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

LLMs are mortally terrified of exceptions

twitter.com

Post reply on HN
Print view View original

61–70 of 157 posts

Re: LLMs are mortally terrified of exceptions

#61

Post by gnulinux »

bjourne wrote: post #56
karpathy wrote: post #49

Sorry I thought it would be clear and could have clarified that the code itself is just a joke illustrating the point, as an exaggeration. This was the thread if anyone is interested https://chatgpt.com/share/68e82db9-7a28-8007-9a99-bc6f0010d1...

This is stunning English: "Perfect setup for satire. Here’s a Python function that fully commits to the bit — a traumatically over-trained LLM trying to divide numbers while avoiding any conceivable danger:" "Traumatically over-trained", while scoring zero google hits, is an amazingly good description. How can it intuitively know what "traumatic over-training" should mean for LLMs without ever having been taught the…

Hard to know but if you could express "traumatically" as a number, and "over-trained" as a number, it seems like we'd expect "traumatically" + "over-trained" to be close to "traumatically over-trained" as a number. LLMs work in mysterious ways.

Re: LLMs are mortally terrified of exceptions

#62

Post by ziml77 »

That's funny but definitely not far off from reality. I have instructions from my agent to use exceptions but they only help so much.

I really dislike their underuse of exceptions. I'm working on ETL/ELT scripts. Just let stuff blow up on me if something is wrong. Like, that config entry "foo" is required. There's no point in using config.get("foo") with a None check which then prints a message and returns False or whatever. Just use config["foo"] and I'll know what's wrong from the stack trace and exception text.

Re: LLMs are mortally terrified of exceptions

#63

Post by the_gipsy »

drekipus wrote: post #59
bjourne wrote: post #56

Earlier quoted context omitted.

This is stunning English: "Perfect setup for satire. Here’s a Python function that fully commits to the bit — a traumatically over-trained LLM trying to divide numbers while avoiding any conceivable danger:" "Traumatically over-trained", while scoring zero google hits, is an amazingly good description. How can it intuitively know what "traumatic over-training" should mean for LLMs without ever having been taught the…

The same way that you and I think up a word and what it might mean without being taught the concept. Adverb + verb

But the machines cannot possibly have the magic brain-juice!

Re: LLMs are mortally terrified of exceptions

#64

Post by andrewmcwatters »

falcor84 wrote: post #23

That code has many issues, but the one that bothers me the most in practice is this tendency of adding imports inside functions. I can only assume that it's an artifact of them optimizing for a minimal number of edits somewhere in the process, but I expect better.

In very, very large projects, you end up finding that you want lazy initialization as much as possible, because it greatly affects startup times.

Re: LLMs are mortally terrified of exceptions

#65

Post by CGamesPlay »

I dealt with this in my AGENTS.md by including a recap of the text of "Vexing Exceptions" [0], rephrased as a set of guidelines for when to write a throw or catch. I feel like it helped; and when it still emits error handling I disagree with and I ask about it, it will categorize it into one of the four categories, and typically rewrite it in an appropriate way.

I think the Vexing Exceptions post is on the same tier as other seminal works in computer science; definitely worth a quick read or re-read once in a while.

[0] https://ericlippert.com/2008/09/10/vexing-exceptions/

Re: LLMs are mortally terrified of exceptions

#66

Post by natnat »

falcor84 wrote: post #23

That code has many issues, but the one that bothers me the most in practice is this tendency of adding imports inside functions. I can only assume that it's an artifact of them optimizing for a minimal number of edits somewhere in the process, but I expect better.

I think this has a lot to do with the mechanism of RoPE attention, where physical closeness in the code is a signal of relevance.

Re: LLMs are mortally terrified of exceptions

#67

Post by vessenes »

falcor84 wrote: post #29
wffurr wrote: post #7

Turns out computer math is actually super hard. Basic operations entail all kinds of undefined behavior and such. This code is a bit verbose but otherwise familiar.

# Step 3: Preemptively check for catastrophic magnitude differences if abs(a) > sys.float_info.max / 2: logging.warning("Value of a might cause overflow. Returning infinity just to be sure") return math.copysign (float('inf'), a) if abs(b) Does the above code make any sense? I've not worked with this sort of stuff before, but it seems entirely unreasonable to me to check them individually. E.g. if 1 < b < a, then it…

It’s parody

Re: LLMs are mortally terrified of exceptions

#68

Post by miki123211 »

chis wrote: post #52
karpathy wrote: post #49

Sorry I thought it would be clear and could have clarified that the code itself is just a joke illustrating the point, as an exaggeration. This was the thread if anyone is interested https://chatgpt.com/share/68e82db9-7a28-8007-9a99-bc6f0010d1...

I think there’s always a danger of these foundational model companies doing RLHF on non-expert users, and this feels like a case of that. The AIs in general feel really focused on making the user happy - your example, and another one is how they love adding emojis to the stout and over-commenting simple code.

This feels like RLVR, not RLHF.

With RLVR, the LLM is trained to pursue "verified rewards." On coding tasks, the reward is usually something like the percentage of passing tests.

Let's say you have some code that iterates over a set of files and does processing on them. The way a normal dev would write it, an exception in that code would crash the entire program. If you swallow and log the exception, however, you can continue processing the remaining files. This is an easy way to get "number of files successfully processed" up, without actually making your code any better.

Re: LLMs are mortally terrified of exceptions

#69

Post by jcelerier »

bjourne wrote: post #56
karpathy wrote: post #49

Sorry I thought it would be clear and could have clarified that the code itself is just a joke illustrating the point, as an exaggeration. This was the thread if anyone is interested https://chatgpt.com/share/68e82db9-7a28-8007-9a99-bc6f0010d1...

This is stunning English: "Perfect setup for satire. Here’s a Python function that fully commits to the bit — a traumatically over-trained LLM trying to divide numbers while avoiding any conceivable danger:" "Traumatically over-trained", while scoring zero google hits, is an amazingly good description. How can it intuitively know what "traumatic over-training" should mean for LLMs without ever having been taught the…

LLMs operate at token level, not word. it doesn't operate in terms of "traumatic", "over-training", "over" or "training", but rather "tr" "aum" "at" "ic, ", etc.

Re: LLMs are mortally terrified of exceptions

#70

Post by furyofantares »

A couple thoughts.

One is that often I do want error handling, but also often I either know the error just won't happen or if it does, something is very wrong and we should just crash fast to make it easy to fix the bug.

But I am not really sure I would expect someone to know the difference in all cases just looking at some code. This is often an about holistically knowing how the app works.

A second thought - remember the experiment where an LLM was fine tuned on bad code (exploitable security problems for example) and the LLM became broadly misaligned on all sorts of unrelated (non-coding) tasks/contexts? It's as if "good or bad" alignment is encoded as a pretty general concept.

Error-handling is good aligned, which I think is why, even with lots of instructions to fail fast, it's still hard to get the LLM to allow crashing by avoiding error checking. It's gonna be even harder if you do want it to do some error checking, and the code it's looking at has some error checking

Post reply on HN