archibaldJ

What if he only took it before his career, or before joining Netflix? And is it because his role is not an engineer/designer/etc so the history of having taken medically prescribed ketamine could mean greater risk during the operations designated to him that are of business-type nature?

What if my engineers microdose on ketamine for productivity? And will the HR decides differently if it is lsd, dmt, etc? So because of Steve Jobs these naturally have better "branding" to them?

What if Kevin Baillie made a documentary film out of this and became "the Steve Jobs of ketamine"? Do things on this planet work this way?

put them into a Cartesian Closed Category and you an emulate the whole thing with LLMs.

I still think it can somewhat be argued that what Claude Code is doing may still be considered justifable. Dark patterns like this in business practices are not rare and isn't this like the mildest kind already.

But I wonder if there will ever be a day when VSCode, etc, would decide to engage in similiar practice but for the collection of business & research intels, etc... that will be the true cyberpunk era and the information dark age.

This can be solved when the ARC puzzle is cracked (https://arcprize.org/play) so we can automate correctness-checking like in coq but for program synthesis.

Anyone else having trouble making sense of Figure 5 (model-proposed task and response of predict input)?

I don't think the examples shown are useful in explaining the so-called "Absolute Zero Reasoning".

I think the interesting thing here though is the notion of `am a jealous God`.

Note: Exodus 20:5 and Similar Passages: The concept of "visiting the iniquity of the fathers on the children to the third and fourth generation" is often understood in the context of the collective nature of ancient Israelite society. It emphasizes the idea that the consequences of sin can affect subsequent generations, particularly in a communal and covenantal context. This does not necessarily mean direct punishment, but rather the natural repercussions and influence of one generation's behavior on the next.

So one interpretation is that: these people's descendants are not directly punished bust just less favorable by God given the jealous nature of God (i.e. the jealous nature to the degree that is expressed in Exodus 20:5 if we take Exodus 20:5 at face value as composed by Moses at the cultural-political time when he received the Commandments where God proclaimed that He is a jealous God)

On the other hand, regarding Ezekiel 18:20, it is a verse that is a part of a broader discussion in Ezekiel 18 that emphasizes individual responsibility when it was composed by the prophet Ezekiel ~6th century BCE, during the Babylonian exile. So once again, one good interpretation can be that due to the cultural-political background at that time, God's message to humanity (for the betterment of humanity) was in a tone where He emphasized less on His jealous aspect and more on the individualism of sins - which is interestingly very similar to the idea of karma formulated in Shakyamuni's Dharma, which was also around that time in human history.

Nah.. it doesn't work that way.

True Innovation comes from the heart, and most mainland Chinese have lost touch of their hearts - you can even see that foreshadowed in ài the character for Love (爱/愛)。The simplified version just pretended the heart 心 is not in the center of it..

You may have step-by-step progress in China, like going from 110 to 111.

But there have never been and will never be any 0 to 1 kind of innovation in China.. maybe at most gimmicks and copying orchestrated by the Beijing State itself.

thanks for the assurance ;) I will stick to it then, since I’ve already spent time to look into its internal APIs too

I guess I’m just tempted because Docusaurus looks really nice

So how does astro's Starlight compare to it as of Sep 2024? I've invested a bit time to configure Starlight for a new proj but now I'm a bit tempted to switch over

Would love feedbacks from anyone who have been with both Starlight and Docusaurus; thanks!

e.g. are there anything great about Starlight that I should stick to it? SEO-wise, etc.

To be fair that is not the wordings in the original paper. https://arxiv.org/abs/2401.11682

I think the pop-sci writer simply didn’t choose a good wording - the original paper is simply about how the (consciousness-orchestrated?) synchronized activities of millions of neurons may be linked to “cylindrical cavity formed by a myelin sheath can facilitate spontaneous photon emission from the vibrational modes and generate a significant number of entangled photon pairs”.

When we don’t have a way to define consciousness and its (let’s called it) orchestration, the notion of “communication” is none-sensible. It’s like if we look at Shor’s algorithm, its BQP efficiency isn’t really due to any “communication” among the q-bits - but more as a kind of “probabilities collapsing” as they go through these quantum gates.

Td;dr I think the phys.org writer is just trying to make it sound more exciting by unfortunately using a miss-leading word

i think the freedom of zooming in and out is what deeper level of meditation feels like

computationally, what appears as the experience of “consciousness” is thus Anattā (not-self)

https://en.m.wikipedia.org/wiki/Anattā

This looks really interesting; any possibility the researchers will release some code soon ?

good points.

Realistically, it feels like the actual utility of an open-source project is based on:

1. it being educational: so everyone can look into its source & learn from its design pattern, etc, or build upon or borrow parts (eg to be modified) and to be used in their own projects - but the practicality of it will really depend on how decoupled and well-designed the system is

2. in favour of competition (so more possible start-ups / big corps can clone their systems/services) and as consumers we will obviously benefit from that

3. llm can access & train on its source code

I think point 3 is most interesting. And I’m also super curious how true point 2 is and to what extend

Point 1 is really cool too - esp when it is done wonderfully (Linux, React for example) but it really depends on so many levels

As a researcher of LLMs and a student of semiotics, I find this piece to be cultural-politically interesting, especially from a socio-cybernetic perspective

this will go so well with TreeTV : https://youtu.be/Wv-P2aN3-b4?si=g6YLSswosShVcV84

yeah I heard about lib like XState; just not sure what values it actually provides for a team to decide to use it in a real-world app

I don’t really understand what values such design pattern can bring worth (that will outweigh its own complexities)

my experience working with DAGs programmatically (ie not as an abstraction (like in React) but actually handling the edges&nodes of a graph-based abstraction in code) is that it looks nice theoretically but in practice top-down (conceptual) approach like this often tends to over-complicate things

would love to see some real-world examples where an actual dev team, etc, find values in graph-theortical-based abstractions like this

I see. Thanks!

So Phind's main mission is to overtake Google right? ;)

hi; great work. so is this more fine-tuning on Phind-CodeLlama-34B-v2?

will there be api access soon?

also: will it be open-source at some point? thanks

I thought AI advisers were LLMs for the formation of AI government eg https://0a.io

maybe that is what we need - true AI advisers for a blockchain-based decentralised government to combat the exact problems as mentioned in the article

but the question remains: can that actually fix politics? Or will things be the same (in terms of lobbying, etc) or worse off?

no no you came out of the universe and you return to the universe

nothing is free !

everything that made up of you had always been here since the beginning :)

As funny as this is, there is something trippy in what you said.

Actually this is why LSD is so important - it helps you better recognise the voices repeating apple, pear, banana in your everyday life

we are all always about to die:

1. everyday billions of your cells die and are dying

2. death is the single most real/important matter in life, and about to or not really only depends on how relativistic you want to be

3. can always swiftly get real because you never know when your last day is

hmm interesting. I just realized the magical Thread is actually about local maximum i.e. the best foreseeable thing right now that doesn't require much effort to achieve.

Even when we don't think we're pulling the thread, we often are. The only way is to constantly remind yourself of an imaginary/unforeseeable global maximum that you can work hard to acheive..

But even then that could just be another local maximum that I've tricked myself into thinking that's all there is. But if you aim too high, it's exhausting and may have the adverse effects (e.g. you end up failing while also missing out your local maximum) - and so you would want to hang more into the foreseeable local maximum (which may not be too bad a thing to be honest, esp when you are all burnt out).

This is truly the hardest problem in life.

.. and then here is the strange question: is trying to achieve global maxmium and immersing yourself in work and becoming a workaholic (so more get done while time passes quicker) the acutal pulling of the thread?

Nice!

Does traceloop support OpenTelemetry Protocol File Exporter?

I'm the maintainer of Insomnium (https://github.com/ArchGPT/insomnium) and I'm building a prompt orchestration framework for automated software development that I plan to integrate into Insomnium. (The orchestration framework will also be open-source soon) Traceloop cloud looks good but I think for simple cases my users will prefer to have a 100% local solution.

any smooth way to get this work with javascript? would love to use this in a project but my inferences are all in js

hmm intriguing; I wonder how is Cursor/Tabnine doing in this aspect

or are they suffering similiar lose?

I asked them about this in their discord. they didnt give me straight-forward answers. that’s when I knew I was right from the start that this whole thing was a show

https://github.com/ArchGPT/insomnium - and soon with local llm support

disclaimer: I'm the maintainer and my day job is fine-tuning llm

p.s. I will always keep insomnium 100% local & FOSS