Hacker News Reader

Comment by abixb | Hacker News Reader

The amplifying mirror analogy works best here. LLMs are ultimately a reflection of your own interactions with its weights, the tone you use, the structure with which you construct your prompt, aspects of an issue you tend to focus on, your breadth of vocabulary and world knowledge and whatnot.

People who (carefully) use it as an extension of their own mind and senses will very likely thrive, and those who use it as a replacement for their minds and their senses will struggle.

One of the Claude skills I made Claude itself generate was the 'learning a concept across tiers' skill -- from ELI5 level to a PhD level, and it triggers whenever I ask it a very general question on a complex topic that isn't my bread-and-butter. The fact that I'm able to choose explanation level from a super smart LLM (that's available 24x7) that can explain any topic under the sun would've been mind-bogglingly sci-fi-ish just 4 years ago in 2022.

View on HN

When I get out of my domain, I always ask it to describe things to me "like I'm a senior CS student who doesn't know any industry terms". I find it really easy to skim past the definitions I know and very useful to get the explicit clarification on unfamiliar terminology. Telling it to talk down to you a little also seems to calm down it's tendency to talk like it's trying really hard to convince you it's smart like a 8th grader trying to use every single vocab word they learned last week.

I've been using a prompt that comes from the opposite direction for non-programming stuff: "Assume the user is an expert in all fields; while this is clearly logically untrue on a literal level, the user prefers to see a field's technical terminology and then ask the AI about terms the user does not understand rather than get an inaccurate statement about some issue."

Whether you have to reassure the LLM that this is obviously untrue, I don't know, but they do have a knowledge baseline to know it's not true and I have a sneaking suspicion it would be less effective without that.

This has ended up in some of the most interesting incidental knowledge exploration I've ever done. A recent example is that I was asking about some stretches and it started talking about how useful they are for the sarcomeres, which I had not heard of. Now I have.

I'm not saying this is better, just that it is different. I think there's a time and a place for both approaches.

Off topic, but stretching is absolutely useless for ... stretching. And vital for retaining and enhancing usable range of motion (not static stretching, still, but dynamic stretching / mobility under tension).

The wildest thing is, there's no evidence that I can find that static stretching does... anything? It increases pain tolerance through the range of motion, but not any more than just asking people to try harder etc. And it doesn't build up, so after you stretch once for a given day, you're done.

Edit: after looking into it a bit further there's actually a cochrane review that stretching does bupkis, fascinating. It's also the first "Good" evidence graded cochrane review I've seen.

I think you probably just don't like to stretch.

Can you share the source? This seems dangerously wrong.

I see a specific review making the very narrow claim that it can have detrimental effects on power/strength activities immediately following static stretching.

But by and large I'm finding a great deal of evidence for a wide range of other benefits, particularly in range of motion and injury prevention. I can't find anything widely damning, and I certainly don't see any reviews contesting the validity of the very vast body of research supporting the many benefits of static stretching.

https://www.researchgate.net/publication/377653878_The_benef...

Most of this comment is not true. Static stretching does increase flexibility over time - it’s not the most efficient tool for it, but it’s certainly effective.

Presumably the Cochrane study you’re referring to is “Stretching to prevent or reduce muscle soreness after exercise”, which as it’s title suggests, investigates a narrow question and does not support you claim that “stretching does bupkis”.

"And vital for retaining and enhancing usable range of motion (not static stretching, still, but dynamic stretching / mobility under tension)."

I'm doing it for knee pain. It's possible it's a placebo because there is a slight strength training component to the whole thing (it's several exercises), although the knee pain has persisted through a lot of strengthening of my leg in general. This falls under "don't care" as the program as a whole works and I'm not worried enough about the details to try to optimize it. All the stretches fall under "dynamic stretching under tension" so maybe it's not something covered by the study you reference last.

It’s not a placebo. I’m not sure exactly which Chocrane review the GP is talking about but if it’s the one I’m thinking of, the headline is that stretching doesn’t help muscle soreness [1] (delayed onset muscle soreness specifically).

You should absolutely keep stretching to help with your injury, and there is decades of “sport science” to back that up.

[1] https://www.cochrane.org/evidence/CD004577_stretching-preven... - there’s another one about joint deformities which is far more interesting, but less relevant to this discussion

Static stretching increases tendon length, among other things. The stretch MUST NOT cause pain, and MUST be steady for the duration of stretch. Some suitable duration increments are 10 sec. (absolute minimum), 30 sec., 60 sec. Multi-minute stretches are suitable if they can be held without discomfort.

Notably, longer tendons are counter-productive for energy return. The fastest runners tend to have increased tightness along the backs of their legs.

I can tell you if I don't stretch, my muscles feel bad and stiff. No doubt this is partly because I did karate for a few years, where stretching to increase your range of motion is absolutely required. You're not going to "just try harder" into a high kick.

That's very interesting. It kind of mirrors the experience of having a very smart friend.

Edit: https://xkcd.com/2501/

Sometimes when I want AI to explain something technical, I say "explain it like I'm a junior engineer" -- just to get it to start with the high level like a human being would.

This was part of my system prompt for a while, but the LLM would just make forced, unhelpful CS analogies.

> People who (carefully) use it as an extension of their own mind and senses will very likely thrive, and those who use it as a replacement for their minds and their senses will struggle.

Claude code is entirely vibed. Someone posted some of the prompts they used: barely comprehensible typo-addled half phrases.

This seems to be the opposite of your experience.

I am sure some variant of the infinite monkey theorem applies. Anthropic devs have unlimited access to Claude, so they are no doubt lazy when it comes to prompt writing and just brute-force the agent to do what they want.

This was true some time ago but nowadays I don't get this impression. Seems like whatever style I type in, the LLM is already pre-prompted to respond in "its" "own" "style".

As an unapologetic generalist[1] this has also been my experience. Many tools that would have been "eh maybe if I get bored over Thanksgiving holiday" have become "hold on, gimme fifteen minutes".

Tiny, isolated, but awesomely useful CLI scriptlets, for me, seem to be the sweet spot. Little shining rays spreading out from the veins of my own familiarity.

The downside, the Achilles Heel of LLMs, so far as I can tell, is using the system to assist in maintaining large, sprawling, and largely pointless legacy codebases. Somewhere you have to keep many many many stupid things alive. I swear I can almost sense Claude's frustration with some of this shit. Then you get frustrated, and then Claude wants to agree with you so it acts even more frustrated, and the gyre thus widens. You're just cussing at everything with a machine. Which can be fun - Claude is often surprisingly funny - but not productive.

[1] I know absolutely nothing about positively everything, but have the attention span of a squirrel.

I've found Claude Code absolutely amazing for the sorts of 100-500 line data cleaning/analysis/visualization tasks that used to take me a couple hours to knock out. They're often self contained (boss wants a graphic for a slide or some numbers), and I tell it which packages I would prefer it to use.

On the other hand, I've been using it to make small changes to a ~4000 line codebase, and it takes a lot of wrangling to keep changes in scope.

Today I'm translating a 5,000 line VB6 codebase to C#, and I've been spending the day chopping up the job, passing it to Claude, and manually validating it.

I went to VB.NET first using a previous generation of LLM's (that was quite manual back then) and then from VB.NET to C# or just keeping the VB.NET around worked very well. The code was not highly complex but more than just CRUD. The porting from VB6 to VB.NET included building tests which helped.

ELI5 is the way to go. I have Claude break down high level physics "as if I'm a farmer standing in a field" - works beautifully.

I tried that but Claude thought I was already outstanding in my field.

You literally made me Laughing out Loud :D

> I have Claude break down high level physics "as if I'm a farmer standing in a field" - works beautifully.

Careful now: https://www.youtube.com/watch?v=TMoz3gSXBcY

There's an irony here in using Gemini to summarize this 40 minute video right from the page.

What about breaking down farming as if I'm a physicist standing in a lab?

Farming is physics, silly.

Does it? ChatGPT had no problem telling me that dark energy is the Higgs VEV.

The amplifying mirror is not just a characteristic, but the fundamental driving force of LLMs. With every token it consumes, its primary goal is to understand who you are and what you intend. If you type Hola thats sufficient to tell it so much more than if you type hi.

LLMs do not have goals and cannot understand. Please don't anthropomorphize.

One does not need to be human to understand. Animals understand. Saying “he” or “she” would be closer to anthropomorphism than the parent comment, but still does not sufficiently satisfy the definition of anthropomorphism (giving human form).

I see you pushing back on related topics in a few comment threads. I’m not sure if I can define “understanding” but I think “intelligence” is a fair synonym for this conversation. I see your passion for clarifying that LLMs are not intelligent, and I’m sensing you’re conflating that with “consciousness” and trying to hedge against other people inadvertently making that same error.

Intelligence is not uniquely human. And not constrained to consciousness. Even consciousness itself, under fringe definitions like exhibiting non-deterministic behavior, is not constrained to living beings. Systems, for example, exhibit intelligence all the time - groups of people, colonies of ants, weather systems, self-organized criticality like blooming forest fires or abelian sand piles, evolutionary processes, and so on.

Your counter argument - that LLM responses are “just” statically probable (i.e. predictions) - actually supports this notion. Many intelligent statements we make are probably right or probably wrong, in part or in whole. We as humans are far more than statistical computers, but our intelligence in isolation doesn’t seem far off from that of LLMs (not mechanistically, but rather qualitatively).

3Blue1Brown recently posted a video called something like “Compression is Intelligence”. Worth a watch.

You're arguing semantics, but I think that expanding your definition of intelligence to include ant colonies and weather systems is going to make it completely useless to say anything at all about an LLM.

The problem with all of this is that because an LLM gives a facsimile of language and conversation people believe it can "comprehend" or "understand" or whatever word you want to use. This problem isn't new, AI research has had this issue for decades. The risk is that this makes people misunderstand and offload their decision making to it, treating it like a person to be instructed or reasoned with. It isn't, and can never be.

Your LLM is not "smart" and it cannot "explain", it is a program that can output a statistically average text of an explanation. Don't mistake the capabilities of these models for intelligence.

It explained something to me today. So your point is disproven. What is intelligence then if not explaining things. They can be intelligent without consciousness if that’s what you mean

    It explained something to me today. So your point is disproven.

No. It produced a statistically likely answer.

    What is intelligence then if not explaining things.

"the ability to learn, understand, and make judgments or have opinions that are based on reason"[0]

LLMs by definition cannot learn, understand, make judgements, or have opinions. If you find the model to be a tool that is useful to you, by all means. But do not anthropomorphize.

[0] https://dictionary.cambridge.org/dictionary/english/intellig...

You're just talking semantics. To me if I read text that helps me understand something, that's an explanation. The black box it comes from is irrelevant.

LLMs do learn, understand, and make judgments. You're just pushing back on the underlying mechanism, which instead of reason/sentience is instead statistics and weights. I'd call that intelligence but I don't really care if you want to call it something else.

If you believe that an LLM can understand and make judgements then you have fundamentally misunderstood what an LLM is. I'd strongly suggest you read up on the internals.

That sounds useful. Can you share that skill?

Not the OP but you can whack this into your prompt and get most of the way there: "no jargon goes unearned, nothing gets dumbed down, every abstraction touches ground"

A punchy tricolon containing 2 analogies that don’t quite make sense. That’s some S-tier AI-mimicking. Nice!

One could say that it being an LLMism is...load-bearing :)

Here's the kicker: it's not just load-bearing — it's a whole new paradigm support column.

And honestly, it's amazing.

Clean.

One I got "That settles it, and it flips the picture"

well not very settled then is it

I have been using this project recently to make myself tutorials on things I want to learn: https://github.com/devenjarvis/lathe