6 ms·
Having the chat logs enter the training data and having them have a meaningful influence on the ultimate result the model produces are very different things. T
by CSMastermind 8d ago
Having the chat logs enter the training data and having them have a meaningful influence on the ultimate result the model produces are very different things.
The text for all the Goosebumps books are certainly in the training data and to some small amount influenced the solve. But their contribution was so vanishingly small it would seem absurd to say R L Stein should have recourse for contibuting to the solve.
- pbmonster 8d agoBut this is different, right? The equivalent would be taking a (fully offline) LLM and asking it about the ending of one specific Goosebumps book, and it revealing the twist. And although that specific book was (probably) only once in the training data, a high parameter LLM can usually "remember" the twist.
- derangedHorse 7d agoThe only way it would be able to tell you the ending is if it was somehow given more importance in pretraining, loaded into context, or represented in multiple sets of training samples. I have a blog that I make very LLM friendly and usually load posts up into context when I’m working on something relevant. I’ve also opted to improve models for everyone. Despite this, the model can’t recognize my site or any of my posts when I ask it to recall without internet usage (I also turn memory off btw).
- pbmonster 7d agoThat's not correct for a SOTA model with trillions of parameters. Those have immense amounts of knowledge trained into their parameters. Try it. It "remembers" the ending of random books, Goosebumps and otherwise. That's the entire point of why knowledge cutoff is so important (if you use them offline). I find it entirely believable that the Navier-Stokes conversation was auto-flagged as high value training data, and burned-in the models knowlegdge base.