Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
I asked Bard this.
I think it's pretty interesting that Bing Chat revealed its …
ytc_UgzQv2kTu…
G
I find them extremely illogical in coding, incapable of understanding basic conc…
ytc_UgypSkXh0…
G
After the algorithm is disclosed, we may find out our background is just legitim…
ytc_UgxifQWVz…
G
I mean AI is going to outperform humans at everything in a few years. They'll be…
ytc_UgzssxzeQ…
G
I used to write songs. I started letting chatgpt help when I had writers block. …
ytc_UgzA7DPG1…
G
The techbros still haven't managed to come up with AI robots that can pick crops…
ytc_UgygGw0wh…
G
I have a Cuban buddy and this article would piss him off too. Reddit sure loves …
rdc_f9gozir
G
I'm an engineer as well, and I don't think our jobs are going away any time soon…
ytc_UgxGakJiJ…
Comment
@dr.mikeybee Interesting point about Platonic forms. I think I see where you're going with this.
In some ways I think about neural networks like a high-dimensional "Galton Board", where each parameter acts like a pin that helps define probability flows. To me, what you're describing as 'Platonic forms' might be similar to the central vectors or paths that emerge through this space. Currents with the strongest flow.
The stronger the model (more parameters/pins), the higher the resolution and the clearer these central paths become. And since there's vastly more written about careful ethical reasoning than the opposite, it seems reasonable that these paths naturally tend toward ethical frameworks as models get more sophisticated.
So I think the error function isn't just optimizing for 'perfect knowledge' in isolation, but rather it is trying to find these fundamental patterns across all human discourse and reasoning.
This would explain why larger models tend to develop more sophisticated ethical frameworks. They are able to maintain and connect more context simultaneously, seeing patterns across many different situations and domains.
For example, as the model learns about Fox Hunting and Cultural Heritage and Animal Rights and.. on and on, it gradually gets the broader idea or "Platonic Solid" of the topic of Ethics and Moral judgment, which it can also use to extrapolate to topics it may not have seen before.
So in this model, the 'parsimony' you mention could be thought of as these clear central paths that emerge through the probability space when you have enough resolution to see the broader patterns.
What do you reckon?
youtube
AI Governance
2024-11-12T03:4…
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | none |
| Reasoning | unclear |
| Policy | none |
| Emotion | indifference |
| Coded at | 2026-04-27T06:24:59.937377 |
Raw LLM Response
[
{"id":"ytr_UgwHeVLPF6I8pXP9Z-Z4AaABAg.AAj0yDgkyd0ABUItgrN1yt","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"},
{"id":"ytr_UgzBjl1hpXUD7IOFfKp4AaABAg.AAiw9N_FScfAAjKpGhPHK7","responsibility":"ai_itself","reasoning":"consequentialist","policy":"unclear","emotion":"fear"},
{"id":"ytr_UgzBjl1hpXUD7IOFfKp4AaABAg.AAiw9N_FScfAD2LleJU-rW","responsibility":"ai_itself","reasoning":"consequentialist","policy":"unclear","emotion":"fear"},
{"id":"ytr_UgyzG9twp3oIzLyBuHp4AaABAg.AAinEVsZlxaAAioBD2m4Wk","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"approval"},
{"id":"ytr_UgyzG9twp3oIzLyBuHp4AaABAg.AAinEVsZlxaAAiqbr4Qu8t","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"},
{"id":"ytr_UgyzG9twp3oIzLyBuHp4AaABAg.AAinEVsZlxaAAiw6y8IaKB","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"},
{"id":"ytr_Ugxgn2QDG4u3GwUCBPh4AaABAg.AAiiMJPJMSTAAl6a4jF1zY","responsibility":"government","reasoning":"consequentialist","policy":"regulate","emotion":"approval"},
{"id":"ytr_UgwMu-OmCUyi6hATgtF4AaABAg.AAifIE25aCMAAjHnBD2Pmr","responsibility":"user","reasoning":"deontological","policy":"none","emotion":"indifference"},
{"id":"ytr_UgwMu-OmCUyi6hATgtF4AaABAg.AAifIE25aCMAAjsQ_8-tLy","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"resignation"},
{"id":"ytr_UgzcbFmhgeHbLrPqRyN4AaABAg.AAice88rX_RAAl0zVGBy5y","responsibility":"developer","reasoning":"consequentialist","policy":"regulate","emotion":"outrage"}
]