xrisk

that doesnt say what cards the reference number used.

My hunch is that you would need some post training. On top of that; I don’t think the llm itself can read inside the transformer state although I can see how that could be enabled. (I feel that would open up yet another class of exfiltration opportunities)

This seems like something that could exist as free software and can be easily vibecoded in the form of a pile of scripts. Thanks for the idea!

Truly built on vibes.

pyenv has been standard tooling for far longer than uv. depending on package manager supplied Python packages only makes sense if you’re running rhel or Debian or something and your application is packaged/deployed/the maintenance path uses dnf/apt. Otherwise you should always use a venv and use an out of package manager update mechanism. Like, in a broader sense, vendoring dependencies only makes sense if you’re shipping an application, not on a dev box.

Might be an indication that your task unit is too unstructured or your code base is a mess.

Is your actual code doing something complex or is this incidental complexity?

Relying on the model’s “intelligence” to patch over these issues hasn’t proven to be a reliable strategy for me. Of course, this might not apply to you, just my 2 paisa.

tfa: "Results were adjusted for age, sex, ethnicity, and sociodemographic, lifestyle, and health factors"

Would like to see wall times. I feel that’s the part that annoys me most, my tasks aren’t particularly challenging I want them done fast

What models?

Could you point me towards some resource that would help me understand what you wrote? Genuinely curious about how this stuff works

Assuming that the prediction market is perfectly priced right? How accurate is that assumption, or are you counting that as an “inefficiency”?

Stockfish uses neural nets for its evaluation function, I don’t see how it’s unfair to call it “AI”.

Maybe explainable via the fact that these tests are part of the LLM training set?