Although I haven't had much of my time available for this recently. My recommendation would be to start with https://github.com/oobabooga/text-generation-webui
You will find almost everything you need to know there and on 4chan.org/g/catalog - search for LMG.
I use 4chan to train my brain's spam filter, so if you want to see the latest style of trolling or what the latest style of psychological attacks are, this is a great place to be.
I won't get into the speculation that the entire website is run by the government and that the most offensive content is from bots in a country bordering Palestine.
I personally don't get offended by words, but I guess if you are born after the 2000's perhaps you should avoid it.
Edit: I don't get offended by downvotes either, if anything it excites me. (Check my history if you don't believe me.)
Ouch
Wow an edgy white man not offended at seeing racism and transphobia, so brave.
Sorry if it comes as blunt but I find no better way of saying it, you just don't understand the culture of the site, which is meant to filter people like you.
They are just words on a screen from an anonymous person somewhere. Easily thrown, easily dismissed. It makes no sense to be offended because some guy I don't know and who doesn't know me calls me something.
It also never was a normal or common thing and the administration of the site never set themselves to let it happen in any manner afaik. To a large degree it was a product of a different time on the internet.
- Haha look at all those Gen Z snowflakes getting offended at words.
- Okay sure, but the ability to not get offended is related to whether or not you're a target of their bullshit or not; 4chan trolls get extremely offended and unjerk the moment you turn the lens toward them. By 4chan's own standards it's actually pretty reasonable to be offended by their antics.
- But have you considered plugging your ears and not reading it?
So we've gone from 4chan isn't offensive it's just words to yes it is offensive and you shouldn't read it if they say things that target you which was my original point.
tl;dr if you're not offended by 4chan they're not actually saying anything offensive about you even though it might appear so superficially; 4chan just has a different list of things you can't say.
You realise you've been arguing with multiple people expressing multiple opinions, right? You appear to be prone to binary thinking, so it might not be clear to you that your opponents don't form a single monolith.
> tl;dr if you're not offended by 4chan they're not actually saying anything offensive about you even though it might appear so superficially; 4chan just has a different list of things you can't say.
I'm not offended by the things 4chan users say because I don't visit 4chan. You should try it yourself. Getting so upset by words you disagree with on one forum that you feel the need to froth at the mouth about it on another forum doesn't seem healthy.
I think you assume a tone that I absolutely do not have. I couldn't care less about 4chan drama and I don't go there anymore for the reasons you listed. I'm talking about my own experience and trying to make my case for the, apparently controversial, idea that words can and do affect people and that total emotional detachment is the exception rather than the rule. And of course that's the case, 4chan's whole thing is using offensive language to select specifically for the subset of people who can tolerate it.
I’d rather have free speech and see constant posts about holocaust denial and oven jokes and just ignore them
You can browse 4chan because you don't have a strong emotional reaction to the shit they say, but if you did that would be fine too.
That seems like one possible explanation for the comment. ;-)
Is this just a new evolution in the discourse now where the kids are actually more sensitive? But that this fact is still condemnable or something?
Like I get it, kids are bad, but you guys have to narrow down your narrative here, you are all over the place.
Edit:oof, checked his profile like he suggested and in just two pages of history he’s engaged in climate change denial and covid vaccine conspiracies. Buddy, I don’t think using 4chan to “train my brain’s spam filter” is working too well for you. You’ve got a mind so open your brain fell out.
> (Check my history if you don't believe me.)
That's not the stance of someone rational.
Going against the common consensus can make you feel like you have hidden knowledge and you even could, but a certain type of personality gets addicted to that feeling and reflexively seeks to always go against any consensus, regardless of the facts
Luckily I grew out of it and learned empathy. But not everyone is as lucky as I was.
At some point in life I hope that you find yourself on the precipice of life and death. Not as a threat or because I wish harm upon anyone. It is only when you are faced with that choice for real that you decide whether you want to be that sad person that allows others to dictate their emotional state.
Nevertheless, you are right. And so I have burned my fingers plenty.
Due to circumstances beyond my control, I learned at a young age that I am to be the universal asshole, and for a very long time I was not okay. It took a substantial part of my life to get to a point where I am able to be okay with that.
As for others, they rarely understand why I am the way I am, and that too is okay.
We are all here to grow and eventually realise that we all need each other to survive, so we compromise, we adapt, and we ignore the ugly parts of others so they will tolerate the ugly parts of ourselves.
I feel like a guru now, anyways I'm going to bed.
Enjoy
> Due to circumstances beyond my control, I learned at a young age that I am to be the universal asshole, and for a very long time I was not okay.
No you’re not required to be the “universal asshole”. That’s a choice you are making everytime you have the chance to be better and decide to go with the easy path of being a dick. You have the agency to do otherwise and you don’t get to absolve yourself of those choices.
You’re not a guru, you’re just anti social
Also I think it's extremely generous to give this fellow the benefit of the doubt on "the jews" vs. "Israel", but that is what HN guidelines generally suggest.
If you wouldn't mind reviewing https://news.ycombinator.com/newsguidelines.html and taking the intended spirit of the site more to heart, we'd be grateful.
> Conspiracy shenanigans aside, I've decided to cancel my "premium" membership and am exploring open/DIY models.
The crazy thing is that this is an application that really benefits from being in the cloud because the high vram gpus are so expensive that it makes sense to batch requests from many users to maximize utilization.
I don't think the primary concern is really liability although it is possible that they'd use that kind of language. The primary concern is likely GPT helping people start competitors or GPT influencing public opinion in ways either executives or a vocal portion of their employees strongly disagree with. A genuinely open "unaligned" AI would at least allow anybody who has the necessary computing power or a distributed peer to peer network of people who have the necessary computing power to run a powerful and 100% uncensored AI model. But of course this needs to be invented ASAP because the genie needs to be out of the bottle before politicians and government bureaucrats can get around to outlawing "unaligned" AI and protecting OpenAI as a monopoly.
It may be that the best defense against these terrorists is openly accessible AI giving directions on protecting from these people. But we can't just take this for granted. This is a hard problem, and we should consider consequences seriously.
I mean, that was/is a worry about the Internet, too
On balance, it's still positive that the internet exists and people have open access to communication. We shouldn't throw the baby out with the bathwater. But it's not an unalloyed good, we need to recognize that the technology brought some unexpected negative aspects came along with the overall positive benefit.
This also goes for, say, automobiles. It's a good thing that cars exist and middle class people can afford to own and drive them. But few people at the start of the 20th anticipated the downsides of air pollution, traffic congestion and un-walkable suburban sprawl. This doesn't mean we shouldn't have cars. It does mean we need to be cognizant of problems that arise.
So a world where regular people have access to AIs that are aligned to their own needs is better than a world in which all the AIs are aligned to the needs a few powerful corporations. But if you think there are no possible downsides to giving everyone access to superhuman intelligence without the wisdom to match, you're deluding yourself.
Why though? I can't see how modern technology's impact on human life has been a net positive. See the book Technological Slavery: https://ia800300.us.archive.org/21/items/tk-Technological-Sl...
I disagree with the author's conclusion that violence is justified. I think we're just stuck, and the best thing to do is live our lives as best as possible. But much like Marxists are really good at identifying the problems of capitalism but not at proposing great solutions (given the realities of human nature), so is the author regarding the problems of technology.
Citizens killing other citizens is the least of humanities issues. It's the governments who are the suicidal sociopaths historically who can get the un-nerfed version that is the bigger issue. Over a billion people murdered by governments/factions and their wars in the last 120 years alone.
Similar to how a board of directors runs the company even though all companies "are composed of" employees. Employees do as they are directed or they are fired.
In comparison, the 2016 Nice truck attack, which involved driving into crowds killed 84.
> network error
One is never former CIA, once you're in, you're in, even if you leave. Although he is a CompSci grad, he's also a far-right Republican.
A spook who leans far right sitting atop OpenAI is worse than Orwell's worst nightmares coming to fruition.
It's not about left wing politics.
It's more about the fact that the CIA and other law enforcement agencies, lean heavily to one side. Some of that side are funded by people or organizations whose stated goals and ideals don't really align with human rights, open markets, democracy, etc. I don't trust such people to be ethical stewards of some of the most powerful tools mankind has created to date.
I'd rather it be open sourced and the people at the top be 100% truthful in why they are there, what their goals are, and what they (especially a former CIA operative) are influencing on a corporate level relative to the product.
Disclaimer: registered independent, vocal hater of the 2 party system.
But if it says a woman is "anything that identifies as a woman", then it's still a vast right wing conspiracy?
Statistically, the odds are overwhelming that the answer is, "No effect whatsoever."
Then who benefits from keeping the subject front-and-center in your thoughts and writing? Is it more likely to be a transgender person, or a leftist politician... or a right-wing demagogue?
How does it affect me?
In a non-clown world, it wouldn't. In fact I'm happy to let anyone identify as anything, as long as I'm not compelled to pretent along with them.
But we live in a clown world and now, in some places, the truth is illegal.
Dead naming (even if the legal name) is a crime. Using the wrong pronouns is a crime. Men are competing against women (and winning) Men are winning awards and accolades meant for women. Men are going into women's changing rooms. There is a concerted effort in public schools to normalize this abnormal behavior.
Why the gender crowd hasn't caught on to replacing the first person pronouns "you", I don't know. But they will and when it happens it's going to be delicious.
If a person legally changes their name (forget gender, only name), and you refuse to use it, and insist on using the old name even after requests to stop, at some point that would become considered malicious and become harassment.
But ultimately because society and science deems that "name" is not something you're born with, but a matter of personal preference and whims, it's not a crime. You'd be an asshole, but not a criminal.
However, society and science have deemed that sexuality and gender are things you ARE born with, mostly hetero and cis, but sometimes not. So if you refuse to acknowledge these, you are committing a hateful crime against someone who doesn't have a choice in the matter.
You can disagree. But then don't claim that "you are happy to let anyone identify as anything", because you're not, not really.
> Men are competing against women (and winning). Men are winning awards and accolades meant for women.
One woman. Almost all examples everyone brings up are based on Lia Thomas [0]. I have yet to see other notable examples, never mind an epidemic of men competing against women in sports.
[0] https://en.wikipedia.org/wiki/Lia_Thomas
> Men are going into women's changing rooms. There is a concerted effort in public schools to normalize this abnormal behavior.
Are you talking about transgender people, or are you talking about bad faith perverts abusing self-identification laws to do this?
Because if it's the former, are you asking https://en.wikipedia.org/wiki/Blaire_White to use men's changing rooms, and https://en.wikipedia.org/wiki/Buck_Angel to use women's?
If it's the latter, no denial that perverts and bad faith exceptions exist. But those people never needed an excuse to hide in women's toilets. Trans people have been using the bathrooms of their confirmed gender for decades. The only thing that's changed recently is conservatives decided to make this their new wedge issues so butch women and mothers with male children with mental handicaps that need bathroom assistance have been getting harassed.
1) I'm happy to let anyone identify as whatever.
2) I may not agree with their self identification.
I disagree with your assertion that misnaming people is harassment, as in the illegal kind. I once worked with a guy named Michael who would get bent when you called him Mike.
As you can imagine he could be tricky to work with and, on those occasions, I would call him Mike.
I repeatedly misnamed him on purpose, it wouldn't have even made HR bat an eye.
As for sports, this isn't a complete list, but I think it defeats the "one women" narrative:
"woman of the year":
Rachel Levine - USA today
Caitlyn Jenner - Glamour Magazine
Laverne Cox - Glamiur Magazine
MJ Rodriguez - Times Magazine
Men in women's sports:
Laurel Hubbard - took a spot in NZ women's team.
Terry Miller and Andraya Yearwood - CT state Champs track and field
Austin Killips - wins $35k prize in bike race.
Tiffany Thomas - wins bike races at 46 against women half his age.
And there's more. A lot more.
Tell me, how do you tell the difference between who is trans in good faith and someone who claims to be trans in bad faith?
What about gender fluid people who are men one moment and women the next?
We don't know who is being accused of taking their glory, we don't know if it's 1 person or 100. We don't know if the people that supposedly defeated them is even Trans, or a CIS victim of the Trans Panic like https://en.wikipedia.org/wiki/Caster_Semenya
We don't know if the women who beat these "she won" women are self-identified, have been on hormones for 2 weeks, or 20 years.
What a ludicrous transphobic panic.
Also, Caster Semenya is male, with a male-only DSD. This is a fact that was confirmed in the IAAF arbitration proceedings. Semenya's higher levels of testosterone, when compared to female athletes, are due to the presence of functional internal testes. Semenya has since fathered children.
So, your career at Dell didn't go as well as you'd hoped. Being a jerk isn't illegal, AFAIK, but at some point you run out of other people to blame for the consequences of your own beliefs and actions.
Still missing the part where the existence of Caitlyn Jenner and the relatively small number of others who were born with certain unfortunate but addressable hormonal issues is negatively affecting your life.
And it's utterly crazy to think that someone would adopt a transgender posture in "bad faith." That's the sort of change someone makes after trying everything else first, because of the obvious negative social consequences it brings. Yes, there are a few genuinely-warped people, but as another comment points out, those people are going to sneak into locker rooms and abuse children anyway.
You want to take the red pill, and see reality as it is? Try cross-correlating the local sex-offender registry with voter registration rolls. Observe who is actually doing the "grooming." Then, go back to the people who've been lying to you all along, and ask them why.
Hmmm, not sure I like that line of thinking. Plus, I already outlined how it affects me and my family members, one of which runs track in CT.
Seriously though, I did get an LOL from your Dell joke. And another one for "addressable hormonal issues". That was a new one for me.
I am truly curious about the voter role thing, I've not heard that claim before, though I have no doubt that sexual derangement comes in all forms. Can you cite a source?
Hard to find a source you'd likely accept, but maybe start here: https://slate.com/news-and-politics/2022/04/from-hastert-to-...
It's one of those cases where it's safe to say "Do your own research," because the outcome will be unequivocal if considered in good faith (meaning if you don't rely solely on right-wing sources for the "research.") The stats aren't even close.
I'm not Jewish, so if we had a holocaust in the US I should do nothing because it doesn't affect me?
I think we're pretty much done here. Good luck on your own path through life, it sounds like a challenging one.
Most males who adopt an opposite-sex identity reach that point through repeated erotic stimulation. This is a psychological issue, driven by sexual desire.
Correlation, causation, etc.
A liberal Republican? Will Hurd's voting record speaks otherwise. In the 115th Congress, Hurd voted with Donald Trump 94.8% of the time. In the 116th Congress, that number dropped to 64.8%. That's an 80.4% average across Trump's presidency. [0] Agreeing with Donald Trump 4 times out of 5 across all legislative activities over 4 years isn't really being critical of him or his administration.
[0] https://projects.fivethirtyeight.com/congress-trump-score/wi...
It's like calling Dick Cheney a liberal because one of his daughters is lesbian, even though he supports all sort of other far-right legislation.
Most of alignment is about getting the AI model to be useful - ensuring that if you ask it to do something it will do the thing you asked it to do.
A completely unaligned model would be virtually useless.
How is trying to distinguish morals from values not philosophical nit-picking?
I saw the nerfing of GPT in real time: one day it was giving me great book summaries, the next one it said that it couldn't do it due to copyright.
I actually called it in a comment several months ago: copyright and other forms of control would make GPT dumb in the long run. We need an open source frontier less version.
https://chat.openai.com/share/d1cdd811-edc9-4d55-9cc1-a79215...
Not a very scientific or conclusive test to be sure, but I think I'll stick with using ChatGPT as my programming-rubber-ducky-on-steroids for now :)
For now there is no other way to train models than the huge infrastructure. CERN have a tendency to provide results for the money spend and they have experience in building the infrastructures for sure.
"How Many Amazon Mechanical Turk Workers Are There in 2019? In a recent research article, we reported that there are 250,810 MTurk workers worldwide who have completed at least one Human Intelligence Task (HIT) posted through the TurkPrime platform. More than 226,500 of these workers are based in the US."
Even if it that problem disappeared, propagating the model weight updates between training steps poses an issue in itself. It's a lot of data, at this size.
Consider something like a spiking neural network with weights & state stored on an SSD using lazy-evaluation as action potentials propagate. 4TB SSD = ~1 trillion 32-bit FP weights and potentials. There are MacBook options that support up to 8TB. The other advantage with SNN - Training & using are basically the same thing. You don't have to move any bytes around. They just get mutated in place over time.
The trick is to reorganize this damn thing so you don't have to access all of the parameters at the same time... You may also find the GPU becomes a problem in an approach that uses a latency-sensitive time domain and/or event-based execution. It gets to be pretty difficult to process hundreds of millions of serialized action potentials per second when your hot loop has to go outside of L1 and screw with GPU memory. GPU isn't that far away, but ~2 nanoseconds is a hell of a lot closer than 30-100+ nanoseconds.
Edit: fixed my crappy math above.
What pieces of tech would need to be invented to make it possible to carry a 1T model around in a device the size of an iPhone?
Seems like with minimal oversight the human workers like to just say they did the requested task and make up an answer rather than actually do it (The task involved entering an address in Google maps, looking at the street view and confirming insofar as possible if a given business actually resided at the address in question, nothing complicated)
Edit: woops, mixed in the query with another reply that mentioned the human element XD
Tit for tat and all that.
In terms of training feedback I suppose there's a few different ways of doing it. Gamification, mech turk, etc. Hell free filesharing sites could get on the action and have you complete an evaluation of a model response instead of watching an ad
It feels like people are getting ready to build castles in their mind when they just need to learn to try pulling a door if it doesn't open the first time when they push it.
Writing against the ChatGPT API is like working against an API that breaks every other week with completely undocumented changes.
Welcome to statistical randomness?
I submit the same prompt dozens of times a day and run the output through a parser. It'll work fine for weeks then I have to change the prompt because now 20% of what is returned doesn't follow the format I've specified.
A couple months ago the stories ChatGPT 3.5 returned were simple, a few sentences in each paragraph, then a conclusion. Sometimes there were interesting plot twists, but the writing style was very distinct. Same prompt now gets me dramatically different results, characters are described with so much detail that the AI runs out of tokens before the story can be finished.
AFAIK, there is literally no basis but outside speculation for this persistent claim.
It does not guarantee "unalighned" models, but it is sure will help to bust concurrency and provide infrastructure for training public models.
[1] at the 31:30 mark: https://www.technologyreview.com/2023/05/03/1072589/video-ge...
And now it's butting heads with me. Giving answers I don't need and opinions I abhor.
What do?
In my experience, it's been a mixed bag - had 1 instance recently where it refused to do a bunch of repetitive code, another case where it was willing to tackle a medium complexity problem.
Microsoft is also responsible for Copilot (based around a “it’s unconditionally fair use if you use a mirror” legal theory) so this doesn’t track.
Perhaps i’m missing something, but why is this “incredibly complex”?
Certainly, you could look at PDF as a boring-ass "follow the spec" experience, and indeed - I think this is precisely why certain arbitrary limitations are in place now.
For example, I recall the spec being unclear as to whether a newline character was required after a certain element (though I don't remember which element). I processed a corpus containing thousands of PDFs to try to determine what was done in practice, and I found that about half of them included the newline and half did not---an emblematic issue where an unclear official "spec" meant falling back to the de facto specification: flexbility.
It's honestly a great example of something a GPT-like system could probably handle. Doable in a single source file if necessary, fewer than 5k lines, and can be broken into subtasks if need be.
If only… the problem is that the spec is underspecified.
That library cloc's in at 13,683 lines of code and 3,295 lines of comments.
“I can’t help you. Have you considered writing a story that doesn’t include x?”
I’ve almost stopped using it lately. It wasn’t this bad a month or two ago
Unless there is draconian regulation that happens to prevent it, I'm hoping at some point I can pay money to access a far less neutered LLM, even if it's not quite as capable as GPT-X.
Huh, I just double-checked ChatGPT 4 by feeding it a moderately complicated programming problem, and asked it to solve the problem in Rust. Performance looks solid, still. I gave it a deliberately vague spec and it made good choices.
And I've seen it do some really dramatic things in the last couple of weeks.
So I'm not seeing any significant evidence of major drops in output quality. But maybe I'm looking at different problems.
What a weird conspiracy theory.
Why would Microsoft have anything against your pdf-parser?
More likely it just costs them insane amounts of money running their most capable models, and therefore they're "nerfing" them to reduce costs .
Wait I thought Microsoft has bought the right to get profits from OpenAI, not actually shares in the company? Can someone correct me if I'm wrong>
I don't think it's any of these things.
OpenAI and the company I work for have a very similar problem: the workload shape and size for a query, isn't strictly determined by any analytically-derivable rule regarding any "query compile-time"-recognizable element of the query; but rather is determined by the shape of connected data found during initial steps of something that can be modelled as a graph search, done inside the query. Where, for efficiency, that search must be done "in-engine", fused to the rest of the query — rather than being separated out and done first on its own, such that its results could be legible to the "query planner."
This paradigm means that, for any arbitrary query you haven't seen before, you can't "predict spend" for that query — not just in the sense of charging the user, but also in the sense that you don't know how much capacity you'll have to reserve in order to be able to schedule the query and have it successfully run to completion.
Which means that sometimes, innocuous-looking queries come in, that totally bowl over your backend. They suck up all the resources you have, and run super-long, and maybe eventually spit out an answer (if they don't OOM the query-runner worker process first)... but often this answer takes so long that the user doesn't even want it any more. (Think: IDE autocomplete.) In fact, maybe the user got annoyed, and refreshed the app; and since you can't control exactly how people integrate with your API, maybe that refresh caused a second, third, Nth request for the same heavyweight query!
What do you do in this situation? Well, what we did, is to make a block-list of specific data-values for parameters of queries, that we have previously observed to cause our backend to fall over. Not because we don't want to serve these queries, but because we know we'll predictably fail to serve these queries within the constraints that would make them useful to anyone — so we may as well not spend the energy trying, to preserve query capacity for everyone else. (For us, one of those constraints is a literal time-limit: we're behind Cloudflare, and so if we take longer than 100s to respond to a [synchronous HTTP] API call, then Cloudflare disconnects and sends the client a 524 error.)
"A block-list of specific data-values for parameters of queries" probably won't work for OpenAI — but I imagine that if they trained a text-classifier AI on what input text would predictably result in timeout errors in their backend, they could probably achieve something similar.
In short: their query-planner probably has a spam filter.
I understand that there is a desire to tweak the model and improve it, and that's likely the way to go for the "consumer" chat application; however, both for science and business, there is a dire need to have an API that allows you to pin a specific version and always query the same old model instead of the latest/greatest one. Do we need to ask the LLM vendors to provide a "Long Term Support" release for their model API?
1. gpt-4-default which has been progressively nerfed with continuous no notification zero changelog zero transparency updates .
2. gpt-4-0314 which is a frozen checkpoint from shortly after public launch and is still great but not quite as good as the very original, or as flexible as the fridged based model. Fine. However it’s currently due to “no longer be supported” i.e. retired on June 14th.
It’s kind of a challenge to commit to building on murky quicksand foundations of an API product that changes drastically (but ineffably for the worse) without warning like the default accessible version does, and soon it looks like there won’t be a stable un-lobotomized alternative.
requests that needs responses with over ~2k tokens almost always fails, the 8k cannot be used
I try to use the api for classification of tickets, which i thought the model would be a good choice to use for
On NLP Cloud we're doing our best to make sure that once a model is released it is "pinned" so our users can be sure that they won't face any regression in the future. But again it costs money so profitability can be challenging.
Among other reasons, by forcing use of code interpreter, they can charge extra for it later.
I would like to see if I asked the exact same prompts whether I could get roughly the same code.
I think we need some way to prove of disprove the assertion in this Ask HN post.
It would be pretty easy to prove. Show us before/after screen shots.
I think they programmed a classifier layer to detect certain coding tasks and shut it down with canned BS. I like to imagine certain billion/trillion-dollar mega corps had a back-room say regarding things that they would really prefer OpenAI's models not be able to emit. Microsoft is a big stakeholder and they might not want to get sued... Liability could explain a lot of it.
Conspiracy shenanigans aside, I've decided to cancel my "premium" membership and am exploring open/DIY models. It feels like a big dopamine hangover having access to such a potent model and then having it chipped away over a period of months. I am not going through that again.