User permission required: 'View sources' AND 'Review and label'
Although training using 'Search' is not one of the main steps outlined in the Explore phase of the training, it can still be a useful training tool and can be used at any point throughout the training process, though it should be used sparingly. As explained in the article explaining using search in Discover (here), too much search can bias your model, which is something you want to avoid.
Searching for terms or phrases in Explore works the same as searching in Discover (here). One of two key differences is that in Explore you must review and label search results individually, rather than in bulk like Discover.
The other key difference is that Explore provides a helpful approximation of the number of verbatims that match your search terms, as shown in the example below when searching for 'location'.
This can be a useful way to gauge whether there will be sufficient examples for a certain label, by searching for a few relevant terms or phrases and seeing how many approximate matches there are in the dataset.
You can search in Explore by simply typing in your search term in the search box at the top left of the page.
Example search query in Explore
Example search query in Explore on a sentiment enabled dataset