3 ms·
There's some wild disconnect here. The author found a bug that causes the chatbot to "babble": > The chatbot started out fine – for the first few words in its
by rpdillon 2y ago
There's some wild disconnect here. The author found a bug that causes the chatbot to "babble":
> The chatbot started out fine – for the first few words in its response. Then it descended into a babble-like madness. Which went on and on and on and on and … on. Somehow, it couldn't even stop babbling.
Anyone that's tweaked parameters on local LLMs via e.g. llama.cpp can tell you that it's easy to have a model fall into babbling with the wrong params. I often get this when I try out a new model and restart my server, but my test webUI retains settings from a previous GGUF because I didn't refresh.
But let's be clear: the bug is that the model babbles. But here's the intro that the author justifies with this bug:
> It would appear that the biggest technological innovation since the introduction of the world wide web a generation ago has been productized by a collection of fundamentally unserious people and organizations who appear to have no grasp of what it means to run a software business, nor any desire to implement any of the systems or processes needed to affect that seriousness.
Good grief. I feel like I live in a world where hyperbole is the order of the day, every day.
- mewpmewp2 2y agoTo get babble all you have to do is set Temperature to 2 on gpt-4o. Or any gpt probably. You could also get babble without LLM if you want by using some faker library. Or maybe some lorem ipsum generator. I'll make a report to faker library that what it generates doesn't make sense.
- malux85 2y agoYeah it reads like an inexperienced/junior dev found a bug in a production system for the first time, and has a grandiose vision of themselves as saviour of humanity, rather than realising that bugs in software is the normative state of all software development