They mention pretraining too, which surprises me. I thought that was prohibitively expensive?
It's feasible for small models but, I thought small models were not reliable for factual information?
Typical stages of training for these models are:
Foundational:
- Pretraining - Mid/post-training (SFT) - RLHF or alignment post-training (RL)
And sometimes...
- Some more customer-specific fine-tuning.
Note that any supervised fine-tuning following the Pretraining stage is just swapping the dataset and maybe tweaking some of the optimiser settings. Presumably they're talking about this kind of pre-RL fine-tuning instead of post-RL fine-tuning, and not about swapping out the Pretraining stage entirely.