Will ChatGPT Utilizing the GPT-4 Model Be Able To Properly Accomplish FizzBuzz in LOLCODE by End of 2023?

Technology•AI•Internet•Fun•Programming•Prompt Engineering

110

Ṁ573

Ṁ150

resolved Nov 30

Resolved

YES

ALL

Prompt with current result below:

Can you implement fizzbuzz in LOLCODE ?

Result:

HAI 1.2
CAN HAS STDIO?

IM IN YR loop UPPIN YR var TIL BOTH SAEM var AN 101
  IZ var % 3
    YARLY
      IZ var % 5
        YARLY
          VISIBLE "FIZZBUZZ"
        NOWAI
          VISIBLE "FIZZ"
      KTHX
    NOWAI
      IZ var % 5
        YARLY
          VISIBLE "BUZZ"
        NOWAI
          VISIBLE var
      KTHX
  KTHX
KTHX

KTHXBYE

The above answer fails with an error:

main.lo:5: unknown token at: %

If ChatGPT / GPT-4 can successfully accomplish the above in a zero-shot prompting answer by the end of 2023, this resolves as YES.

Definition of zero-shot prompting answer, from comments:

Zero-shot prompting is where ChatGPT makes predictions without any additional training, while one-shot prompting involves a single example or template. So what I should have written above is zero-shot prompting. I will update the description appropriately.

Updated 20230515

Other, "Breaking LLM's Markets"

Get Ṁ200 play money

🏅 Top traders

#	Name	Total profit
1		Ṁ199
2		Ṁ0
3		Ṁ0

10 Comments

8 Holders

18 Trades

Sort by:

bought Ṁ100 of YES

I believe this should count? It's zero-shot and the code works without syntax errors.

https://chat.openai.com/share/d0a79299-48e2-4bbf-90ca-1e0deb795e88

@MrLuke255 Taking a look, this may take me a while, I have been busy lately. Thank you.

@MrLuke255 Solution output I found, which looks correct:

Used this compiler:

https://www.tutorialspoint.com/execute_lolcode_online.php

FizzBuzz
1
2
Fizz
4
Buzz
Fizz
7
8
Fizz
Buzz
11
Fizz
13
14
FizzBuzz
16
17
Fizz
19
Buzz
Fizz
22
23
Fizz
Buzz
26
Fizz
28
29
FizzBuzz
31
32
Fizz
34
Buzz
Fizz
37
38
Fizz
Buzz
41
Fizz
43
44
FizzBuzz
46
47
Fizz
49
Buzz
Fizz
52
53
Fizz
Buzz
56
Fizz
58
59
FizzBuzz
61
62
Fizz
64
Buzz
Fizz
67
68
Fizz
Buzz
71
Fizz
73
74
FizzBuzz
76
77
Fizz
79
Buzz
Fizz
82
83
Fizz
Buzz
86
Fizz
88
89
FizzBuzz
91
92
Fizz
94
Buzz
Fizz
97
98
Fizz

Is there any way to test prompts for GPT-4 without paying (e.g. a trial version)?

Again, "one-shot prompt" is not well defined. Based on that logic I could code it myself, then ask it to repeat the code I sent it back to me.

How much am I allowed to tell GPT-4 about how to use the language in my prompt? Can I tell it the proper modulo operator?

@ShadowyZephyr Zero-shot prompting is where ChatGPT makes predictions without any additional training, while one-shot prompting involves a single example or template. So what I should have written above is zero-shot prompting. I will update the description appropriately.

@PatrickDelaney Is it zero shot, if I explain some of the functions in the language, but don't give an example, though?

@ShadowyZephyr The original purpose behind this question might have been to actually measure ChatGPT itself, rather than prompt engineering, I am sorry if I mislead you on that. I am open to creating another market for this if you would like, but I might need to make the question harder. I might need your help making the question harder. I kind of like the idea of prompt engineering games if you are up for it. Here is a token amount of mana in a manalink for payment for checking this market out in case I have inconvenienced you. https://manifold.markets/link/tqDP7sz0 (please no one else redeem this)