pstorm

8 ms·

61.
by pstorm 4y ago
This is cool! I was going to build something similar this weekend, but you beat me to it. I wonder if, in the short term at least, this will affect how we modularize code. I've already started thinking how, for a few instances at least
62.
by pstorm 4y ago
It is 1/100 of a percent [1]. Just a different way of saying .25% [1] https://en.wikipedia.org/wiki/Basis_point
63.
by pstorm 4y ago
I flew two flights last week with a lap-child and didn't see this. The in flight info just indicated to hold on tight in an emergency.
64.
by pstorm 4y ago
I just search Obsidian plugins and there are 3 that might work for you: - https://github.com/louis030195/obsidian-ava - https://github.com/brianpetro/obsidian-smart-connections - https:/&#x2
65.
by pstorm 4y ago
Hey, I tried it out - it's pretty cool. I haven't sprung for GPT4, so I tried really hard to get the tokens under 4k, but never made it. Despite that, the output of this was spot on. If this could utilize the api, it would be even
66.
by pstorm 4y ago
> don't @ me I'm @ing you :) This code only checks for the longest word, not the highest scoring word. It neglects the x3 and x2 tile spaces.
67.
by pstorm 4y ago
That's a trivial example, sure. Think of saying "make this image more red, flip it vertically, then crop the bottom 25%, and finally output it with 80% compression as a jpg". That would take 15 minutes to figure out how to wr
68.
by pstorm 4y ago
You are getting a surprising amount of backlash from this, but I think you are right. There may be better tools for the job, but general tools tend to win out as they get "good enough"
69.
by pstorm 4y ago
I think I have the same bug catcher! No matter how we treat bugs, humans as a whole kill a heck of a lot of bugs. Even unintentionally, like by driving our cars or destroying habitats. While I will conceed that there is some doomsday thinki
70.
by pstorm 4y ago
Think of how we treat ants or cockroaches. If they at all inconvenience us or even appear in our home, we kill them. We go as far as developing specialized poisons they take back to their nests to try to kill the whole colony. It isn't
71.
by pstorm 4y ago
If the context window grows from 32k to 1m, maybe the entire history would fit in context. It could become a cost concern though.
72.
by pstorm 4y ago
True. I should have clarified that the workaround used for many NLP tasks, utilizing libs such as Langchain, will become obsolete. And after further thought, obsolete is wrong. More likely just used for more niche needs within NLP.
73.
by pstorm 4y ago
Good point! I realized after I wrote the comment above, that I will still be using them in a service I'm working on to keep price down, and ideally improve results by providing only relevant info in the prompt
74.
by pstorm 4y ago
Reading the press release, my jaw dropped when I saw 32k. The workaround using a vector database and embeddings will soon be obsolete.
75.
by pstorm 4y ago
It is a way to interact with their chat api: https://platform.openai.com/docs/guides/chat/introduction It already exists, but according to their docs current chatGPT "does not always pay strong attention
76.
by pstorm 4y ago
They won't, but that doesn't mean some other business won't automate legal counsel and assume risk. If, down the line, GPT (or some other model) has empirically been proven to be more accurate than legal assistants and lawyer
77.
by pstorm 4y ago
If React + Node is your preferred stack, I would say NextJS on Vercel + Supabase is the closest to meeting your requirements. You'd still have a local env, but the other points are taken care of. Auth is particularly nice with this com
78.
by pstorm 4y ago
I always assumed it was "f*k you". But after a little digging, it turns out it's not "fight yard" either. According to this comment ( https://www.rockpapershotgun.com/community/profile/dmgky
79.
by pstorm 4y ago
If it's public, Can you share the repo? This is really good, and I'd love to see how you did it. I imagine you are using a vector store/search on the backend, just cutrious!
80.
by pstorm 4y ago
Where do you find someone like this?
81.
by pstorm 4y ago
Curious why you suggest Auth0 and Supabase. What is wrong with Supabase auth?
82.
by pstorm 4y ago
Really cool. I'll check this out for my next project. Also, I chuckled at the giant button that says "Click this gigantic button to experience how Cleavr..." :)
83.
by pstorm 4y ago
This seems really powerful. Nice job! React hooks get a lot of hate, but now that I've spent a couple years with them, I like the modularity and composability of them. This feels familiar in that regard. I'll have to try this out
84.
by pstorm 4y ago
I was wondering why some circles seemed great but got low scores. To test your tip, I just made a square and got 92%