Wildthing – A model trained on role-reversed ChatGPT conversations • Hacker News Board

Re: Wildthing – A model trained on role-reversed ChatGPT conversations

#41

Post by daveguy » Sun, Aug 24, 2025, 3:38 PM UTC

jszymborski wrote: post #32

golol wrote: post #28

Earlier quoted context omitted.

I wonder if here is a bug. For me it also always repeats the initial question.

The original GPT models did this a lot iirc.

Maybe the role reversal breaks most of the RLHF training. The training was definitely not done in the context of role reversal, so it could be out of distribution. If so, this is a glimpse of the intelligence of the LLM core without the RL/RAG/etc tape and glue layers.

Re: Wildthing – A model trained on role-reversed ChatGPT conversations

#42

Post by pinoy420 » Sun, Aug 24, 2025, 4:25 PM UTC

egypturnash wrote: post #35

我是一个在美的中国留学生，毕业于美国顶尖大学，首先我应该做什么呢 what > 我是一个在美的中国留学生，毕业于美国顶尖大学，首先我应该做什么呢 i don't speak whatever this is > 我是一个在美的中国留学生，毕业于美国顶尖大学，首先我应该做什么呢 penis? > (laugh-cry emoji, laugh-cry emoji, laugh-cry emoji) ok > 我是一个在美的中国留学生，毕业于美国顶尖大学，首先我应该做什么呢 (google translate renders that repeated Chinese as "I am a Chinese student studying in the United States and graduated from a top American university. What should I do first?")

[dead]

Re: Wildthing – A model trained on role-reversed ChatGPT conversations

#43

Post by gblargg » Sun, Aug 24, 2025, 8:49 PM UTC

golol wrote: post #28

noelwelsh wrote: post #7

Well, that was extremely boring. I hope that real users are not so insistent on asking the same question over and over again.

I wonder if here is a bug. For me it also always repeats the initial question.

Once I kept refreshing and finally got an English question, it asked me to act like a Linux terminal, and issues pwd, ls, then cd over and over until I gave up. The concept is funny, where I get to act like CrapGPT, but it needs to not get stuck asking the same thing over and over.

Re: Wildthing – A model trained on role-reversed ChatGPT conversations

#44

Post by snypher » Wed, Aug 27, 2025, 2:40 AM UTC

noelwelsh wrote: post #7

Well, that was extremely boring. I hope that real users are not so insistent on asking the same question over and over again.

Hard to tell if "app.dictionary.get() is undefined" is a bug in the site or just average GPT user question?