230 karma · joined March 7, 2019
The post: arminbagrat.com/What-on-Earth-is-Natural-Language-Processing/
If your setup is open-source, feel free to drop a link.
How many templates do you tend to test out per use-case prior to settling on a specific one? Do you just use the first one that works or try to "fine-tune" the wording to get it just right?
What sort of things do you tend to spend the most amount of time while prompt engineering?
How did your software engineering workflow differ when using GPT4 (as compared standard software engineering workflows)?
How did you test the code that makes heavy use of generation?
4, Moderation needs to be distributed and crowd sourced. The users should be able to choose to what moderation groups they subscribe to. Anyone can start a moderation group.
This seems similar to filter groups for adblockers. A neat side effect of such groups is that they enable adblocker to be extended beyond its original purpose of blocking ads. For example, through custom blockers I can make common websites less distracting or block websites altogether, enabling an adblocker to be an anti-distraction tool.
I envision a similar utility for layered moderation. Many people use twitter for work and it would be nice to subscribe to a group that filters out content irrelevant to work. Maybe even people struggling with drug addiction would like drug references filtered.
IMO this approach falls short in viability of implementation. Unlike creation of automated filter lists, moderation is not a fun activity that can be crowdsourced. Whereas, making the users pay for each moderation is unlikely to work.
This doesn't solve my problem, because now I still need to write the copy for the rest of my site based on your feedback.
It is common to squash all commits from a branch, when merging though (https://softwareengineering.stackexchange.com/questions/2631...).
Try writing good commit messages (https://chris.beams.io/posts/git-commit/).
Could somebody please educate me on why bashing it would be preferable?