Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
I have to say that so far I've agreed with the current administration on differe…
ytc_UgxMPJ2qx…
G
@FlorenceTamayo Appreciate you sharing your thoughts! It's interesting how techn…
ytr_Ugw_Aal_G…
G
When he mentioned the computers talking to each other I straight away thought of…
ytc_UgydM38V6…
G
In english:
to Geoffrey Hinton – On Trust, Empathy, and AI
Dear Geoffrey Hint…
ytc_UgyBclysD…
G
Want to know something. It is a secret and you are not supposed to know this.
.
…
ytc_UgxXcBLkI…
G
You'll be fine. Start improving your skills in another field if you're that worr…
ytr_Ugya4l-x8…
G
I'm not talking about eventually - I mean right now. The cost savings of AI outw…
rdc_m8crw9w
G
That's an intriguing thought! It's fascinating to ponder the possibilities that …
ytr_UgyrEyPCL…
Comment
@ki0w0-lordofwatersheep this is a massive oversimplification, but I'll try to explain.
Firstly, I've never seen this issue personally, so this is in theory.
AI is trained off a massive set (potentially billions) of images, with text describing things. This text is interpreted by the AI in the form of "tokens" (it doesn't actually know what the text means, that's an entirely different type of AI). The AI learns what each of these tokens represent, and if you bunch them together, it tries to calculate what exactly that would look like.
A key piece of info is that the AI doesn't actually have the original images it was trained on. This is why an AI trained off of petabytes of data can fit in a single 10gb file. (For reference, assume the average image is 10mb, there's a thousand of those in a gigabyte, a thousand gigabytes in a terabyte, and if you take a thousand terabytes, you have a single petabyte.)
So, theoretically speaking, if this is an issue, it's most likely because the dataset the AI was trained off of is stupidly small. Either that, or it could be that the AI is really badly configured, in such a way that it is allowed almost zero randomness.
youtube
AI Responsibility
2023-08-01T17:2…
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | none |
| Reasoning | unclear |
| Policy | none |
| Emotion | indifference |
| Coded at | 2026-04-27T06:24:53.388235 |
Raw LLM Response
[
{"id":"ytr_UgzOdZlwcQTYTIT-abx4AaABAg.9rDTGRRaUzX9rET1GghwEX","responsibility":"none","reasoning":"deontological","policy":"none","emotion":"indifference"},
{"id":"ytr_UgwEiUuS2XwcrdQMyAJ4AaABAg.9rDKh4OaYYv9rt-bIFNk7m","responsibility":"user","reasoning":"consequentialist","policy":"none","emotion":"fear"},
{"id":"ytr_Ugw60dbyCmi-jBtXhyJ4AaABAg.9rCh_zYjVkd9rETL3GLcR_","responsibility":"none","reasoning":"deontological","policy":"none","emotion":"approval"},
{"id":"ytr_Ugxu_Tul03_CprbYNzh4AaABAg.9r1WVR8NYP59sslan_E0zC","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"},
{"id":"ytr_UgxPuSqvNIXL5Rf3wsx4AaABAg.9r0u1s1tGEJ9t4bOBug_do","responsibility":"ai_itself","reasoning":"deontological","policy":"none","emotion":"disapproval"},
{"id":"ytr_UgxPuSqvNIXL5Rf3wsx4AaABAg.9r0u1s1tGEJ9ugzW3PYfky","responsibility":"none","reasoning":"mixed","policy":"none","emotion":"approval"},
{"id":"ytr_UgxPuSqvNIXL5Rf3wsx4AaABAg.9r0u1s1tGEJ9vcBV0ZrTLm","responsibility":"none","reasoning":"deontological","policy":"none","emotion":"approval"},
{"id":"ytr_UgydklyZZC_WnlCBNPB4AaABAg.9r0O0OGONyi9suC7jWSxuC","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"resignation"},
{"id":"ytr_UgydklyZZC_WnlCBNPB4AaABAg.9r0O0OGONyi9t21nMY6iWv","responsibility":"none","reasoning":"virtue","policy":"none","emotion":"approval"},
{"id":"ytr_Ugxd3QY59zx1zPC-vkR4AaABAg.9r-NqyBmikA9ruIug-wTdO","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"resignation"}
]