Save notebook or code outside of an experiment

[bfirsh]

Why

When working in a notebook, a lot of work you do is not training, and it is very easy to lose your work in a notebook. It should be possible to save the state of your notebook or code in Replicate, even if you're not running an experiment.

How

Perhaps there should be some kind of checkpoint abstraction that is not attached to an experiment.

In a notebook, perhaps this could be saved automatically when running a cell. You could imagine scrubbing back and forth in time, where experiments are "milestones" in that timeline.

Related

#309