Hacker Newsnew | past | comments | ask | show | jobs | submit | IOT_Apprentice's commentslogin

What languages are supported?

If they don't mention it, it's probably just Americanese, right?

The US military has been murdering fishermen off the coast of South America this year, after firing on the boats, disabling them they were shown no quarter and kill shot orders were given AND EXECUTED. Those are WAR CRIMES. The US military has done war crimes alongside the IDF over during this current year in Iran. Furthermore, the President of The United States has publicly stated he will destroy the Iranian civilization. He has ordered new very large bombs be built to Drop on Iran. Do you care to acknowledge these facts?

No you aren’t. Your repeated comments show otherwise. You have trivialized their deaths, ignored other findings and demonstrated a distinct lack of empathy. Do better.

Are the schoolgirls dead, alongside some of their parents Anna teachers? Did Hegseath eliminate the review process pre stikes to determine if a site is civilian in nature, the CHMR? Yes they did. Ghouls like you have posted so much in this thread ignoring the actions to the point of cheerleading the mass murder of children. Where are the administrators like Dan G to bring you to heel?

Why instead of answering specific question you switch on explicit personal insults?

Mastodon support? Discord scans everything posted on it.


I tried it. I asked where Bruce Lee was born. It stated he was born in Hong Kong. I challenged it and it went further naming a hospital there. I stated he was born in San Francisco and it apologized and then said his father was a missionary traveling in America, which was also wrong. Bruce’s father was a famous Cantonese Opera singer and actor.

This model had zero information right, while being fast in responding.

Unacceptable.


It gave the correct answers to both questions for me:

> Bruce Lee was born in San Francisco, California, USA on November 27, 1940.

> Bruce Lee's father was a Chinese opera singer

That being said, this is not a good test. It is a language model (a very small one), not an encyclopedia.

ChatJimmy interface is just a tech demo. Without tool calling functionality we can't expect it to be factually correct.


if it's baked into silicon how can you two get different answers?


It still works the same way other LLMs do, by outputting the probability distribution over the possible completions (The weather is ... (sunny (50%), cloudy (50%))). Then the next token is sampled from this probability distribution (in our example the next word could be "sunny" or "cloudy" equally likely), which can result in different outputs every run.


Could the model or algorithm be changed to make it deterministic somehow? It could help a lot if there were reproduceable outputs from deterministic baked-in silicon.


You can make any LLM deterministic by dropping the temperature hyperparameter to zero.

This will generally make them suck, though, a little bit of randomness is necessary for proper function.


You can also use a fixed seed for your prng. A hash of the input text (up to the current turn) should do.


But since it's so fast you can just ask it 100 times where Bruce Lee was born, and statistically you'll get the correct answer. We could call it "mixture of idiots". /s


That's not what speed is useful for.

I just pasted your comment and its whole inheritance chain to it, started my comment, and asked to generate a total of 9 completions, 3 from each of {current & next word, current paragraph, current paragraph + rewrite the entire paragraph}.

Half of the answers were perfectly good (ironically, not the "next word" ones!), but the important bit, they came back near-instantly ("Generated in 0.024s - 14,163 tok/s", the page says). Slightly more powerful model while keeping this under a second, and this could easily become a qualitatively different form of autocomplete/text suggestion. Running in the background every couple keystrokes, or every time user stops typing for more than 500ms.


>That's not what speed is useful for.

>I just pasted your comment and its whole inheritance chain to it,

Good idea. Only problem is it doesn't work. I just did the same thing with exactly this prompt:

>did the user IOT_Apprentice participate in the thread below and if, number and quote all of their comments. Only just number and quote the comments or write "Did not participate", do not add any commentary. Quote any comments by this user verbatim, exactly as input. Thread:

followed by pasting the thread[1]

And received the answer "IOT_Apprentice did not participate in the thread."[2] in 0.001s, even though they have literally the last comment in my quote and it's clearly legible.

It's particularly insidious because the understanding and thinking that is required to follow my requested answer format exactly is substantial - so based on the fact that it gets the format right and clearly understood the assignment, I would be inclined to believe that it would also be correct!

So to use your example, it's not just autocomplete, it's autocomplete that confidently returns "No matching results" in 0.001 seconds, even though there is a search term matching what you put in, right in the prompt itself that was sent to it. That is much worse than useless.

[1] prompt: https://ibb.co/CKVmRvtd

[2] result: https://ibb.co/BKdRKmyD


Using LLMs for information retrieval is the most stupid thing one can do. Especially when old methods work much better.


She is a scholar of the subject and wrote a well regarded translation of the Odyssey. She is a subject matter expert. And Nolan mentioned her in reference to his movie. I read the whole review and now I’ll read her book.


I am from the USA and 401k + IRAs are the terms here and yes they are going to be used. Your point?


There’s a relevant structural difference between 401(k)s/IRAs and pensions. You get to choose how to invest your personal retirement accounts, while pensions are institutional investors with active managers. An individual retiree can adjust their retirement account’s risk profile over time, but a pension fund needs to be generating sufficient revenue to cover any upcoming disbursements. There are also often other constraints on the pension fund, such as being substantially invested in the sponsoring company or only investing in investment-grade securities. If a single significant investment blows up, this can threaten the solvency of the fund for all beneficiaries. There have been pension funds that have zeroed out their unvested beneficiaries due to insolvency.

My point in bringing this up is that the fact that most American retirements are self-directed rather than fixed-benefit means that the catastrophic scenario implied by the OP is less likely.


Protect them from what? The known associates of trafficking associated with Maxwel & Epstein have not been fully exposed or prosecuted, instead their names are hidden. Nothing is happening. This is an attempt to force identity in all things. It should be stopped.


I’d just like to buy steamOS and install it on my ryzen 9 desktop and my Ryzen laptop.


No need to wait and buy anything. Try any gaming focused Linux distributions and you can already have this. Check out CachyOS, Nobara, Bazzite ...


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: