Comment by ijk

In aggregate? Signs point to yes. For the general purpose SFT base models. We see some evidence even with RNNs vs Transformers. You're essentially finding a function that models language. Use the same optimization function, get a similar result.

However, the RL and especially the RLHF does a lot to reshape the responses, and that's potentially a lot more varied. For the training that wasn't just cribbed from ChatGPT, anyway.

Lastly, it's unlikely that you'll get the _exact same_ responses; there's too many variables at inference time alone. And as for training, we can fingerprint models by their vocabulary to a certain extent. So in practical terms there's probably always going to be some differences.

This assumes our current training approaches don't change too drastically, of course.