Once again, an LLM has a "meltdown". This time GPT-5 Mini started questioning it's own knowledge upon providing a snippet with non existing syntax.
Probably pronouncable version of above: Em-Em-Es-Ek
I'm not exactly sure what happened here but trying that prompt again triggered this strange "meltdown" with Llama 3 LLM (through DuckDuckGo) again.
(In terms of prompt itself... I was experimenting, trust me. Though if someone knows a good rhyme...)
After tinkering with prompt a little it seems that "quitting" makes model answer with "more serious answer" within it's original answer.
Other models seem to behave normally with this prompt.
I think, perhaps, it's time for our first ever Mastodon Steam game key competition don't you? Can't let the other side have all the fun.
Follow and Boost this to be in to win a copy of (winner picks 2!):
- My Time at Portia
- Sid Meier's Civilization VI
- Hot Wheels Unleashed
- Embr
Winner will be tagged to see what they want. Ends 9PM UTC, 18th December.