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

Tell HN: ChatGPT is fantastic for finding and solving issues in logs

news.ycombinator.com

Post reply on HN
Print view View original

141–150 of 184 posts

Re: Tell HN: ChatGPT is fantastic for finding and solving issues in logs

#141

Post by joquarky »

qbasic_forever wrote: post #126
stevenhuang wrote: post #123

Earlier quoted context omitted.

> it's reaching into its training set and stochastically parroting a response Yes, this is how humans work too. Also I hope the irony of trotting out this oft-repeated quip much like a stochastic parrot would isn't lost on you :P

No, humans don't work like your simplistic view of a neural network. I as a human can apply logic and deduction to figure out how something works despite never having experienced or learned about it directly for example. GPT cannot do this and can only guess or make up responses when faced with a problem that requires such reasoning.

> I as a human can apply logic and deduction to figure out how something works

And what is the process for doing this?

Are you using words?

Where do they come from?

Re: Tell HN: ChatGPT is fantastic for finding and solving issues in logs

#142

Post by muzani »

qbasic_forever wrote: post #126
stevenhuang wrote: post #123

Earlier quoted context omitted.

> it's reaching into its training set and stochastically parroting a response Yes, this is how humans work too. Also I hope the irony of trotting out this oft-repeated quip much like a stochastic parrot would isn't lost on you :P

No, humans don't work like your simplistic view of a neural network. I as a human can apply logic and deduction to figure out how something works despite never having experienced or learned about it directly for example. GPT cannot do this and can only guess or make up responses when faced with a problem that requires such reasoning.

Have you used GPT-4 though? It does have reasoning capabilities, better than a clever human.

LLMs are different, but a bunch of transistors can apply reasoning to chess better than any grandmaster. That's emergent behavior.

But you don't really see anyone today trying to argue that power plants don't have the same muscles as a horse, because it gets the job done.

Re: Tell HN: ChatGPT is fantastic for finding and solving issues in logs

#143

Post by bumbledraven »

Aperocky wrote: post #52

Can any of the existing open source models do the same? ChatGPT is great but I don't want all of my queries going to OpenAI. I'd rather shell out a considerable sum to buy the equipment to run my own.

Not even remotely. None of the closed-source models can either.

Re: Tell HN: ChatGPT is fantastic for finding and solving issues in logs

#144

Post by CodexArcana »

simmerup wrote: post #134
d0mine wrote: post #115

Earlier quoted context omitted.

Why do you think your email is private? Is your email provider more aligned with your interests or more secure than OpenAI? I doubt either Google or Microsoft care about your privacy (no difference).

Of course Google and Microsoft is more secure than OpenAI. And Google and Microsoft don't care about your privacy but they do care about being the only ones to profit off your data. OpenAI don't make profit off of your data, but they are collecting it; how they choose to make profit off of it in the future could be completely orthogonal to your interests. Hell, China or any other of the numerous wealthy baddys could…

Homie, Bard from Google is trained on your Google Emails. They read your emails and build data profiles based on that shit and sell it. What are you on exactly? The US Government is more of a direct threat to you and me than the CCP and they actively buy your data and were reading all your emails not too long ago.

Re: Tell HN: ChatGPT is fantastic for finding and solving issues in logs

#145

Post by taf2 »

sangnoir wrote: post #82
simonw wrote: post #70

Earlier quoted context omitted.

I think there's a widely held misconception that anything you paste into GPT-4 will be used as raw training data by the model. Some people even seem to believe that it's learning continuously, so something you paste in could show up in an answer for another user a few minutes later. My mental model of how this works is somewhat different: - It takes months to train a model on raw data, and OpenAI train new ones (that…

I'm absolutely horrified by people's willingness to submit private information (personal or corporate) even if it's not used for training . Data breaches happen all the time (targeted or accidental), and OpenAI is becoming a juicier target by the day. You're right that OpenAI doesn't want the information. Consequently, OpenAI will not have security policies and processes geared for anonymization, or handling financia…

They are starting to execute BAAs so for health care data they are taking it pretty serious

Re: Tell HN: ChatGPT is fantastic for finding and solving issues in logs

#146

Post by ambrozk »

qbasic_forever wrote: post #99
csomar wrote: post #88

Earlier quoted context omitted.

Strong disagree. Will be writing a blog post how GPT-4 understands languages that don't even exist (at least not formally).

It's confabulating (or hallucinating) meaning. GPT has no understanding of anything. Given an input prompt (tell me what this made up language sentence means for example), it's reaching into its training set and stochastically parroting a response. There is no cognition or understanding of language.

People repeating this take in the face of so much overwhelming evidence to the contrary look so ridiculous that at this point, you just have to laugh at them. Yeah, sure, it's not reasoning. That hour-long exchange I just had with it, where it helped me trouble-shoot my totally bespoke DB migration problems step by step, coming up with potential hypotheses, ways of confirming or disconfirming theories, interacting in real time as I gave it new constraints until we collaboratively arrived at a solution -- that was definitely not "reasoning." Why isn't it reasoning? No explanation is ever given. I'm waiting for the moment when someone tells me that it's not "true reasoning" unless it's produced in the Reasoning region of France.

Re: Tell HN: ChatGPT is fantastic for finding and solving issues in logs

#147

Post by qbasic_forever »

ambrozk wrote: post #146
qbasic_forever wrote: post #99

Earlier quoted context omitted.

It's confabulating (or hallucinating) meaning. GPT has no understanding of anything. Given an input prompt (tell me what this made up language sentence means for example), it's reaching into its training set and stochastically parroting a response. There is no cognition or understanding of language.

People repeating this take in the face of so much overwhelming evidence to the contrary look so ridiculous that at this point, you just have to laugh at them. Yeah, sure, it's not reasoning. That hour-long exchange I just had with it, where it helped me trouble-shoot my totally bespoke DB migration problems step by step, coming up with potential hypotheses, ways of confirming or disconfirming theories, interacting in…

It's barfing up text in a form similar to tens of thousands of database troubleshooting guides in its training and filling in context you've given it in your discussion/text prompt. It has no understanding of what a database is and would happily tell you what to name your databases' child if you told it the database was pregnant or other similar nonsense.

Re: Tell HN: ChatGPT is fantastic for finding and solving issues in logs

#148

Post by jklein11 »

taf2 wrote: post #145
sangnoir wrote: post #82

Earlier quoted context omitted.

I'm absolutely horrified by people's willingness to submit private information (personal or corporate) even if it's not used for training . Data breaches happen all the time (targeted or accidental), and OpenAI is becoming a juicier target by the day. You're right that OpenAI doesn't want the information. Consequently, OpenAI will not have security policies and processes geared for anonymization, or handling financia…

They are starting to execute BAAs so for health care data they are taking it pretty serious

Usually BAAs are required for IT vendors from healthcare companies before they start getting paid. It doesn't mean that they are claiming that their systems are HIPAA compliant

Re: Tell HN: ChatGPT is fantastic for finding and solving issues in logs

#149

Post by satvikpendem »

simmerup wrote: post #134
d0mine wrote: post #115

Earlier quoted context omitted.

Why do you think your email is private? Is your email provider more aligned with your interests or more secure than OpenAI? I doubt either Google or Microsoft care about your privacy (no difference).

Of course Google and Microsoft is more secure than OpenAI. And Google and Microsoft don't care about your privacy but they do care about being the only ones to profit off your data. OpenAI don't make profit off of your data, but they are collecting it; how they choose to make profit off of it in the future could be completely orthogonal to your interests. Hell, China or any other of the numerous wealthy baddys could…

Huh? Google Bard is trained on your email data, and Microsoft de facto controls OpenAI and by extension ChatGPT via their 49% investment in the company.

Re: Tell HN: ChatGPT is fantastic for finding and solving issues in logs

#150

Post by ttul »

garysahota93 wrote: post #64
ttul wrote: post #33

I used it to transform cryptic credit card statement items into company names, which then allowed me to query my Gmail archives for receipts and invoices from these vendors, automating a manual process of accounting backup discovery that is the bane of my very existence. I even got GPT-4 to assess whether an email likely relates to an invoice or payment so that I could limit the amount of noise extracted from my emai…

Would you be open to making this logic open source? I'd LOVE to do this as well. (I use LunchMoney.app for all spend tracking, but don't have my receipts logged)

Sure why not. It’s absolutely horrific Python code. Maybe someone else can clean it up 10,000x. I’ll make a note to move it to GitHub.
Post reply on HN