Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

GPT-Prompt-Engineer

github.com

Post reply on HN
Print view View original

131–140 of 166 posts

Re: GPT-Prompt-Engineer

#131

Post by therein »

QuantumGood wrote: post #127
jstanley wrote: post #72

Earlier quoted context omitted.

That's definitely begging the question. If you're prepared to accept that GPT-4 can answer questions just as well as humans can, why do you even need to do prompt engineering?

Humans benefit from good communication too. For example, annual U.S. deaths from medical errors is in the hundreds of thousands. Much of it is due to miscommunication. Is this akin to poor human-to-human prompt engineering? Of course, humans will rush and not attempt better communication, and you can take all the time you wish with an AI. And AI will continue to incorporate better prompt engineering that you won't ha…

You're forgetting what you may consider to be factual, self-evident and a priori is your opinion.

You may be under the impression that annual U.S. deaths from medical errors being in the hundreds of thousands miscommunicates but that is truly your opinion. You are merely jumping to conclusions at places another person might not.

And going on to rely on the LLM to validate your perspective is a lossy process. It may not lose your perspective but it loses someone else's and you don't even seem to notice or care.

Re: GPT-Prompt-Engineer

#132

Post by jes »

rmilejczz wrote: post #111
jes wrote: post #96

Earlier quoted context omitted.

You might enjoy this article on the Therac-25 [1]. It's kind of the standard example of how errors in software can wind up harming people. I have written medical device software for about 30 years. In my experience, delivering high quality software for Class B and Class C devices is both challenging and expensive. https://en.wikipedia.org/wiki/Therac-25

Every software developer should know this story, it is a humbling and important lesson. Yes luckily most of can’t ship code that accidentally kills people but we can absolutely empathize with the conditions which led to it happening. I recommend reading the entire postmortem, http://sunnyday.mit.edu/papers/therac.pdf yes it is quite long but if you write code in any capacity it’s worth the read

Thank you so much for the link to the postmortem. I will be sharing it and discussing it with my colleagues. We are currently working on the embedded software for an AED.

Re: GPT-Prompt-Engineer

#133

Post by amelius »

oefnak wrote: post #119
albrewer wrote: post #110

Earlier quoted context omitted.

The post you're replying to is a case in point. This time it's cow eggs; what next?

Some people believe the earth is flat, but they can still provide useful work.

These people typically subscribe to a very limited number of conspiracy theories.

Re: GPT-Prompt-Engineer

#134

Post by fatso784 »

typpo wrote: post #120
fatso784 wrote: post #65

This tool doesn’t benchmark based on how a model actually responds to the generated prompts. Instead, it trusts GPT4 to rank prompts simply in terms of how well it imagines they will perform head-to-head. Thus, there’s no way to tell if the chosen ‘best prompt’ actually is the best, because there’s no ground truth against actual responses. Why is this so popular, then (more popular than promptfoo, which I think is a…

Thanks for mentioning promptfoo. For anyone else who might prefer deterministic, programmatic evaluation of LLM outputs, I've been building this for evaluating prompts and models: https://github.com/typpo/promptfoo Example asserts include basic string checks, regex, is-json, cosine similarity, etc. (and LLM self-eval is an option if you'd like).

No problem! I guess I will make a plug myself --we've been working on a similar 'prompt engineering' tool, ChainForge (https://github.com/ianarawjo/ChainForge). It's targeted towards slightly different users and use cases than promptfoo --geared more towards early-stage, 'quick-and-dirty' explorations of differences between prompts and models for less experienced programmers, versus the kind of continuous benchmarking and verification testing power that promptfoo offers.

I particularly like promptfoo's support for CI, which I haven't seen anywhere else, and is very important for developers pushing prompts into production (esp since OpenAI keeps updating their models every few months...).

Re: GPT-Prompt-Engineer

#135

Post by l33t233372 »

FishInTheWater wrote: post #49
brunoluiz wrote: post #34

Isn’t engineering an exact science while prompt engineering is completely not? Although, even software engineering being an exact science, it is a funny one: most of us don’t get certified as like, let’s say, mechanical engineers do. Would they say we are engineers? So perhaps the “engineer” term got overloaded in recent years?

Prompt "engineering" is just writing prayers to forest faeries. Whilst BASIC/JavaScript/etc are all magic incantations to a child, a child will soon figure out there's underlaying logic, and learn the ability to reason about what code does, and what certain changes will do. With prompts, it's all faerie logic. There is nothing to learn, there are only magic incantations that change drastically if the model is updated…

You cannot predict results in _any_ domain with 100% accuracy, especially not in most engineering domains.

Why do you think rockets explode, bridges collapse, etc.

Re: GPT-Prompt-Engineer

#136

Post by jstanley »

therein wrote: post #131
QuantumGood wrote: post #127

Earlier quoted context omitted.

Humans benefit from good communication too. For example, annual U.S. deaths from medical errors is in the hundreds of thousands. Much of it is due to miscommunication. Is this akin to poor human-to-human prompt engineering? Of course, humans will rush and not attempt better communication, and you can take all the time you wish with an AI. And AI will continue to incorporate better prompt engineering that you won't ha…

You're forgetting what you may consider to be factual, self-evident and a priori is your opinion. You may be under the impression that annual U.S. deaths from medical errors being in the hundreds of thousands miscommunicates but that is truly your opinion. You are merely jumping to conclusions at places another person might not. And going on to rely on the LLM to validate your perspective is a lossy process. It may n…

This is an excellent example.

The post you replied to was saying that the deaths were caused by miscommunication, but you interpreted it to mean that stating the number of such deaths is somehow a miscommunication itself!

Re: GPT-Prompt-Engineer

#137

Post by l33t233372 »

FishInTheWater wrote: post #116
broast wrote: post #112

Earlier quoted context omitted.

I don't get where this sentiment comes from. I build software specifically on the concept of predictable results from llm's being composable. Sure, the results are not deterministic in that 100% of the time the exact prompt returns the exact same result, but you can tune your prompts so that 100% of the time they give you a valid result in the result category you were seeking, and with a specific probability distribu…

but you can tune your prompts so that 100% of the time they give you a valid result in the result You can't though, that's the issue. Illustrative here are tokens like "SolidGoldMagikarp", but this does happen to "normal" sequences of tokens as well. There is no filter you can build to keep out such mistakes, any set of otherwise normal tokens could trigger the model to produce wrong output. Because of how large thes…

You can reasonably expect an LLM to respond appropriately often. Which percentage of the time depends on the details, but it’s not much more magic than expecting the bridge you built to hold up.

Re: GPT-Prompt-Engineer

#138

Post by namaria »

l33t233372 wrote: post #128
namaria wrote: post #33

Earlier quoted context omitted.

> Run a bunch of experiments and see which does better. That's empiricism. Scientific method implies formulating falsifiable theoretical claims, producing meaningful analysis of experimental conditions, publishing peer-reviewed and reproducible results.

Does the scientific method include peer review? I thought that was a recent phenomenon.

Peer review has been a thing since, albeit in a much less rigorous way, at least the XVII century. The so-called 'republic of letters' was a world (europe) wide society of philosophers, mathematicians, and early experimenters sharing results and data in writing. Mathematicians have been publishing proofs and challenging each other in Europe since at least the Italian Risorgimento...

Re: GPT-Prompt-Engineer

#139

Post by duskwuff »

fatso784 wrote: post #69
donkeyboy wrote: post #67

Earlier quoted context omitted.

There is a paper on arxiv saying that GPT4 correlation with human evaluators on a variety of tasks with strongly positive. I am also uncomfortable with it, but using GPT4 as a grader is not as bad as you think.

You’re missing the point here. It’s not even getting the LLM’s opinion on evaluating the responses to the prompts (which itself is fraught for some tasks, and benchmarks are known to be limited —even OpenAI admits this, it’s why they made evals). It’s one level abstracted from that. It’s evaluating what the LLM thinks of how well the prompt will do, in purely hypothetical terms. That’s hogwash —different LLMs perform…

> Here is what HN was talking about, nearly three months ago -the exact same type of ‘auto-prompt-gen’ tool.

I was reminded of the same thing. What a lot of it boils down to is that LLMs have no innate ability to self-reflect. They can pretend to do it, but no more effectively than an untrained human would.

Re: GPT-Prompt-Engineer

#140

Post by olddustytrail »

namaria wrote: post #101
judge2020 wrote: post #97

Earlier quoted context omitted.

That's if you're trying to become a scientist. You can do science even if your experiment doesn't contain any of those elements besides trying multiple things and observing the outcomes.

Sure, the same way I can practice medicine by putting leeches on people. I'm not trying to become a physician so it's alright to call that medicine.

Well, not you, but other people could because leeches are still used because they work.

You're trying to correct people on a subject you know nothing about.

Post reply on HN