MarginaliaFeb 27

Did some local LLM-based labeling of the HN comment corpus, tasking the model to classify how likely it is that each comment is:

* Using AI-like grammar (e.g. it's not X, it's Y)
* Using AI-like markup (e.g. em-dashes)
* Trying to shill something
* Trying to influence public opinion

Still a bit work in progress, but here are some preview data.

Show threadMarginalia
Just noticed I labelled the y-axis documents, and not comments. In my defense, it's almost always documents in my line of work.
Feb 27 at 6:59pmWeb