Tenoke

19 ms·

211.
by Tenoke 3y ago
The feelings are just how the underlying probabilities are presented to the conscious part of your thinking. You are clearly not consciously noting what your neurons are actually doing.
212.
by Tenoke 3y ago
Yes, it makes sense except that the rating will be provisional and the variance really high until you have a lot of comparisons.
213.
by Tenoke 3y ago
People are reporting that if you bring a bunch of substances they only at best check 1 if you are one of the first and that there's an absurd limit of only 12-40 (I believe) total samples checked per week across all users for Berlin wh
214.
by Tenoke 3y ago
It seems rather naive to think that's the reason why there was so much progress then rather than say due to there being more low hanging fruit.
215.
by Tenoke 3y ago
You can with paid Colab. You just can't on the free tier.
216.
by Tenoke 3y ago
I (sadly) doubt Lemmy will be it in terms of a reddit replacement (if reddit is to actually lose it's users). I just really hope it's not discord who will replace it but I fear they have a good shot. And while discord is great
217.
by Tenoke 3y ago
I was playing with it yesterday and it's not bad. I'd much rather use it for e.g. YouTube videos than risk getting copyright claimed for using something that already exists.
218.
by Tenoke 3y ago
This doesn't download the context and I want the post text and the comments surrounding mine at minimum.
219.
by Tenoke 3y ago
Does anyone have a quick and easy to use package for downloading all your comments and the comments surrounding them for context? Last time I looked into it, it was a massive pain and I definitely wouldn't want to just delete and los
220.
by Tenoke 3y ago
>Excluding child mortality, the average life expectancy during the 12th–19th centuries was approximately 55 years So still worse by whole decades than now. https://en.m.wikipedia.org/wiki/Life_expectancy#:~:text=Excl
221.
by Tenoke 3y ago
>till got to lower prices than bitcoin transactions Visa and Mastercard have 1+% fee, so it's lower for some transactions than bitcoin but higher for others.
222.
by Tenoke 3y ago
The Simulation Theory doesn't break any known physics and still abides by the same framworks as to why the universe or life etc. came to be. It's not about explaining why we are here like Creatonism, but a probabilistic argument a
223.
by Tenoke 3y ago
There is this tapping on your head trick[0] that legitimately works for many people (me included) to clear up your tinitius but only for a few seconds after it. If you are curious what silence sounds you can try it. 0. https://ww
224.
by Tenoke 4y ago
A better comparison would be other EU countries and I'm seeing food inflation estimates for ~22% in Germany at the top of Google. 0. https://tradingeconomics.com/germany/food-inflation#:~:text=... . 1. https:&#x2F
225.
by Tenoke 4y ago
I finetuned one in 2020[0] to play around with and the results still seemed a bit worse than a gpt of comparable size. 0. https://svilentodorov.xyz/blog/reformer-99m/
226.
by Tenoke 4y ago
The quality with reformer is much much worse, it's not really comparable.
227.
The Three Factors Framework for Goal Achievement(svilentodorov.xyz)
1 points by Tenoke 4y ago|0 comments
228.
Exploring Lichess data on improvement using e4 vs. d4 and blitz vs. rapid(svilentodorov.xyz)
1 points by Tenoke 4y ago|0 comments
229.
by Tenoke 4y ago
99.9% of the time sharing mice studies has negative value. Too many of those findings don't pan out for humans for non-researchers to get anything useful out of.
230.
by Tenoke 4y ago
I played a game against it yesterday (it won) and the only time it made an ilegal was move 15 (the game was unique according to lichess database from much earlier) so I just asked it to try again. There's variance in what you get but y
231.
by Tenoke 4y ago
In the bottom it should say the version. Does it say March 14th version (gpt-4) or March 13th version (gpt-3.5)?
232.
by Tenoke 4y ago
Because there were large models trained on huge amounts of data yesterday yet they couldn't do it.
233.
by Tenoke 4y ago
I'd much prefer a lower chance of dying to more accountability for whoever is responsible but higher chance.
234.
by Tenoke 4y ago
A big motivation, and at minimum a solid chunk of the RLHF tuning does seem to have been alignment-focused.
235.
by Tenoke 4y ago
You cannot buy USD on chain to directly transact with it and other crypto, no.
236.
by Tenoke 4y ago
The API is useful because not everyone has quick 10+gb vram gpus lying around.
237.
by Tenoke 4y ago
I've tried a lot of nootoripcs, most I give up on but lion's mane noticably reduces my brain fog.
238.
by Tenoke 4y ago
I've tried it as well last year (free) and it starts off meh but it's kind of cool in a tamagochi-like way how you can train it to care and talk about specific things more. It just takes a while. I deleted it but thought that if t
239.
by Tenoke 4y ago
With enough tries it gives wrong ones to the exact same question, too so I don't see an improvement in that direction.
240.
by Tenoke 4y ago
I'm doing it via the openai library in the way they have in its docs. >completion = openai.ChatCompletion.create(model="gpt-3.5-turbo", messages=[{"role": "user", "content": "Who is Scott

More ›