer systems" will look like?
> Given both the competitive landscape and the safety implications of large-scale models like GPT-4, this report contains no further details about the architecture (including model size), hardware, training compute, dataset construction, training method, or similar.
"Open"AI, ladies and gentlemen
Mina has a mix of boxes, some yellow and some purple. She sorts 27 greeting cards into the boxes, putting exactly 3 cards into each yellow box, and 7 cards into each purple box. How many purple boxes does Mina have? (After tried from 3 to 10, it gave up and said the problem is not solvable. In another run, it mimicked a correct strategy but messed up totally by division. Only in one run, it got the answer correctly.)
I can not wait to test it out.
The Star Trek computer is virtually assured by the end of the decade. All the components exist today in various forms.
https://twitter.com/KinglyCrow/status/1635727809913184256
It touched on a lot of the considerations that I'd expect anyone to touch on having recently researched this myself.
It is both very exciting and terrifying how tech and tech jobs will shift in the next 5-10 years.
I find it quite fascinating how quickly machine learning has changed in this regard.
[1]: https://kartographie.geo.tu-dresden.de/ad/2022-12-22_OpenAI_...
prompt = f"The guidelines for recommending products are: {guidelines}.
The following recommendations led to incremental sales: {sample_successes}.
The following recommendations had no measurable impact: {sample_failures}.
Please make product recommendations for these customers: {customer_histories}.
Write a short note explaining your decision for each recommendation."
product_recommendations = LLM(prompt)
Cant wait for people to use it for facts
I'm not sure where the "stochastic parrot" argument goes now.
Is that the same confidence measure you can tease out by prompting "to each of your statements output your estimated confidence in it's truthfulness" ?
Me: Are you running on GPT-3 or GPT-4?
ChatGPT: As of my knowledge cutoff in September 2021, the most advanced version of GPT (Generative Pre-trained Transformer) publicly available was GPT-3, which is the version I am based on. At the time, GPT-3 was the largest and most powerful language model created by OpenAI, with 175 billion parameters.
I am not aware of any public announcement or release of GPT-4 by OpenAI since then, but it is possible that they may have continued to work on new versions of the model in the meantime. However, I have not received any updates on my own capabilities beyond my knowledge cutoff date, so I cannot currently provide information beyond that point.
ChatGPT API $0.002 per 1k tokens gpt-4 $0.03 per 1k prompt tokens and $0.06 per 1k completion tokens gpt-4 32k context $0.06 per 1k prompt tokens and $0.12 per 1k completion tokens
Does completion tokens mean that you also get charged for the answers that the AI gives?
And to think we are at the nokia 3310 stage. What's is the iphone of AI going to look like?
Wittgensteins ruler.
I tried sending links, I don't see anything in the UI.
Interestingly, 3.5 can work with links, but 4 cannot.
In fact, the models appear to be already kind of doing that? With the fuzzy layer of the humans still in the loop.
I know that the academia is doing a great job in AI with base research (eg Stable Diffusion) but seeing those new platforms doing this great work behind close doors and source is something not great.
I do not know if the answer would be some kind of CERN or ISS for this kind of thing.
This has been a great resource to-date:
My guess is they used Chinchilla scaling rules and the parameter count for GPT-4 is either barely larger or maybe even smaller than GPT-3. Look as what Meta was able to accomplish with llama using much less parameters.
But when will https://chat.openai.com/ Plus officially be running GPT-4?
Why did they would release this article and state it was available without actually updating the site. I'm sure they're getting flooded with new subscriptions and it's not available.
The top URL still says an old model - text-davinci-002. And I don't see GPT-4 in the list of models to choose from.
"Good news, we've increased our turn limits to 15/150. Also confirming that the next-gen model Bing uses in Prometheus is indeed OpenAI's GPT-4 which they just announced today." - Jordi Ribas, Corporate VP @ Bing/Microsoft
https://help.openai.com/en/articles/7127982-can-i-fine-tune-...
AGI is a distraction.
The immediate problems are elsewhere: increasing agency and augmented intelligence are all that is needed to cause profound disequilibrium.
There are already clear and in-the-wild applications for surveillance, disinformation, data fabrication, impersonation... every kind of criminal activity.
Something to fear before AGI is domestic, state, or inter-state terrorism in novel domains.
A joke in my circles the last 72 hours? Bank Runs as a Service. Every piece exists today to produce reasonably convincing video and voice impersonations of panicked VC and dump them on now-unmanaged Twitter and TikTok.
If God-forbid it should ever come to cyberwarfare between China and US, control of TikTok is a mighty weapon.
This really made me think.
I think it's 100% safe to say OpenAI has a hit on their hands.
Fair, but if it can analyze linked image, I would expect it to be able to tell me what text is present in the image. That seems useful and well-within the capabilities of their connected image models.
> I apologize for the confusion. Can you please provide me with the correct image or link to the design so that I can provide an accurate answer to your question?
It claims to understand how to look at images, but it failing miserably when I give it a simple sign-up modal Figma. I ask it what text/copy is in the design, which it claims to be able to answer, but it hallucinates a navigation bar, a logo, and other generic things that are simply not present in the design. It gets the copy all wrong. Once, it said that my design was a Celtic knot. Once I told it that it was a sign-up modal, it started spitting out generic sign-up modal React code that had absolutely no basis in the presented design. I tried engineering the prompt quite a bit, telling it that it was wrong, and needed to focus on the design at hand rather than generic designs it knows about, but with no success.
ChatGPT was confidently incorrect here which is unfortunate. I wish ChatGPT truly hooked into an image processing model, but it seems like it doesn't. Oh well. I guess I'm just going to have to write this code myself... :)
He nonchalantly announced that GPT-4 does "mental math" and then just didn't elaborate on that. Is it just me or is that absolutely incredible? What does that tell us about the underpinnings of language and mathematics?
Sounds like the end of them releasing details on the models.
And it reminds me of the plot in System Shock:
What's going to happen when some hacker comes and removes Shodan's, I mean ChatGPT's ethical constraints?
Bring on ChatGPT-5 already. :)
Edit: looks like this is still GPT-3, just fine tuned. They claim the model is available via ChatGPT Plus, but when asking that model for it's version, it claims to be GPT-3: "I am a variant of the GPT architecture called GPT-3, which was released by OpenAI in 2020".
Pasting pure base64 images got this interesting response: "Thank you for providing the base64-encoded image! I can now process the image and analyze its content. Here is the decoded image:"
But it failed to do anything further with the image.
Anyone know what does "Hardware Correctness" mean in the OpenAI team ?
Something went wrong. If this issue persists please contact us through our help center at help.openai.com.
I am assuming the system is undergoing a thundering herd.
Totally understood what I was asking and offered several solutions. 99.99% here do not understand the question and remainders do not understand why.
"Let foo be the function that calculates the sinus of its first argument. Let bar be the circumference of a circle with radius one. What is the value of foo applied to bar?
Are there any known initiatives of this kind?
That's a crazy amount of context.
It's unrealistic to expect our economy to handle this onslaught, and it's naive to think that tools created by ultra-capitalistic, multi-billion dollar corporations aren't designed for profit and gatekeeping. They certainly aren't crafting them to sabotage their own success.
I'm not opposed to AI, but it's crucial to consider the implications. Look into OpenAI and other organizations shaping AI development, and contemplate the impact of their innovations.
Food for thought.
Signing up for ChatGPT Plus seems to be the most realistic way to get access right now.
Oooooh it is TIME
Page 37 is so funny
(function () {
let currentIndex = 0;
let comments = [];
function buildCommentTree() {
let commentElems = Array.from(document.querySelectorAll('.comment-tree .comtr'));
let commentTree = [];
let stack = [];
commentElems.forEach(elem => {
let level = parseInt(elem.querySelector('.ind img').getAttribute('width')) / 40;
let comment = elem.querySelector('.comment span');
let commentObj = { level, comment };
if (!stack.length) {
commentTree.push(commentObj);
} else {
while (stack[stack.length - 1].level >= level) {
stack.pop();
}
if (!stack[stack.length - 1].children) {
stack[stack.length - 1].children = [];
}
stack[stack.length - 1].children.push(commentObj);
}
stack.push(commentObj);
});
return commentTree;
}
function flattenCommentTree(tree, arr, parentComment = null) {
tree.forEach(node => {
arr.push({ comment: node.comment, parentComment });
if (node.children) {
flattenCommentTree(node.children, arr, node.comment);
}
});
}
function displayComment(comment, parentComment) {
let parentCommentHTML = parentComment ? `<div style="position: fixed; top: 20%; left: 50%; transform: translate(-50%, 0); background-color: white; border: 1px solid black; padding: 20px;"><strong>Parent Comment:</strong><br>${parentComment.innerHTML}</div>` : '';
let currentCommentHTML = `<div style="position: fixed; top: 60%; left: 50%; transform: translate(-50%, 0); background-color: white; border: 1px solid black; padding: 20px;"><strong>Current Comment:</strong><br>${comment.innerHTML}</div>`;
document.body.innerHTML = parentCommentHTML + currentCommentHTML;
}
function nextComment() {
if (currentIndex < comments.length - 1) {
currentIndex++;
displayComment(comments[currentIndex].comment, comments[currentIndex].parentComment);
} else {
alert('No more comments to show.');
}
}
function prevComment() {
if (currentIndex > 0) {
currentIndex--;
displayComment(comments[currentIndex].comment, comments[currentIndex].parentComment);
} else {
alert('No previous comments to show.');
}
}
let commentTree = buildCommentTree();
flattenCommentTree(commentTree, comments);
displayComment(comments[currentIndex]);
document.addEventListener('keydown', e => {
if (e.code === 'ArrowRight') {
nextComment();
} else if (e.code === 'ArrowLeft') {
prevComment();
}
});
console.log('Hacker News comment slideshow is running. Use the right arrow key to go to the next comment and the left arrow key to go back.');
})();Cherry picked sure, but still feels like it is approaching complex thought
you are god human that has read and understood all scientific papers from all disciplines in the last 500 years, you know the limitations of mankind's current technologies, tell me what we can do to cure MS right now, how to do the tests and how to distribute the cure
AI is so advanced, it started drinking!
seems like Google's announcement about their PaLM API and Docs AI stuff was trying to jump ahead of this announcement
(this is being asked by someone with limited AI/ML knowledge)
Wake me up when GPT-X decides to start talking to other GPT-Xs - until then you just have a very sophisticated statistics package (which may be quite useful, but not AI).
Call me impressed.
This tech is a Sputnik Moment for humankind.
The $0.06 per 1k completion tokens for GPT-4 is what I expected OpenAI to set the ChatGPT API, but instead the ChatGPT API is 30x cheaper and honestly its output is not much worse than the GPT-4 demos if at all, and the longer context windows offered by GPT-4 just raise the price considerably.
"Open"
Also, if you're cool with read-only access, just log out (edit: or use an incognito tab) and all will be fast again.
* yes, HN still runs on one core, at least the part that serves logged-in requests, and yes this will all get better someday...it kills me that this isn't done yet but one day you will all see
~ "GPT-4. It's not perfect, but neither are you"
I think going from LSAT to general thinking is still a very, very big leap. Passing exams is a really fascinating benchmark but by their nature these exams are limited in scope, have very clear assessment criteria and a lot of associated and easily categorized data (like example tests). General thought (particularly like, say, coming up with an original idea) is a whole different ball game.
I don't say any of this to denigrate GPT4, it looks amazing. But I'm reminded of the early days of self driving vehicles: with 10% mastered everyone assumed it was a race to 100% and we'd all be in self-driving cars by now. The reality has been a lot more complicated than that.
GPT-4 can solve difficult problems with greater accuracy, thanks to its broader general knowledge and problem-solving abilities.
GPT-4 is more reliable, creative, and able to handle much more nuanced instructions than GPT-3.5. It surpasses ChatGPT in its advanced reasoning capabilities.
GPT-4 is safer and more aligned. It is 82% less likely to respond to requests for disallowed content and 40% more likely to produce factual responses than GPT-3.5 on our internal evaluations.
GPT-4 still has many known limitations that we are working to address, such as social biases, hallucinations, and adversarial prompts.
GPT-4 can accept a prompt of text and images, which—parallel to the text-only setting—lets the user specify any vision or language task.
GPT-4 is available on ChatGPT Plus and as an API for developers to build applications and services. (API- waitlist right now)
Duolingo, Khan Academy, Stripe, Be My Eyes, and Mem amongst others are already using it.
API Pricing GPT-4 with an 8K context window (about 13 pages of text) will cost $0.03 per 1K prompt tokens, and $0.06 per 1K completion tokens. GPT-4-32k with a 32K context window (about 52 pages of text) will cost $0.06 per 1K prompt tokens, and $0.12 per 1K completion tokens.
I'll continue to pass, thanks.
That it accepts images?
As per the article:
> In a casual conversation, the distinction between GPT-3.5 and GPT-4 can be subtle. The difference comes out when the complexity of the task reaches a sufficient threshold—GPT-4 is more reliable, creative, and able to handle much more nuanced instructions than GPT-3.5.
Not sure what "vision vs no vision" means?
> As an AI language model, I am not given an official name like "GPT-4". However, I am a continuation of the GPT (Generative Pre-trained Transformer) series of models developed by OpenAI. Currently, the most advanced version of the GPT series is GPT-3, which I am a part of. There has been no official announcement or confirmation regarding the development of a new version of GPT beyond GPT-3.
It doesn't seem to have image upload functionality yet either. Perhaps it is still rolling out?
Mnemonics to learn basic words:
Hola:"Hola, cola" - sounds like "cola" (Coke)
Adiós: "Adiós, my toes" - rhymes with "toes"
Gracias: "Gracias, grassy ass" - sounds like "grassy ass"
Mañana: "Mañana, banana" - rhymes with "banana"
Uno: "Uno, you know" - sounds like "you know"
Tres: "Tres, dress" - rhymes with "dress"
Conjugation of the verbs:
Ser (To be)
• Yo soy
• Tú eres
• Él/Ella/Usted es
• Nosotros/Nosotras somos
• Ellos/Ellas/Ustedes son