How do we reproduce your stories, by a more practical Q&A chat model?

[oliverbob]

If we can create models targeted to a custom dataset but not about "stories", then perhaps it would become useful.

For instance, how do we make it work for a Bible/scriptures dataset. And secondly, will it be reliable enough to produce the right results on a given prompt? It is more practical to have samples that can accurately respond to conversational prompts.

Thanks.

[GilesBathgate]

#462