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

Try to guess if code is real or GPT2-generated

doesnotexist.codes

Post reply on HN
Print view View original

41–50 of 108 posts

Re: Try to guess if code is real or GPT2-generated

#41

Post by _coveredInBees »

I got 4/4 GPT-2 guesses right. It is impressive but the "tell" I've found so far is just poor structure in the logic of how something is arranged. For example: a bunch of `if` statements in sequence without any `else` clauses with some directly opposing prior clauses. Another example was repeating the same operation a few times in individual lines of code which most human programmers would write in a simpler manner.

It's harder to do with some of the smaller excerpts though, and I'm sure there are probably examples of terrible human programmers who write worse code than GPT-2.

Re: Try to guess if code is real or GPT2-generated

#44

Post by moyix »

lostmsu wrote: post #32
moyix wrote: post #24

Earlier quoted context omitted.

Sure, it's here in JSON format: https://moyix.net/~moyix/csrc_dataset_large.json.gz

I am curious, how were you able to feed the GPUs? Did you simply preload the entire dataset into RAM (it certainly seems possible)? Did you preapply BPE? Did you train your own BPE?

Yep, I trained my own BPE using HuggingFace's tokenizer library. During training I didn't keep the entire dataset in memory because even on an RTX8000 the full dataset + model weights + data used by the optimizer (ADAM) is too big.

Re: Try to guess if code is real or GPT2-generated

#45

Post by dzdt »

et1337 wrote: post #14

This looks like overfitting to me. Some of the GPT samples were definitely real code, or largely real code. One looked like something from Xorg, another like it was straight from the COLLADA SDK. It’s really hard to define what “truly new code” is, if it’s just the same code copy pasted in different order. Blah blah Ship of Theseus etc.

I got some code related to VICE emulator. It looked pretty real, referring to concepts that make sense in the context of a C64 emulator, but the results said it was GPT not real code. It even had the correct GPL license matching that project. It seems the GPT model has learned quite a bit about the real projects it was fed as input.

Re: Try to guess if code is real or GPT2-generated

#46

Post by emteycz »

skissane wrote: post #40
Felk wrote: post #34

I got a function that assigned the same expression to three variables. Then it declared a void function with documentation stating "returns true on success, false otherwise". Apparently that code was written by a human, which makes me either doubt the correctness of that website, or the quality of the code it was fed with

First code it showed me had getXXX() methods returning void, each of which contained nothing but a printf using the same string variable with no apparent connection to XXX, along with invalid format strings. Surely code this nonsensical has to be generated. Yet when I clicked "GPT2" it said I was wrong.

Don't underestimate the power of failed merges and indifference

Re: Try to guess if code is real or GPT2-generated

#47

Post by dataflow »

psyklic wrote: post #38
Felk wrote: post #34

I got a function that assigned the same expression to three variables. Then it declared a void function with documentation stating "returns true on success, false otherwise". Apparently that code was written by a human, which makes me either doubt the correctness of that website, or the quality of the code it was fed with

Same thought here - apparently humans read from uninitialized arrays immediately after declaring them! That said, it is still a pretty fun website :)

I actually ran into a case where I wanted to do this, but was forced not to.

What was the scenario? I had a couple of small, fixed-size char buffers and I wanted to swap their valid portions, but the obvious choice of swap_ranges(a, b, a + max(na, nb)) would run into this issue. (n.b. this wouldn't be correct for non-POD types anyway, but we're talking about chars.)

On top of it being annoying to not be able to do the convenient thing, it made life harder when debugging, because the "correct" solution does not preservs the bit patterns (0xCC/0xCD or whatever) that the debug build injects into uninitialized arrays, therefore making it harder to tell when I later read an uninitialized element from a swapped-from array.

Re: Try to guess if code is real or GPT2-generated

#48

Post by moyix »

psyklic wrote: post #38
Felk wrote: post #34

I got a function that assigned the same expression to three variables. Then it declared a void function with documentation stating "returns true on success, false otherwise". Apparently that code was written by a human, which makes me either doubt the correctness of that website, or the quality of the code it was fed with

Same thought here - apparently humans read from uninitialized arrays immediately after declaring them! That said, it is still a pretty fun website :)

Since these are snippets from a random position in the file, it's possible that the code that initialized them was outside the snippet?

Re: Try to guess if code is real or GPT2-generated

#49

Post by moyix »

dzdt wrote: post #45
et1337 wrote: post #14

This looks like overfitting to me. Some of the GPT samples were definitely real code, or largely real code. One looked like something from Xorg, another like it was straight from the COLLADA SDK. It’s really hard to define what “truly new code” is, if it’s just the same code copy pasted in different order. Blah blah Ship of Theseus etc.

I got some code related to VICE emulator. It looked pretty real, referring to concepts that make sense in the context of a C64 emulator, but the results said it was GPT not real code. It even had the correct GPL license matching that project. It seems the GPT model has learned quite a bit about the real projects it was fed as input.

It has entirely memorized a bunch of common open source licenses, a bunch of contributor names/emails, and so on. However when I've tried to locate the actual code it's producing in the training data it's not there.

Re: Try to guess if code is real or GPT2-generated

#50

Post by wnoise »

lelandbatey wrote: post #26
hntrader wrote: post #18

Earlier quoted context omitted.

Are you presenting real samples and GPT2 samples to users with equal probabilities? EDIT another poster guessed GPT2 each time and found the frequency was 80 percent

I guessed GPT2 each time, 200 times in a row and only found that GPT2 was correct 89/200 times, so about 45% was GPT2 for me.

In [2]: scipy.stats.binom_test(89, 200, 0.5) Out[2]: 0.13736665086863936

Unusual to be this lopsided (1-in-7), but not crazy.

Post reply on HN