08:57:06

who to follow

🗨️ 1 ♺ 0 🤍 0


09:27:08

is there a simple way to scrape LessWrong? I want all the posts (i have my reasons)

🗨️ 1 ♺ 0 🤍 2


09:52:40

reading now, you should too

🗨️ 0 ♺ 2 🤍 1


10:08:41

SimClusters means we're all living in the influencer dimension now https://t.co/aj8JZXsOZk

🗨️ 1 ♺ 1 🤍 7


10:12:05

basically, Twitter uses a latent space of simulated "communities" as the topology of the network. the top 1% of users are considered "influencers". if a group of influencers have a lot of shared incoming connxns they become a "community"

🗨️ 1 ♺ 0 🤍 4


10:13:57

these simulated clusters go on a gigantic sparse matrix with all types of content: users, tweets, hashtags, "Events". this way they can use nearest-neighbor analysis rather than heavy matrix multiplications, as well as representing all content in the same latent space

🗨️ 1 ♺ 0 🤍 3


10:15:55

all Users can belong to one or more of these Communities, and so can all Tweets. Influencers can change the shape of these communities, through their actions, but only indirectly: attracting or repelling different followers. if you're not an influencer, you're basically a tweet

🗨️ 1 ♺ 0 🤍 3


10:17:44

oh that's right in the linked post lol

🗨️ 1 ♺ 0 🤍 2


10:19:44

all girls born after 1996 know how to do

🗨️ 0 ♺ 1 🤍 6


11:32:13

made chicken noodle soup with green chile yesterday. already paying "the price". am I eating it again today? you bet my ass

🗨️ 0 ♺ 0 🤍 1


12:02:11

Daveed Diggs is a better rapper than Eminem

🗨️ 0 ♺ 0 🤍 0


12:05:23

slapping the TL whole-hand like a prayer wheel

🗨️ 0 ♺ 0 🤍 1


12:18:23

read SimClusters here

🗨️ 0 ♺ 0 🤍 1


13:47:05

running SGRank on a 600,000 word text so I have more time to chill on the porch. why iterate fast when you could iterate chill

🗨️ 1 ♺ 0 🤍 0


13:52:27

(laughs in ghostemane)

🗨️ 1 ♺ 1 🤍 1


13:57:52

I think the laptop is melting

🗨️ 2 ♺ 0 🤍 0


14:09:57

it finished but it didn't "work", haha, here we go again https://t.co/S27P6Nut2G

🗨️ 1 ♺ 0 🤍 1


14:22:09

for all that it doesn't look appreciably different from PosolitionRank results 🤨

🗨️ 1 ♺ 0 🤍 0


14:23:04

position* wtf

🗨️ 0 ♺ 0 🤍 0


14:45:06

running seven million words through a full spaCy pipeline takes a lot of RAM. too much RAM, in fact

🗨️ 0 ♺ 0 🤍 0


15:46:45

about to accelerate this process

🗨️ 0 ♺ 0 🤍 3


15:48:58

what shibboleths are key to the rationalist diaspora? for example, "epistemic status" or "utility function". or "basilisk"

🗨️ 2 ♺ 3 🤍 3


15:50:25

why? no reason... https://t.co/fseVRtlJo9

🗨️ 1 ♺ 0 🤍 0