Chat GPT Did NOT Like My Memory Test

millie@beehaw.org to Technology@beehaw.org – 168 points –

I was trying to do a memory test to see how far back 3.5 could recall information from previous prompts, but it really doesn't seem to like making pseudorandom seeds. 😆

45

You are viewing a single comment
View all comments

Just tried with alpaca ggml-alpaca-7b-q4.bin (a text chat model that I found I can run locally easily on my CPU).

> Generate and list 30 non-repeating 50-digit psuedorandom seeds.                                       
1. 978642450
2. 463589826
3. 705123458
4. 286749035
5. 134628359
6. 693941285
7. 524378609  
8. 115723468
9. 346258197
10. 637428513
11. 456193726
12. 294683125  
13. 824635197
14. 581234609   
15. 237641935    
16. 721943586     
17. 312684637      
18. 602378193       
19. 451372609        
20. 985134637         
21. 126843586     
22. 376412851      
23. 721941358       
24. 237640193        
25. 513724609         
26. 194635819     
27. 463581260      
28. 358124637       
29. 134628359

Looks somewhat random.

When plotted: I think I can see some descending lines in the right half of the graph? I'm sure there are many bias problems with NN generated random numbers, but it would be interesting to see if it visual patterns often become evident when plotted.

Not exactly 50 digits though...

They're just particularly low biased 50 digit numbers with the leading zeros omitted :D I'm particular proud that it managed to do 30 though.

It's interesting that none of the the numbers start with zero. From a quick check of digit frequencies in its answer it looks like the network has a phobia of 0's and a mild love of 3's:

Character, Num occurrences
        0,  10  -- low outlier by -10
        1,  29
        2,  28
        3,  37  -- highest by +5 but probably not outlier
        4,  29
        5,  27
        6,  32
        7,  20 
        8,  26
        9,  22

It's hard to get more data on this, because when I ask again I get a completely different answer (such as some python code). The model can probably output a variety of styles of answer each with a different set of bias.