I use full length vectors (512 dimensions) and have seen very fast lookups with pgvector (HNSW index) and sqlite-vec on 20k vectors. I think any decent vector database should be able to handle 44k entries… which one are you using now?
Thanks! Yeah embedding is simple enough and my needs were small enough that I didn’t want to pay. Both llama-server and ollama are great options, and if container size isn’t an issue you get a greater variety running what you want with sentence transformers.
Aren’t most inventions anti-labor by this rationale? We employ fewer farmers than we used to, but this is good for society because people who would be farmers are now working on other things.
> 5 - 9 means strong effects, definitely not placebo.
It’s impossible for anyone to say this convincingly about their own experience. If it were easy to tell whether an effect was due to placebo, we wouldn’t need blinded trials!
I do think it’s common for managers (especially new ones) to desire too much validation from their reports. James Stockdale offered some advice here in a 1980 Proceedings article [1]
> [Leaders] must be fair and they may be compassionate, but they cannot be addicted to being loved by everybody. The man who needs to be loved is an extortionist’s dream. That man will do anything to avoid face-to-face unpleasantness; he will sell his soul down the river for praise. He can be had.
https://ewatchbill.com - a fair schedule generator that uses simulated annealing to minimize a ‘unfairness’ heuristic. I wrote it for my friends in the Navy who have to write duty bills every month.
"If responsibility is rightfully yours, no evasion, or ignorance or passing the blame can shift the burden to someone else. Unless you can point your finger at the man who is responsible when something goes wrong, then you have never had anyone really responsible."