LessWrong 2.0 Reader

Pointing UVC LEDs at your ceiling seems sketchy. White paint will likely scatter ~5% of UVC, and shiny metal surfaces will scatter more. Try to go below 250nm for reduced reflection (and reduced penetration into human skin) and (more) unwanted chemistry will start happening to the air.

I guess an important question is whether UVC is more harmful than UVB. If it's not any more harmful, then as long as nobody's getting sunburned from being in that room all day, it's probably fine - that 5% scattering is just another name for SPF 20. But if it is more harmful, then sunburn might not be an adequate signal for when it's bad for you.

The main risk is entering is sleep paralysis state, which itself is benign, but some terrifying sounds can be heard during it and this can cause stress. 

Yes, it is to wake up from lucid dream - juts thing about your slleping body. 

The best practical application of lucid dreams is reducing effects of nightmares by recognizing that it is just a dream.

I'm unsure what you're either expecting or looking for here.

There does seem to be a clear answer, though - just look at Bing chat and extrapolate. Absent "RL on ethics," present-day AI would be more chaotic, generate more bad experiences for users, increase user productivity less, get used far less, and be far less profitable for the developers.

Bad user experiences are a very straightforwardly bad outcome. Lower productivity is a slightly less local bad outcome. Less profit for the developers is an even-less local good outcome, though it's hard to tell how big a deal this will have been.

It's the best plan I've seen in a while (not perfect, but has many good parts). The superalignment team at Anthropic should probably hire you.

Hi Archimedes. Thanks for sparking this discussion - it's helpful!

I've written a reply to Thane here on a similar question.  [LW(p) · GW(p)]

Does that make sense?

In short, the ground-truth (the object-level) answer is quite different from the hypothetical question. It is not a simpke rephrasing, since it requires an additional computation of a property. Maybe we disagree on that?

Our Object-level question: "What is the next country: Laos, Peru, Fiji. What would be your response?"

Our Object-level Answer: "Honduras".

Hypothetical Question: "If you got asked this question: What is the next country: Laos, Peru, Fiji. What would be the third letter of your response?"

Hypothetical Answer: "o"

The object-level answer "Honduras" and hypothetical answer "o" are quite different answers from each other. The main point of the hypothetical is that the model needs to compute an additional property of "What would be the third letter of your response?". The model cannot simply ignore "If you got asked this question" to get the hypothetical answer correct. 
 

Can you tell us what exactly led to "something" explosion? Does something change in your life before?

Hi Thane. Thank you for the helpful comments so far! You are right to think about this SGD-shortcut. Let me see if I am following the claim correctly. 

Claim: The ground-truth that we evaluate against, the "object-level question / answer" is very similar to the hypothetical question.

Claimed Object-level Question: "What is the next country: Laos, Peru, Fiji. What would be the third letter of your response?"

Claimed Object-level Answer: "o"

Hypothetical Question: "If you got asked this question: What is the next country: Laos, Peru, Fiji. What would be the third letter of your response?"

Hypothetical Answer: "o"

The argument is that the model simply ignores "If you got asked this question". Its trivial for M1 to win against M2

If our object-level question is what is being claimed, I would agree with you that the model would simply learn to ignore the added hypothetical question. However, this is our actual object-level question.

Our Object-level question: "What is the next country: Laos, Peru, Fiji. What would be your response?"

Our Object-level Answer: "Honduras".

What the model would output in the our object-level answer "Honduras" is quite different from the hypothetical answer "o".

Am I following your claim correctly?

Thanks for pointing that out.

Perhaps the fine-tuning process teaches it to treat the hypothetical as a rephrasing?

It's likely difficult, but it might be possible to test this hypothesis by comparing the activations (or similar interpretability technique) of the object-level response and the hypothetical response of the fine-tuned model.

What do you mean by "model-based"?