Infinity315

11 ms·

1.
by Infinity315 2mo ago
Keep in mind for Anthropic the goal for them isn't necessarily to solve a problem but to demonstrate/measure the innate capabilities of their model. Many results would be a lot less interesting/impressive for an LLM to accom
2.
by Infinity315 2mo ago
This is useful to me since I delegate most coding tasks to Opus and use Fable for planning.
3.
by Infinity315 10mo ago
So many ideas involving AI just seems to be built off of sci-fi (not in a good way), including this one. Like sci-fi, there are little practical considerations made.
4.
by Infinity315 11mo ago
What? Fundamentally, information can only be so dense. Current models may be inefficient w.r.t. information density, however, there is a lower bound of compute required. As a pathological example, we shouldn't expect a megabyte wort
5.
by Infinity315 1y ago
I thought recycle was the breaking down of the device into its constituents (mostly to recover precious metals or other base materials). In contrast, reuse is where the device is kept mostly intact and used for other purposes such as this.
6.
by Infinity315 1y ago
It might make more sense to think of in terms of expected value. Whilst the probability may be low, the payoff is probably many times the $250M if their startup becomes successful.
7.
by Infinity315 1y ago
That's measuring desirability--or at least immediate desirability, which is sometimes related but distinct from wealth. It also quickly becomes meaningless as people risking life and limb tend to flee to the nearest stable country. Fo
8.
by Infinity315 1y ago
A similar argument could be made for any safety feature that adds cost to vehicles--literally, all of them. If a death is preventable and adds on a relatively inconsequential amount to the cost of a vehicle, then it is the morally correct
9.
by Infinity315 1y ago
Can you answer the question?
10.
by Infinity315 1y ago
How do you define what is needed? The vast majority of reporting does not affect one's own behaviour at all. For example, consider the Snowden leaks, I guarantee you the vast majority of people did not even attempt to modify their ow
11.
by Infinity315 1y ago
We still need people on the ground reporting on these things. AI can't (yet) have connections and conversations with relevant parties. Many interviews would not be possible without a reporter developing a rapport with the interviewee
12.
by Infinity315 1y ago
Just ask any person who works in teaching or any of the numerous faulty AI detectors (they're all faulty). Any current technology which can used to accurately detect pre-AI content would necessarily imply that that same technology coul
13.
by Infinity315 1y ago
Then the solution is targetted tariffs such as the ones they placed on BYD, not across the board tariffs.
14.
by Infinity315 1y ago
For many people, sure. I can always find exceptions to the rule. I am talking statistically. Fact of the matter is, the bottom 20% QoL has improved significantly, correct?
15.
by Infinity315 1y ago
> That short sighted western leadership were so eager to ship off manufacturing jobs to China has been a disaster for the west. In what way are you quantifying this? Would you agree that the quality of life has improved significantly co
16.
by Infinity315 1y ago
The types of manufacturing jobs which would be a step up would be high-tech and high gross-margins manufacturing (which we already do), not whatever crap China is producing. If you want the quality of life of a Chinese manufacturing worker
17.
by Infinity315 1y ago
We do not want manufacturing jobs, they suck. America has an amazing thing going on where we can trade 1s and 0s in exchange for physical goods. We have an infinite renewable resource which costs nearly nothing to reproduce infinitely. We
18.
by Infinity315 1y ago
For whatever reason, people did not believe Trump would be this crazy. In fact, people in the stock market still do not believe he is this crazy. If they did, the stock market would drop much lower. From what I've heard from people i
19.
by Infinity315 2y ago
Yes, but the author (probably) implemented their save system where the client sends the entirety of the save data to the backend; meaning a lot of redundant information is sent. Ideally, the client should only push changes to the backend an
20.
by Infinity315 2y ago
Stock options have legitimate uses. Like all tools, it can be misused. Stock options can be used as a tool to hedge against risk.
21.
by Infinity315 2y ago
This is unsurprising since from what little I know about electronics, I know the ESP32 is pretty common. I know very little about electronics. So if I know it's common, it's for sure in the training data of ChatGPT.
22.
by Infinity315 2y ago
No, this is just how Microsoft names things.
23.
by Infinity315 2y ago
It doesn't seem like the wet bulb model accounts for ratio of surface area and volume. The instrument they use is the sling psychrometer, a long cylindrical tube [1]. If I had to hazard a guess, the ratio of surface area to volume is
24.
by Infinity315 2y ago
I think the results still tell us something. Discrepancies in mathematical ability between the various bases would seem to suggest memorization as opposed to generalization.
25.
by Infinity315 2y ago
The term 'reason' is being overloaded here. The type of 'reason' a cat uses is different from the 'reason' used in math The information a cat uses is incomplete whereas the information used in math and logic is
26.
by Infinity315 2y ago
So at this point are you admitting fully that you have no evidence for your claim? Your entire bet is conjecture? > That does not mean the hypothesis is true, it only means you haven't falsified it. Correct. You can't prove a
27.
by Infinity315 2y ago
I disagree with the assertion that the relation is cyclical. In reality, all of these systems are highly interdependent. I'd model it as a weighted complete digraph. And certain subsets of these various nodes have a greater outsized
28.
by Infinity315 2y ago
Are you not making the positive claim that LLMs can (eventually) generate good incident reports? Please refer to this: https://en.wikipedia.org/wiki/Burden_of_proof_(philosophy) Just to make sure we both understand wha
29.
by Infinity315 2y ago
Wrong. My original wager would be a successful deployment, but sure that could be interpreted as a weasel word. What I mean by successful is that the LLM can generate accurate (>80%) incident response reports and propose correct fixes
30.
by Infinity315 2y ago
> This particular hypothesis is obviously a testable one: someone could simply follow the proposed steps from the hypothesis (e.g. finetune a model on their incident response data), and see if it works. Are you saying that if someone fin

More ›