This is supported in the underlying data, but we don't expose it in the UI at the moment. We'll probably do something a bit different to help people make those queries.

Would it be possible for you to make the trained vectors available for download and independent analysis without having to rescrape and retrain the model?

Maybe it wasn't clear, but we open-sourced everything, including the trained model. See here:

https://github.com/spacy-io/sense2vec

For installation:

$ pip install -e git+git://github.com/spacy-io/sense2vec.git#egg=sense2vec

$ python -m sense2vec.download

This downloads and installs the sense2vec package and model we used in the demo.

We are currently working on finishing the PyPI package. Meanwhile only Linux is supported and the docs are pretty much non-existing. Also, please make sure you have a recent Blas/Atlas package installed (RedHat: atlas, atlas-devel)

Direct link to the model file (~600MB):

https://index.spacy.io/models/reddit_vectors-1.0.1/archive.g...

FYI there are Reddit data dumps publicly available, no need to kill Reddit scraping it:

https://archive.org/details/2015_reddit_comments_corpus

Awesome.