Tiberium

29 ms·

541.
by Tiberium 2y ago
What HDDs do you have for it to be 60TB in total?
542.
by Tiberium 2y ago
First you were talking about those games just existing, now you want me to prove to you that "they don't suck". That doesn't seem fair. Anyway, here you go: https://store.steampowered.com/app/1519310
543.
by Tiberium 2y ago
There are multiple games that use LLMs, available in Steam.
544.
by Tiberium 2y ago
But you're saying that you want a single model to handle all NPCs and the whole world. Of course this isn't possible currently. But using a separate model with separate context for each character is. Also, if you use ChatGPT for p
545.
by Tiberium 2y ago
Are you sure? Models like Claude 3.5 Sonnet are both good at writing and instructions, as long as you set some guardrails for the model, they can be great NPCs.
546.
by Tiberium 2y ago
https://openrouter.ai is exactly that, and they even support payments with crypto.
547.
by Tiberium 2y ago
Telegram's Saved Messages (essentially self-texting) is also commonly used for saving info, somewhat recently they even added tagging and the ability to view messages that you forwarded into there per chat (so you can see all messages
548.
by Tiberium 2y ago
I think the submission is a bit wrong in editing the title from the original. I understood it like this: Chrome has a built-in extension that uses public Chrome APIs that are easily available to other Chrome extensions. The issue described
549.
by Tiberium 2y ago
In what world does "system / tab CPU usage, GPU usage, and memory usage" mean "full access to the system"? Any Chrome extension can access this info easily, the point that the tweet makes is that there's a buil
550.
by Tiberium 2y ago
Ooh, I remember this project just from the video released by the creator :) It gives me an almost-ASMR feel - https://www.youtube.com/watch?v=Y-o3pJEhwD4 ("pyNES - First preview")
551.
by Tiberium 2y ago
Did you release the dataset and the code for testing? It would be interesting to check how 3.5 Sonnet performs on this task.
552.
by Tiberium 2y ago
GPT models shouldn't be used at temp 1 unless you only care about creative writing. They get much worse at factual stuff and code than with lower temperatures. And yes, 3.5 Turbo is less affected by this, which might be the reason why
553.
by Tiberium 2y ago
An important thing to consider is cost - Google Translate is used by hundreds of millions of people and probably has an insane amount of requests done daily, yet it's free for personal use. LLMs are much less economical.
554.
by Tiberium 2y ago
Are you sure it actually called to an external tool to compute the square root? Most big LLMs today can do that by themselves.
555.
by Tiberium 2y ago
Is HN being attacked? This was reposted instantly after the previous submission got flagged.
556.
by Tiberium 2y ago
Can this stop being reposted right after it's flagged? Is this even allowed on HN? I've never seen such behavior on HN before, so I'm curious.
557.
by Tiberium 2y ago
> I myself have formal training as a data scientist, going so far as to dominate a competitive machine learning event at one of Australia's top universities and writing a Master's thesis where I wrote all my own libraries from
558.
by Tiberium 2y ago
Interesting article, sadly due to my exposure to LLMs I couldn't help but notice that the parts about "oinasf" and sakura.sh are AI-edited at least. Kind of a weird choice considering that a lot of the article was clearly hum
559.
by Tiberium 2y ago
A couple lists: https://kycnot.me/?t=service (not just hostings) https://bitcoin-vps.com/ (an extensive list for ones that accept BTC, most accept other coins too) There are hundreds of VPS hosters that acc
560.
by Tiberium 2y ago
Not exactly related to the article itself, but it seems to me to be heavily AI-edited, with a lot of unnecessary fluff, the kind that ChatGPT adds. Honestly it seems like the article could be shortened in half at least. And probably all tho
561.
by Tiberium 2y ago
MoltenVK is an implementation of Vulkan based on Metal so that it can be used on macOS to run Vulkan applications. Asahi Linux is an in-development Linux distribution to make Linux compatible with Apple's M series processors. This spec
562.
by Tiberium 2y ago
Are you sure it was only stored in your space secrets? Not variables (which are public) or stored in the .env file (also public).
563.
by Tiberium 2y ago
LMSys' Cloudflare protection is sadly trivial to bypass - just search for "LMSYS Chatbot Arena: Benchmarking LLMs in the Wild" (with quotes) in Censys and you'll find the origin host. It's a common misconfiguration
564.
by Tiberium 2y ago
Not open source (freeware), but much faster than WinDirStat for NTFS - WizTree https://diskanalyzer.com/ . In short - it scans the actual filesystem metadata directly instead of enumerating files through the OS APIs, which m
565.
by Tiberium 2y ago
https://news.ycombinator.com/item?id=40427620 brings an excellent counterpoint to your post - why won't Taiwan itself drop the unrealistic claims on all of China?
566.
by Tiberium 2y ago
You can easily do that, and it's been already done. Check out: - https://websim.ai - https://worldsim.nousresearch.com/browser Both of those use Claude by the way, because Claude models are way more creativ
567.
by Tiberium 2y ago
And to be more exact, GPT-4chan is based on GPT-J (same architecture as GPT-3 whose weights were never released) which only had 6B params and that was back in 2021-2022.
568.
by Tiberium 2y ago
LLMs are deterministic with 0 temperature on the same hardware with the same seed though, as long as the implementation is deterministic. You can easily use the OpenAI API with the temp=0 and a predefined seed and you'll get very deter
569.
by Tiberium 2y ago
I think ultimately text is too structured by itself to be able to understand whether it's AI-generated or not. You can easily tell LLMs to write in a different style or even give examples, and they'll happily oblige. Those tools a
570.
by Tiberium 2y ago
I think by "official" they meant support from the manufacturer.

More ›