[voldacar]

This is simply incredible, I would love to read something novel-length like this. But unfortunately it also generates a lot of banal text, as in some of the other examples in this thread. I wonder if there is a way to optimize for the weird shit.

[exikyut]

Convince someone with a really big pile of TPU credits to train a larger-size model on the full text of the SCP Wiki :>

This could even potentially be interesting enough to attract (small-scale) Kickstarter-level attention.

[voldacar]

Well I don't think that would optimize for weird but good prose. Maybe you could train it on Cormac McCarthy or Thomas Pynchon or something like that

[TeenageMemelord]

Woudn't retraining an existing model on the SCP data suffice?