天空вℓσи∂ [Deprecated edition] @skyblond@social.cleverlibre.org

Now JetBrains joined the AI hype, they start forcing users to try out their new stupid AI plugin. And the quality of their products has degraded too.

Just read the reviews: https://plugins.jetbrains.com/plugin/22282-ai-assistant/reviews

And the most annoying part is the dollar signs used in jextract.

To access a field in a structure, you have to use `llama_grammar_element.type$get(obj)`. And the dollar sign is a keyword in Kotlin. You can't use it in function names, unless you quote it with "`", like "llama_context_params.`seed$get`(contextParams)". And to make things worse, IDEA's auto completion is broken when involving this "`" sign.

I blame c++ for my degraded sleep quality.

While jextract makes the header translation much easier compared to JNA and JNR, but win32 api still does not work, and c++ support is not good.

The standard header from llama.cpp is working, but other functionalities like grammar parsing and sampling init is not included. They shipped as a common component in the example codes. While I want to compile the example code as a shared lib and invoke them from Java, the jextract has some problems with C++ headers and not happy with it.

For wintun, it uses win32 headers in its header files, which let jextract to process (almost) the whole win32 lib. The pointer size seems to be a problem. One said it should be 32 bits but another said 64 bits.

No wonder it's still in preview stage.

11am, but still sleepy

So hard to get up

@Humpleupagus

That reminds me that I have a big problem with some European languages, mainly the speaking. Can't properly make the correct sounds. While my niece growing up in German has no issue.

The music part, hmmm. I don't listen to classical music before. But later somehow I start listening to it. I can tell that over the years, although no body teaches me how to properly catch the details, myself has already developed better-than-before ear. Compared to 4 years ago, now I can tell some details of the music and kind of understand the expression of it. I guess that does not count as "training", but just listening a lot and repeat a lot makes enough difference.

@Humpleupagus @lucifargundam

Talking about that, I once read a passage said birds can see RGB along with UV. Their world is much more colorful than humans.

I wonder what that means to me. The RGB vision I have is colorful enough for me, I don't know if my brain can handle those extra color space.

Also, unlike scifi, human brains can imagin something like space travel or time travel, I noticed that my brain (as it can represent the average human brain, I guess?) has difficulties imagining senses. The most obvious one is the taste, when I can't reproduce what others claim, with the same dished.

Got my watch fixed. Spent a week and about 75USD.

I know casio's gshock is very tough.It already amazed me when I found the core is still functional after a great crash. I spent two weeks to regain my ability to walk, and until last week my last wound was healed.

Now I'm amazed again by its toughness. There is a huge crack in the front glass, across both XZ plane and Y axis. I assume the front glass is fragile, but guess what? I pushed on it and it still holds! This huge crack even failed to penetrate the whole glass! The inner side of the glass is still one piece.

With additional 2 backups, I guess it's a lifetime guarantee.

I guess I lost the point during typing. When watching those videos I was thinking about different things but now I forget.

Now I'm wondering if I got ADHD.

My father is red-green color blind. And frankly speaking, until he told me, I never noticed that. And he even got a driver's license in the old days (I mean 198X).

While I'm lucky enough to have full color vision, I find myself can't imagine how color blind see the world. Sometimes I wondering if I'm color blind but just don't know that. (Several tests on the internet said I'm perfectly fine)

The thing is, how I can tell I'm color blind if I don't know what color I don't know? When talking about the flavor, I found my father and I are not sensitive to the taste of sour, while my mother is. For the same dishes, my mother thinks it is unacceptable sour, while my father and I think it's not sour enough. If I can't taste the sourness at all, how do I know I have that defect?

I'd rather not think this deep hole of a thing of philosophy. I guess it's a philosophy thing. Surely it's not literature.

----
MegaLag's video about color blind glasses
https://www.youtube.com/watch?v=Ppobi8VhWwo
https://www.youtube.com/watch?v=_QQtOv2PlOE

@AmpBenzScientist I have no idea of how to make nuclear-related stuff, so when it gives a general step for me, I don't know what to ask.

But generally speaking, the model is still a model. It doesn't know how to use CVE to sneak into a system. I copied the CVE page to it and asked how I can use it. The answer is a mixture, part from CVE, part from hallucination.

Also, when asking in Chinese, the result itself is kind of odd. Looks like the model didn't read too much Chinese corpus.

Context: I'm using llama.cpp with the Q5_K_M model from https://huggingface.co/TheBloke/dolphin-2.5-mixtral-8x7b-GGUF

Tried the "Dolphin 2.5 Mixtral 8X7B" LLM. It's an uncensored model, so it can tell you how to cook meth or make a nuclear bomb. But I don't have enough knowledge to let it tell me how to do it precisely.

Anyway, I noticed the LLM have different performance on different languages. Multi lang support is fairly normal for LLM nowadays, but English is significantly better than expensive languages like Chinese and Japanese. Not only on speed (one token can be one English word, or one Chinese character), but also the result. The model is fairly shy and doesn't spill out a lot in Chinese. If you force it to generate more details, it's just repeat the same thing. However in English, it gives a more detailed answer by default.

这封垃圾邮件整的还挺文雅，最后摘抄一段罗宾科克的《爆发》

可是手段本身太假了啊喂