On: Large language models often know when they are being evaluated

[ofjcihen]

We put the LLM in a loop with no instructions with whatever tools you want. Now what?

[mewpmewp2 → ofjcihen]

We will observe what it would do. We could write a script to try it out.

[Jensson → mewpmewp2]

It just gets into an endless loop. Human brains are ridiculously good at avoiding those somehow, you almost never see a biological brain stop functioning without being physically damaged. The error handling is so very robust.

[mewpmewp2 → Jensson]

Have you tried it already? What is the endless loop it gets into?

[ofjcihen → mewpmewp2]

Sure, so I just tried it with visual and audio input.

It does nothing. Because there is not impetus for it to do anything by itself.

[mewpmewp2 → ofjcihen]

What do you mean by nothing? How did you put the visual and audio input, which model, how did you loop it etc?

[ofjcihen → mewpmewp2]

It’s preferred method of text.

4o

Maintain context and trigger at 1 second intervals.

It has no desires of its own. Nothing that motivates it. It’s not conscious.

[mewpmewp2 → ofjcihen]

It produced no tokens at all?