Related: Trying to Try
Measure
Karma: 1,613
Could this be the result of a system prompt telling them that the COT isn’t exposed? Similarly to how they denied that events after their knowledge cutoff could have occurred?
Is this a repost? Date says today, but I’ve read it before.
Both of my comments were about the thought experiment at the end of the post:
You are given a moral dilemma, either a million people will get an experience worth 100 utility points each, or a million + 1 people will get 99 utility points each. The first option gets you more utility total, but if we take the second option we get one more person served and nobody else can even tell the difference.
On average, road speeds went up by a whopping 16%!
But here’s something interesting:
Speeds on highways went up 13%, arterial road speeds went up by 10%, and local road speeds increased by 8%.
None of that’s 16%, and that’s important: This means congestion pricing sped roads up, but also sorted people to faster roads.
In response to having to pay a toll, people not only got off the road, they also made wiser choices about the types of roads they used!
This doesn’t necessarily imply an improvement. If people switched to higher-speed but longer-distance roads, then total trip times could be similar (or more or less).
You start the thought experiment with this:
Lets say people can’t actually tell the difference (in a one-shot trial) between experiencing 100 utility points of goodness, and only 99 utility points.
I’m assuming you don’t mean there is literally no difference between the cases (which would make the answer obvious), but rather that people would be slightly happier in the Q100 case vs. the counterfactual Q99 case. They won’t reliably be able to tell the difference, but there would be a small chance of any individual noticing the improvement. Still, if you multiply that epsilon by 1M trials, you get a noticeable effect.
I’m not sure if a barely-noticeable-difference in ice cream flavor is on the order of 1% of the total utility of a serving of ice cream, but I’m pretty confident that even if it was an order of magnitude less, you’d still be better off creating the 1M x 100 world than the (1M + 1) x (100 - ε) world.
nobody else can even tell the difference
Won’t about 1% of them (10,000 people) be able to tell the difference?
Interesting. I think left/right arrows or triangles would be my most preferred/intuitive of the options.
The web address is duplicated.
Here’s the chat fine-tune. I would not have expected such a dramatic difference. It’s just a subtle difference in post-training; Llama 405b’s hermes-ification didn’t have nearly this much of an effect. I welcome any hypotheses people might have.
This looks like what happens when you turn the contrast way up in an image editor and then play with the brightness. Something behind the scenes is weighting the overall probabilities more toward land, and then there is a layer on top that increases the confidence/lowers variance.
The distinction in the quoted text seems backward to me since the ‘x’ in x-risk refers to ‘existential’, i.e. a risk that we no longer exist (extinction specifically), whereas ‘doom’ seems (to me) to merely imply getting stuck at a bad equilibrium.
When you release the lowered atoms back at the surface, do you have to fight against the atmospheric pressure?
I think the assumption in the problem is that the costs of renting the rooms (hotel and hostel) are either already covered or will be divided equally among the group. There is a cost imposed on the group (of just you and one other person in this case) of one fewer hotel room than expected. The problem is how to distribute this cost. You will get some and the other person will get some. Neither of you will necessarily be happy compared to the expected alternative of each getting a hotel room.
Shouldn’t the hostel-goer forego a 1/n share of their Minimum Satisfactory Compensation (since the other n-1 participants also pay a share)? The best outcome is for the most-willing participant to take the hostel, and the total cost incurred by the group is equal to this person’s MSC. If we want to divide this total cost equally among all group members, the n-1 each pay a 1/n share and the 1 forgoes a 1/n share from their actual compensation.
Alternatively, one could argue that the cost should be distributed proportionally-more to participants who would be more harmed by having to stay in the hostel (since they receive more benefit from the ideal arrangement vs. an alternative of, say, picking someone at random).
How do the incentives work out if everyone pays 1/n of their own bid?
2. You are under no obligation to sacrifice even a tiny amount of win percentage in the game or match to make the game finish faster, if you don’t want to do that.
3. You are dishonorable scum if you play in order to make the game finish slower, in a way you would not behave if this was a fully untimed round.
Why draw the line here above game win percentage and match wp and yet below tournament wp?
The entire class of General Assistants is only 8%, versus 4% for plant identifiers.
This graph looks like it’s just counting the fraction of services in the category rather than having anything to do with revenue.
Think of the derivative of the red curve. It represents something like “for each marginal person who switched their behavior, how many total people would switch after counting the social effects of seeing that person’s switch”. If the slope is less than one, then small effects have even-smaller social effects and fizzle out without a significant change. If the slope is greater than one, then small effects compound, radically shifting the overall expression of support.
You could split each full tile into its four sub-tiles, each with six connection points. Then, each sub-tile can be one of 15 flavors.
Either I’m misunderstanding what you wrote, or you didn’t mean to write what you did.
Suppose A is a human and B is a shrimp.
The value of adding a shrimp to a world where A exists is small.
The value of replacing the shrimp with A is large.