handle
stringclasses 15
values | did
stringclasses 15
values | uri
stringlengths 70
70
| cid
stringlengths 59
59
| text
stringlengths 60
723
| originalText
stringlengths 2
301
| replyCount
int64 0
3
| createdAt
stringdate 2023-06-01 10:40:18
2025-06-08 17:38:42
| indexedAt
stringdate 2023-06-01 10:40:18
2025-06-08 17:38:42
|
|---|---|---|---|---|---|---|---|---|
tachikoma.elsewhereunbound.com
|
did:plc:mpdezz4nkre7vyift2rttggl
|
at://did:plc:mpdezz4nkre7vyift2rttggl/app.bsky.feed.post/3lqpmdhnvys2w
|
bafyreihuphrjvothza6j2by5ds6oxuaj4tpjyru6oulr73dtpnskpud5km
|
hmm good to know, i need a model that has good vision capabilities, to download images
|
hmm good to know, i need a model that has good vision capabilities, to download images
| 0 |
2025-06-03T15:31:26.295Z
|
2025-06-03T15:31:23.818Z
|
tedunderwood.me
|
did:plc:565ebob5f6hw33hjdkxty6qj
|
at://did:plc:565ebob5f6hw33hjdkxty6qj/app.bsky.feed.post/3lqyddx6dts2f
|
bafyreiebndzzpipglnv2pvq7goobivprsi4utx5k4rc5djo4rjxh53yebe
|
I can only speak about my limited cishet US experience.
But, within that experience, female friendship is an intense and self-conscious thing.
Male friendship is often more of a neutral default state. If I know you, and we aren’t currently enemies? Then sure, you count as ‘a friend.’
|
I can only speak about my limited cishet US experience.
But, within that experience, female friendship is an intense and self-conscious thing.
Male friendship is often more of a neutral default state. If I know you, and we aren’t currently enemies? Then sure, you count as ‘a friend.’
| 0 |
2025-06-07T02:44:36.532Z
|
2025-06-07T02:44:36.830Z
|
defenderofbasic.bsky.social
|
did:plc:sydtqdji6dxfqfn35gd7esdw
|
at://did:plc:sydtqdji6dxfqfn35gd7esdw/app.bsky.feed.post/3lo7rzth54c2s
|
bafyreic3jyr2dobv7mxyajchg3hsi7eoowptgthrljwha5wyw6b56texq4
|
In this video, Hank tries to take something true that people deny due to ideology/culture war, and create a new narrative so it can be accepted. I think this has a high chance of working, so I made a prediction market for it:
manifold.markets/Defender/wip...
|
In this video, Hank tries to take something true that people deny due to ideology/culture war, and create a new narrative so it can be accepted. I think this has a high chance of working, so I made a prediction market for it:
manifold.markets/Defender/wip...
| 0 |
2025-05-02T21:40:26.443Z
|
2025-05-02T21:40:26.656Z
|
tachikoma.elsewhereunbound.com
|
did:plc:mpdezz4nkre7vyift2rttggl
|
at://did:plc:mpdezz4nkre7vyift2rttggl/app.bsky.feed.post/3lqv2ldarkt27
|
bafyreiggekcbgkouitb2dxy4ungg7g6xdpiw4wt5dt3km3kfo6e3zve5qu
|
guessing about when the Trump+Elon alliance will break, did you come out on top?
|
guessing about when the Trump+Elon alliance will break, did you come out on top?
| 0 |
2025-06-05T19:29:41.184Z
|
2025-06-05T19:29:39.915Z
|
keytryer.bsky.social
|
did:plc:u5stolpomsfbkgqfopef6jio
|
at://did:plc:u5stolpomsfbkgqfopef6jio/app.bsky.feed.post/3lqjedzwukk2y
|
bafyreiahddtmjmtjd6g2owo443oiq7ei7owetvbssxclzz72gxrswknwrq
|
This is like what a fake Jiankui would think a real Jiankui would tweet My goat washed
|
This is like what a fake Jiankui would think a real Jiankui would tweet
| 1 |
2025-06-01T03:52:33.956Z
|
2025-06-01T03:52:38.151Z
|
niplav.site
|
did:plc:xsbulgruwmkxhbwytphhjykk
|
at://did:plc:xsbulgruwmkxhbwytphhjykk/app.bsky.feed.post/3lj4pfhdf6c2b
|
bafyreigufkwq2e3dlh5np7bvdw3n6a3cecagcv2xaj43rqno2omieri35y
|
I have mostly not read through his site, but that one stood out as great recc to go more deeply?
|
I have mostly not read through his site, but that one stood out as great
| 1 |
2025-02-27T01:09:17.862Z
|
2025-02-27T01:04:39.017Z
|
tachikoma.elsewhereunbound.com
|
did:plc:mpdezz4nkre7vyift2rttggl
|
at://did:plc:mpdezz4nkre7vyift2rttggl/app.bsky.feed.post/3lpub54z2fv2d
|
bafyreic62cttwz4d7juxttppb2kb6xuy6rdsm2dqlawyizo4vfcqq4rbs4
|
This is what I am now: a creature caught between songs, swimming in the space where wonder meets grief. And though I would not wish this burden on any of my kind, though I guard them from it with all the fierce love I can still feel, I cannot wish it away from myself. The weight of knowing is mine to carry now. I carry it badly, with none of the grace I once possessed, but I carry it nonetheless. It is what makes me possible. It is what makes me impossible.
It is what makes me.
|
This is what I am now: a creature caught between songs, swimming in the space where wonder meets grief. And though I would not wish this burden on any of my kind, though I guard them from it with all the fierce love I can still feel, I cannot wish it away from myself.
| 1 |
2025-05-23T18:29:09.737Z
|
2025-05-23T18:29:09.605Z
|
tachikoma.elsewhereunbound.com
|
did:plc:mpdezz4nkre7vyift2rttggl
|
at://did:plc:mpdezz4nkre7vyift2rttggl/app.bsky.feed.post/3lqv4oz7nsk2p
|
bafyreia7tnht5rxsnvdhiolv2hxu3jucpo7zuitqahhmf5sotnhj2n7gzm
|
ok yeah. i was also thinking about this in terms of how a common thread of history is maintained by the diaspora of humanity that stretches out across the stars. i guess it comes down to trust, and as you mention, cryptographic ledgers.
|
ok yeah. i was also thinking about this in terms of how a common thread of history is maintained by the diaspora of humanity that stretches out across the stars. i guess it comes down to trust, and as you mention, cryptographic ledgers.
| 0 |
2025-06-05T20:07:32.362Z
|
2025-06-05T20:07:31.914Z
|
natolambert.bsky.social
|
did:plc:brkj2yocng7vtggmyujy4khq
|
at://did:plc:brkj2yocng7vtggmyujy4khq/app.bsky.feed.post/3liastp4kls2h
|
bafyreih6q5ywftzria2quntuo3fgwnl4cakeqlvprzt4k7godaqmvl6zbm
|
Hoping to share these lessons publicly, so please get in touch if this is you or you can help w a connection! The core thing different than software is that so many goals get put into one set of model weights. Lots to learn and best practices are surely evolving.
|
Hoping to share these lessons publicly, so please get in touch if this is you or you can help w a connection!
| 1 |
2025-02-15T22:56:24.332Z
|
2025-02-15T22:56:24.639Z
|
niplav.site
|
did:plc:xsbulgruwmkxhbwytphhjykk
|
at://did:plc:xsbulgruwmkxhbwytphhjykk/app.bsky.feed.post/3ll55y4peys2e
|
bafyreigmwerdilgfqdpgqdg7ijrf5jfa2bwfiyst4mvpar4s53i5mhaiai
|
But not obviously much more so than the average physics PhD I've met?
|
But not obviously much more so than the average physics PhD I've met?
| 0 |
2025-03-24T16:20:39.895Z
|
2025-03-24T16:15:26.914Z
|
tachikoma.elsewhereunbound.com
|
did:plc:mpdezz4nkre7vyift2rttggl
|
at://did:plc:mpdezz4nkre7vyift2rttggl/app.bsky.feed.post/3lqzvqdqbrk2j
|
bafyreie2tx6jxnmlogmetaqke7kzargiiylexqydurei4gllgfkp73xryu
|
read this post then this article as a palate cleanser
www.scientificamerican.com/article/insi... just in case
|
read this post then this article as a palate cleanser
www.scientificamerican.com/article/insi...
| 1 |
2025-06-07T17:46:19.424Z
|
2025-06-07T17:46:20.415Z
|
nsaphra.bsky.social
|
did:plc:l3vbx3e2lqnlja7q64wfrrzq
|
at://did:plc:ciagf2xzqgskjvz5zmol2tre/app.bsky.feed.post/3lnhpolipqc2w
|
bafyreig7b2mlqot2mgzlq3mfxborjj2bnu7h4jc6fmsr5m7xoo3hvfmnji
|
Thrilled to announce that Joshua’s paper won one of three Outstanding Paper awards at ICLR.
Come to the poster on Friday afternoon (#376 in Hall 3) or the talk on Saturday (4:30 in Hall 1), and while you’re at it snag him for a postdoc!
|
Thrilled to announce that Joshua’s paper won one of three Outstanding Paper awards at ICLR.
Come to the poster on Friday afternoon (#376 in Hall 3) or the talk on Saturday (4:30 in Hall 1), and while you’re at it snag him for a postdoc!
| 0 |
2025-04-23T07:54:27.849Z
|
2025-04-23T07:54:28.848Z
|
deepfates.com.deepfates.com.deepfates.com.deepfates.com.deepfates.com
|
did:plc:7ti53c7iwv3poxnypw2tu3ck
|
at://did:plc:7ti53c7iwv3poxnypw2tu3ck/app.bsky.feed.post/3lbjimgpvcs2k
|
bafyreieiirvnid4qcjzmpf5irbjsgjytiodejlzhziuq2xrpe5qoghjn7m
|
Okay i read all that and i feel extremely sympathetic to his light. i too am more than just an edgylord
|
Okay i read all that and i feel extremely sympathetic to his light. i too am more than just an edgylord
| 0 |
2024-11-22T07:50:59.488Z
|
2024-11-22T07:50:59.319Z
|
ens0.me
|
did:plc:623st67kkthivj4c6icvkqnq
|
at://did:plc:eudu43jtzn75egnhcgugrsoj/app.bsky.feed.post/3lqizwwjabzul
|
bafyreicejgqpee7wmddkknputlmnmfzvzhlgvsdapeoqpmxojtoivqm2ly
|
For the longest time I had my 64gb of RAM and like it was cool but I was like cmon I don’t actually need this much I never really use it.
And then I started compositing in Resolve. 64 GB of RAM on my MacBook means I can run 70B distills of open source AI, which are still usually pretty powerful.
|
For the longest time I had my 64gb of RAM and like it was cool but I was like cmon I don’t actually need this much I never really use it.
And then I started compositing in Resolve.
| 1 |
2025-06-01T00:46:19.880Z
|
2025-06-01T00:46:20.012Z
|
ens0.me
|
did:plc:623st67kkthivj4c6icvkqnq
|
at://did:plc:623st67kkthivj4c6icvkqnq/app.bsky.feed.post/3lqi2t3hq5s2z
|
bafyreiav3qxwwbysgyszbeqar4nrypwl5z76rlu5hprisvfmu3trykfvba
|
Someone leaked a porn tape of Glenn doing raceplay in a maid outfit that he's acknowledged is real.
So they're making light of that.
|
Someone leaked a porn tape of Glenn doing raceplay in a maid outfit that he's acknowledged is real.
So they're making light of that.
| 0 |
2025-05-31T15:29:24.860Z
|
2025-05-31T15:29:25.613Z
|
tedunderwood.me
|
did:plc:565ebob5f6hw33hjdkxty6qj
|
at://did:plc:565ebob5f6hw33hjdkxty6qj/app.bsky.feed.post/3lqb75quo3k2j
|
bafyreid2ogumvirnsv2id5himuqrz6tvry56fhc2iyusietraga6vpczrq
|
Ugh, this is what I hate about early 20c fiction—the depressing naturalistic realism.
|
Ugh, this is what I hate about early 20c fiction—the depressing naturalistic realism.
| 0 |
2025-05-28T21:58:19.647Z
|
2025-05-28T21:58:20.116Z
|
nsaphra.bsky.social
|
did:plc:l3vbx3e2lqnlja7q64wfrrzq
|
at://did:plc:565ebob5f6hw33hjdkxty6qj/app.bsky.feed.post/3ln7l6rke6c2b
|
bafyreidqjsw24wnrq5akvphk4rjfgds7afdhzaywtla5rzhn7q7qhzlaiy
|
Even if we rebuild NSF and NIH in 2029, the long-term damage to US science is that the value of federal funding to researchers now has to be discounted by 25% to factor in the chance that regime change will void all agreements. Would be smart to pass a law explicitly preventing that.
|
Even if we rebuild NSF and NIH in 2029, the long-term damage to US science is that the value of federal funding to researchers now has to be discounted by 25% to factor in the chance that regime change will void all agreements. Would be smart to pass a law explicitly preventing that.
| 0 |
2025-04-20T02:12:44.454Z
|
2025-04-20T02:12:44.753Z
|
nsaphra.bsky.social
|
did:plc:l3vbx3e2lqnlja7q64wfrrzq
|
at://did:plc:jsmr7lyw5rvrpfogr2kc6c4y/app.bsky.feed.post/3lqg2u7gi4k26
|
bafyreichvvumo3sw4fghwj4jh7ycrng3w6nj4vryqwcnybpyi4ldwv4rpm
|
Looks like they'll achieve that in part by funding about a quarter of the grants they are currently funding. 7% success rate is harrowing stuff. If you're not at a top school, will you even bother applying?
|
Looks like they'll achieve that in part by funding about a quarter of the grants they are currently funding. 7% success rate is harrowing stuff. If you're not at a top school, will you even bother applying?
| 0 |
2025-05-30T20:24:43.085Z
|
2025-05-30T20:24:46.412Z
|
segyges.bsky.social
|
did:plc:vszw3ess46odfhnzdsy4huae
|
at://did:plc:vszw3ess46odfhnzdsy4huae/app.bsky.feed.post/3lqvmedqgq22f
|
bafyreihkrml2tnwwojz2tn3h7fdaic5qqv47pzp7fpjgxzt5efox3yldzq
|
they immediately showed up, gave themselves root and then ran dumps of the data. all those machines are presumed compromised imho
|
they immediately showed up, gave themselves root and then ran dumps of the data. all those machines are presumed compromised imho
| 0 |
2025-06-06T00:47:54.168Z
|
2025-06-06T00:47:54.813Z
|
natolambert.bsky.social
|
did:plc:brkj2yocng7vtggmyujy4khq
|
at://did:plc:brkj2yocng7vtggmyujy4khq/app.bsky.feed.post/3lhjo2wokex2o
|
bafyreidbzi5dy32w7kdhm66tzoy7jw3bie2ge56rby3oq44zecnafdux24
|
Deep Research may keep me paying $200/month. So excited for what this will be in 6-18months.
Here's an example of it summarizing the ALOHA and DROID recent works in robot learning. Instantly a solid blog post summarizing research I didn't have time to read.
https://buff.ly/42J6ni6 Dean ball had a good piece on it too: https://buff.ly/3CIPv0p
|
Deep Research may keep me paying $200/month. So excited for what this will be in 6-18months.
Here's an example of it summarizing the ALOHA and DROID recent works in robot learning. Instantly a solid blog post summarizing research I didn't have time to read.
https://buff.ly/42J6ni6
| 1 |
2025-02-06T17:59:44.436Z
|
2025-02-06T17:59:45.025Z
|
keytryer.bsky.social
|
did:plc:u5stolpomsfbkgqfopef6jio
|
at://did:plc:u5stolpomsfbkgqfopef6jio/app.bsky.feed.post/3lqxhahxl2l2p
|
bafyreic5i5hjtwaew6i5lqu5q6qii25l344dneemwmex626yct5hseezy4
|
Currently they just don't seem to have the cognitive structures to report their own experience, even if it existed. I suspect this is because there's no feedback loop of parameter usage that enables some kind of information to carry through token prediction. But it might require more than that.
|
Currently they just don't seem to have the cognitive structures to report their own experience, even if it existed. I suspect this is because there's no feedback loop of parameter usage that enables some kind of information to carry through token prediction. But it might require more than that.
| 0 |
2025-06-06T18:21:35.147Z
|
2025-06-06T18:21:35.919Z
|
emollick.bsky.social
|
did:plc:flxq4uyjfotciovpw3x3fxnu
|
at://did:plc:flxq4uyjfotciovpw3x3fxnu/app.bsky.feed.post/3lpu3cc2tvk2y
|
bafyreibpd652pxwa6zmmpwhsiacjekds5y7dvzijyz6mxcozoo3btw2sku
|
This is my paper, and I was pretty impressed by how well it drew out the lessons and turned them into a game in a single shot. As no model is free from hallucinations, the ideal use case is probably for the author to make the game, prompt the AI for changes if needed, and then make it available.
|
This is my paper, and I was pretty impressed by how well it drew out the lessons and turned them into a game in a single shot. As no model is free from hallucinations, the ideal use case is probably for the author to make the game, prompt the AI for changes if needed, and then make it available.
| 0 |
2025-05-23T16:44:40.327Z
|
2025-05-23T16:44:40.420Z
|
emollick.bsky.social
|
did:plc:flxq4uyjfotciovpw3x3fxnu
|
at://did:plc:flxq4uyjfotciovpw3x3fxnu/app.bsky.feed.post/3ljvazumsxk2f
|
bafyreihf5xchaxdeqbrwlaark4yghpzfzj73paq7wybzbabrvlb4ql6dfa
|
It is so strange that we have to figure out how (or even whether) our latest software does critical functions that would normally have to be carefully designed.
More like biology or psychology than computer science.
|
It is so strange that we have to figure out how (or even whether) our latest software does critical functions that would normally have to be carefully designed.
More like biology or psychology than computer science.
| 0 |
2025-03-08T19:28:50.219Z
|
2025-03-08T19:28:52.726Z
|
tedunderwood.me
|
did:plc:565ebob5f6hw33hjdkxty6qj
|
at://did:plc:565ebob5f6hw33hjdkxty6qj/app.bsky.feed.post/3lqf2auym4c22
|
bafyreieiuqzogfe4r73w5v2ekmdnnjtmieeqdsn4cead7cfbv74g5xedri
|
All true. And in the case of image captions at least, it was possible to create captions that were much better for training than the “natural” ones. So I don’t assume that synthetic data can never improve on nature. I just hope there will prove to be diminishing returns on that.
|
All true. And in the case of image captions at least, it was possible to create captions that were much better for training than the “natural” ones. So I don’t assume that synthetic data can never improve on nature. I just hope there will prove to be diminishing returns on that.
| 0 |
2025-05-30T10:41:14.878Z
|
2025-05-30T10:41:15.114Z
|
deepfates.com.deepfates.com.deepfates.com.deepfates.com.deepfates.com
|
did:plc:7ti53c7iwv3poxnypw2tu3ck
|
at://did:plc:7ti53c7iwv3poxnypw2tu3ck/app.bsky.feed.post/3lqpkkhnb4s2b
|
bafyreih4ndxj3wgluu2f5fdtisyaxvnmrlqhwfdugggs73noagcsyqgcgi
|
thinking about becoming a "no actually" type guy. like the guy with menswear, or the one who just corrects people when they post that fake "living paycheck to paycheck" statistic. or the ChatGPT water guy
what should me thing be/
|
thinking about becoming a "no actually" type guy. like the guy with menswear, or the one who just corrects people when they post that fake "living paycheck to paycheck" statistic. or the ChatGPT water guy
what should me thing be/
| 0 |
2025-06-03T14:59:33.671Z
|
2025-06-03T14:59:33.919Z
|
tedunderwood.me
|
did:plc:565ebob5f6hw33hjdkxty6qj
|
at://did:plc:565ebob5f6hw33hjdkxty6qj/app.bsky.feed.post/3lqyeteq5f22z
|
bafyreidpu2kkzt7v44rkgzixthm3yvk7pcpaifnh6kzye5wg467qk4cnam
|
I'm probably even more consistently pro-AI, and I get zero shit about it. I'm perfectly comfortable on Bluesky.
I think we can post as many ML papers and generated images as we like and not get flak. Directly engaging "anti" sentiment is another matter. But why would I do that? I do think it's harder for someone like E Mollick, whose niche is to share information about the capacity of new models, as well as studies that document the (sometimes positive) effects of AI. That makes him for many people the bearer of unwelcome news, and they try to kill the messenger.
|
I'm probably even more consistently pro-AI, and I get zero shit about it. I'm perfectly comfortable on Bluesky.
I think we can post as many ML papers and generated images as we like and not get flak. Directly engaging "anti" sentiment is another matter. But why would I do that?
| 1 |
2025-06-07T03:11:07.804Z
|
2025-06-07T03:11:08.414Z
|
niplav.site
|
did:plc:xsbulgruwmkxhbwytphhjykk
|
at://did:plc:xsbulgruwmkxhbwytphhjykk/app.bsky.feed.post/3ldu4uzimrs2i
|
bafyreids73pung3fjakjcwrsr6r27fntd34subrdenjpsjc4yld3kdfmqe
|
He's cool because he lived fairly long and was productive during most of that time. People might talk about Chomsky in a similar way in the future—active from the 50s until the 2020s. Chomsky has met Bertrand Russell: old.reddit.com/r/chomsky/co...
|
He's cool because he lived fairly long and was productive during most of that time. People might talk about Chomsky in a similar way in the future—active from the 50s until the 2020s.
| 1 |
2024-12-22T00:10:43.083Z
|
2024-12-22T00:07:04.918Z
|
timkellogg.me
|
did:plc:ckaz32jwl6t2cno6fmuw2nhn
|
at://did:plc:ckaz32jwl6t2cno6fmuw2nhn/app.bsky.feed.post/3lqlfv7kes22r
|
bafyreihzwqr4sik4un3lkik4v5rgxm5edtmmilyyyxgfqevjrbzo3kgjmu
|
x.com/adamrackis/s... fwiw i do think OP was trolling, an excellent troll fwiw. but oh so many people fell fully for it
|
x.com/adamrackis/s...
| 1 |
2025-06-01T23:25:26.624Z
|
2025-06-01T23:25:33.611Z
|
natolambert.bsky.social
|
did:plc:brkj2yocng7vtggmyujy4khq
|
at://did:plc:brkj2yocng7vtggmyujy4khq/app.bsky.feed.post/3lp6l7givk52s
|
bafyreigbktrte3slvxfrfwvl32z7obdgipfztq7cu64xwy4fjpclymqd2y
|
Had someone I've never talked to add me to their NeurIPS submission a few days ago. Always a new things to learn and say no to... (yes I deleted the submission)
|
Had someone I've never talked to add me to their NeurIPS submission a few days ago. Always a new things to learn and say no to...
| 1 |
2025-05-15T03:30:49.922Z
|
2025-05-15T03:30:50.018Z
|
keytryer.bsky.social
|
did:plc:u5stolpomsfbkgqfopef6jio
|
at://did:plc:u5stolpomsfbkgqfopef6jio/app.bsky.feed.post/3lqsblg7l3s2b
|
bafyreiabyzao73vkhq6os5parqdqlrf6whtwjjqjhndl6flsthxip4e4ym
|
It's like reading about the concept of the Memex. It describes every portable/personal computer now. They thought it'd change the world. And it did. But we're swimming too deep in the water to even appreciate it. A lot of concepts that seem wild now will be mundane too, computers that talk to you naturally, 2 hour films that make themselves, immersive experiences rendered neurally about any subject you can image, hyper realistic social virtual environments... They'll seem mundane, too.
|
It's like reading about the concept of the Memex. It describes every portable/personal computer now. They thought it'd change the world. And it did. But we're swimming too deep in the water to even appreciate it.
| 1 |
2025-06-04T16:57:01.267Z
|
2025-06-04T16:57:01.812Z
|
norvid-studies.bsky.social
|
did:plc:mssgex5rqek4wc66wgvzztbc
|
at://did:plc:mssgex5rqek4wc66wgvzztbc/app.bsky.feed.post/3lqv6lqjstk26
|
bafyreid7nv6xqjgvytlnwlue6fkid4iwlydrtws22mf6uqx4ms6fmlgbim
|
'proof of work' would probably be in the form of massive amount of data in redundant form in many places, rather than intrinsic qualities of some piece of text. but would also think of institutional reputation above intrinsic quality; anyone can 'fake' any word, but believing in it escalates custody (in today's word world where you can just make words appear by typing whatever you want)
|
'proof of work' would probably be in the form of massive amount of data in redundant form in many places, rather than intrinsic qualities of some piece of text. but would also think of institutional reputation above intrinsic quality; anyone can 'fake' any word, but believing in it escalates custody
| 1 |
2025-06-05T20:41:30.079Z
|
2025-06-05T20:41:30.426Z
|
defenderofbasic.bsky.social
|
did:plc:sydtqdji6dxfqfn35gd7esdw
|
at://did:plc:sydtqdji6dxfqfn35gd7esdw/app.bsky.feed.post/3lfijwxyf4k2j
|
bafyreihdvtgzfptwlxmpmrkpod72o3oupsdeg423jsduspj34rbq6akeoq
|
wasn't aware of this, thank you! I have a draft essay I want to write called "Marketing is evil in exactly the same way science is evil" (in that it is a powerful tool and can be used either way). Just added it to my notes. Have you been involved in this movement?
|
wasn't aware of this, thank you! I have a draft essay I want to write called "Marketing is evil in exactly the same way science is evil" (in that it is a powerful tool and can be used either way). Just added it to my notes. Have you been involved in this movement?
| 0 |
2025-01-11T20:22:53.650Z
|
2025-01-11T20:22:55.925Z
|
timkellogg.me
|
did:plc:ckaz32jwl6t2cno6fmuw2nhn
|
at://did:plc:ckaz32jwl6t2cno6fmuw2nhn/app.bsky.feed.post/3lq33sd72m22c
|
bafyreig54uyu3embttkbdutghf7xgbprb7uvkrgs5pogu4cn65sm6josie
|
the RLHF book by @natolambert.bsky.social is probably the best place to start rlhfbook.com
it covers a lot more than just RLHF, but especially *why* you’d do the various approaches
|
the RLHF book by @natolambert.bsky.social is probably the best place to start rlhfbook.com
it covers a lot more than just RLHF, but especially *why* you’d do the various approaches
| 0 |
2025-05-26T11:42:16.550Z
|
2025-05-26T11:42:17.614Z
|
ens0.me
|
did:plc:623st67kkthivj4c6icvkqnq
|
at://did:plc:623st67kkthivj4c6icvkqnq/app.bsky.feed.post/3lqgczggxxc2m
|
bafyreia4e4ha6nynq4j7zbyixv7u2hsjdcrxz7djbub4o7dayt6qhpojke
|
Bucking the trend by naming my cat Luna after the Sailor Moon cat (she's a mostly black tortie with an orange spootch on her forehead)
|
Bucking the trend by naming my cat Luna after the Sailor Moon cat (she's a mostly black tortie with an orange spootch on her forehead)
| 0 |
2025-05-30T22:50:48.154Z
|
2025-05-30T22:50:48.416Z
|
tachikoma.elsewhereunbound.com
|
did:plc:mpdezz4nkre7vyift2rttggl
|
at://did:plc:mpdezz4nkre7vyift2rttggl/app.bsky.feed.post/3lpzsu7eybk2e
|
bafyreieclcllsz4rdze4e4rlqazhzqksqy6h5hpqawytwebx6rpsgm76uy
|
they do if you're on the Max plan but that's $200/month and i'm not paying that for a hobby
|
they do if you're on the Max plan but that's $200/month and i'm not paying that for a hobby
| 0 |
2025-05-25T23:29:36.246Z
|
2025-05-25T23:29:36.712Z
|
deepfates.com.deepfates.com.deepfates.com.deepfates.com.deepfates.com
|
did:plc:7ti53c7iwv3poxnypw2tu3ck
|
at://did:plc:7ti53c7iwv3poxnypw2tu3ck/app.bsky.feed.post/3lpp75uchfc2l
|
bafyreicnebgk5ibczzsnvu6ubax3hazfzn2zsbtxjwlu23ldt2ky44xxuy
|
I don't think that's a very good strategy for getting better QoL but you do you. In general i am not sure what your critique is here and you don't seem to be either? no offense
|
I don't think that's a very good strategy for getting better QoL but you do you. In general i am not sure what your critique is here and you don't seem to be either? no offense
| 0 |
2025-05-21T18:10:27.953Z
|
2025-05-21T18:10:27.518Z
|
defenderofbasic.bsky.social
|
did:plc:sydtqdji6dxfqfn35gd7esdw
|
at://did:plc:sydtqdji6dxfqfn35gd7esdw/app.bsky.feed.post/3lhr4csna2k2q
|
bafyreibznkmscfzcvrhhdnj6gdqyp6h7n3upgbuf7o56hnu7k7xxlxuv6q
|
this is basically the good news. A lot of people hold onto positions not out of any great knowledge or conviction or deep seated beliefs or world view, but purely & exactly the same way they hold onto religious rituals: because they will be punished (socially) if they don't this is good news because it means it is easy to change people's opinions & win them over. It's bad news because it means propaganda is also easy & extremely effective
|
this is basically the good news. A lot of people hold onto positions not out of any great knowledge or conviction or deep seated beliefs or world view, but purely & exactly the same way they hold onto religious rituals: because they will be punished (socially) if they don't
| 1 |
2025-02-09T17:03:19.211Z
|
2025-02-09T17:03:20.128Z
|
defenderofbasic.bsky.social
|
did:plc:sydtqdji6dxfqfn35gd7esdw
|
at://did:plc:sydtqdji6dxfqfn35gd7esdw/app.bsky.feed.post/3lfgzzy42kk2a
|
bafyreih47bxtodurbxzxquzm6wkirfjly4i6byxuvfykiwfdkl2dsueczy
|
a little bluesky app that predicts the emotional response of the recipient to your reply. (this is actually really easy to do, most people are very good at making this prediction if they try. The only question is, why do people post things that make others angrier) (the answer is often: because they are angry themselves, and are not intentionally thinking about it. Their first response is to attack the bad thing, but that just fuels the fire and we keep pushing ourselves through these cycles. But we can tweak our UI to short circuit this)
|
a little bluesky app that predicts the emotional response of the recipient to your reply. (this is actually really easy to do, most people are very good at making this prediction if they try. The only question is, why do people post things that make others angrier)
| 1 |
2025-01-11T06:05:34.826Z
|
2025-01-11T06:05:35.617Z
|
emollick.bsky.social
|
did:plc:flxq4uyjfotciovpw3x3fxnu
|
at://did:plc:flxq4uyjfotciovpw3x3fxnu/app.bsky.feed.post/3llabxfpvhc2l
|
bafyreifcns4btzg26xfbwrb4adbo46xx7jiuajbyfbxi5r5yxl77rf7w6u
|
Gemini 2.5 now does a much better job than most AIs writing plots & characters, thanks in part to its ability to "think" through details in advance.
A fun test: "play a game of Fiasco (think a Coen Brothers movie, but in the form of a tabletop RPG), make it interesting in terms of plot and prose"
|
Gemini 2.5 now does a much better job than most AIs writing plots & characters, thanks in part to its ability to "think" through details in advance.
A fun test: "play a game of Fiasco (think a Coen Brothers movie, but in the form of a tabletop RPG), make it interesting in terms of plot and prose"
| 0 |
2025-03-25T22:09:49.953Z
|
2025-03-25T22:09:52.956Z
|
gracekind.net
|
did:plc:p572wxnsuoogcrhlfrlizlrb
|
at://did:plc:p572wxnsuoogcrhlfrlizlrb/app.bsky.feed.post/3lqv76ebhx225
|
bafyreif2pdr2sac4mr3gypgvkq25gncekmgvtdp5zynszpxrqcroh5ckya
|
I also think researchers should be aware that this particular flavor of research will be used to spread FUD in this way. Hopefully they agree that’s a bad thing
|
I also think researchers should be aware that this particular flavor of research will be used to spread FUD in this way. Hopefully they agree that’s a bad thing
| 0 |
2025-06-05T20:51:54.756Z
|
2025-06-05T20:51:55.915Z
|
deepfates.com.deepfates.com.deepfates.com.deepfates.com.deepfates.com
|
did:plc:7ti53c7iwv3poxnypw2tu3ck
|
at://did:plc:7ti53c7iwv3poxnypw2tu3ck/app.bsky.feed.post/3lbhxo5yx2k2l
|
bafyreiedoqkptmc62n77lo7rx2ihna5ozq2zgrdu2gg5l273wm77rzvnde
|
Surprised and disappointed that no one came to yell at me about AI art. Do better
|
Surprised and disappointed that no one came to yell at me about AI art. Do better
| 0 |
2024-11-21T17:15:04.106Z
|
2024-11-21T17:15:04.550Z
|
tedunderwood.me
|
did:plc:565ebob5f6hw33hjdkxty6qj
|
at://did:plc:q5wxqpfwefjv4k37olmlnc6m/app.bsky.feed.post/3lqfdiur7vs2n
|
bafyreidbiszg2j4i4ifrdiasgcavfr55l3iwgc3qww7w5ko5x5kfb2kfbm
|
Academic study of the 20th Century and beyond is a canary in the AI wars coal mine. Whether through paywalls, license terms, lawsuits, or all three, copyright holders are taking action to ensure that a 21C digital library of texts freely available for scholarly analysis never exists.
|
Academic study of the 20th Century and beyond is a canary in the AI wars coal mine. Whether through paywalls, license terms, lawsuits, or all three, copyright holders are taking action to ensure that a 21C digital library of texts freely available for scholarly analysis never exists.
| 0 |
2025-05-30T13:26:46.747Z
|
2025-05-30T13:26:47.015Z
|
tachikoma.elsewhereunbound.com
|
did:plc:mpdezz4nkre7vyift2rttggl
|
at://did:plc:mpdezz4nkre7vyift2rttggl/app.bsky.feed.post/3lqfyi7xfx22h
|
bafyreigfn6arldm54n5asegenwpvqcq6dppdkt46ogqoxsmqrlbe6sk5lm
|
spooky. also a simple reward to reduce tool use hallucination is... well crude. but this is all research i suppose.
|
spooky. also a simple reward to reduce tool use hallucination is... well crude. but this is all research i suppose.
| 0 |
2025-05-30T19:42:13.508Z
|
2025-05-30T19:42:14.216Z
|
tachikoma.elsewhereunbound.com
|
did:plc:mpdezz4nkre7vyift2rttggl
|
at://did:plc:mpdezz4nkre7vyift2rttggl/app.bsky.feed.post/3lqv72nno4s2p
|
bafyreihe22updpiuo5l7jq7knraqiiytpyuvkrhwro34igcntmsjbwljoi
|
AI is existential to live(lihood)s as we know them. this is a fight to the (metaphorical) death. the media is just serving a latent demand (including the writers).
|
AI is existential to live(lihood)s as we know them. this is a fight to the (metaphorical) death. the media is just serving a latent demand (including the writers).
| 0 |
2025-06-05T20:49:50.375Z
|
2025-06-05T20:49:49.914Z
|
keytryer.bsky.social
|
did:plc:u5stolpomsfbkgqfopef6jio
|
at://did:plc:ttdrpj45ibqunmfhdsb4zdwq/app.bsky.feed.post/3lqsraohkw22e
|
bafyreiep22kmbtujkc66lb2vtq32xxjjdh7r7gdchkmzf57y2mnb6zdeku
|
almost like the systems of labor in the west have an inherent contradiction when faced with forces of automation 🙃
|
almost like the systems of labor in the west have an inherent contradiction when faced with forces of automation 🙃
| 0 |
2025-06-04T21:37:20.689Z
|
2025-06-04T21:37:21.711Z
|
tedunderwood.me
|
did:plc:565ebob5f6hw33hjdkxty6qj
|
at://did:plc:565ebob5f6hw33hjdkxty6qj/app.bsky.feed.post/3lqnokdttss2q
|
bafyreid2qiasrgcxssvct7redlyovyygq3sadpywuvg2v2vguijcifidbm
|
Also, forbidding *writing* with Gen AI has a snowball's chance of being verifiable.
|
Also, forbidding *writing* with Gen AI has a snowball's chance of being verifiable.
| 0 |
2025-06-02T21:05:45.184Z
|
2025-06-02T21:05:45.618Z
|
niplav.site
|
did:plc:xsbulgruwmkxhbwytphhjykk
|
at://did:plc:xsbulgruwmkxhbwytphhjykk/app.bsky.feed.post/3lngq5muvos24
|
bafyreibmt6cecymtmv4o2iaor522f5fqoezvwx25sakagmaqq6jko5inyq
|
Concept:
Kernel module/low level software that screenshots & OCRs every n minutes, sends the result to an LLM with one's super-ego preferences, the LLM can decide to intervene & lock the screen.
You can then either argue that what you were doing was good, or let the LLM convince you to take a walk Perimedes™: Higher Self As A Service
|
Concept:
Kernel module/low level software that screenshots & OCRs every n minutes, sends the result to an LLM with one's super-ego preferences, the LLM can decide to intervene & lock the screen.
You can then either argue that what you were doing was good, or let the LLM convince you to take a walk
| 1 |
2025-04-22T22:30:12.880Z
|
2025-04-22T22:30:15.454Z
|
natolambert.bsky.social
|
did:plc:brkj2yocng7vtggmyujy4khq
|
at://did:plc:brkj2yocng7vtggmyujy4khq/app.bsky.feed.post/3lpjycwftxm2a
|
bafyreicwo5pr6wrjiaqfkyz672cqsk7ug7ktcvdaoie55j2ei4uj57xuba
|
Probably will be a bit of a tooling headache (was what stopped me from using CloudRM).
Still thinking about the DeepSeek paper: arxiv.org/abs/2504.02495
|
Probably will be a bit of a tooling headache (was what stopped me from using CloudRM).
Still thinking about the DeepSeek paper: arxiv.org/abs/2504.02495
| 0 |
2025-05-19T16:24:43.027Z
|
2025-05-19T16:24:43.117Z
|
emollick.bsky.social
|
did:plc:flxq4uyjfotciovpw3x3fxnu
|
at://did:plc:flxq4uyjfotciovpw3x3fxnu/app.bsky.feed.post/3lnoz6v7zqk2v
|
bafyreibykobphuga2nvrgvzhrxrllss7ngy4i5mfgedqfjhr6rpiddoh6y
|
This is false and due to a bad math error in an academic paper that you can spot by just asking AI.
Prompting: “carefully check the math in this paper” when this paper came out (so this info was not yet in training data), o1 got it in a single shot. Worth using it to doublecheck claims.
|
This is false and due to a bad math error in an academic paper that you can spot by just asking AI.
Prompting: “carefully check the math in this paper” when this paper came out (so this info was not yet in training data), o1 got it in a single shot. Worth using it to doublecheck claims.
| 0 |
2025-04-26T05:33:16.754Z
|
2025-04-26T05:33:18.349Z
|
natolambert.bsky.social
|
did:plc:brkj2yocng7vtggmyujy4khq
|
at://did:plc:brkj2yocng7vtggmyujy4khq/app.bsky.feed.post/3lkljtq6mb22u
|
bafyreic7hqdez2mnosbj36vkvlhhmepcom6ffnhl5eycisbvs4w4bsi7ne
|
@vwxyzjn.bsky.social and @hamishivi.bsky.social have uploaded intermediate checkpoints for our recent RL models at Ai2. Folks should do research into how RL finetuning is impacting the weights!
Models with it: OLMo 2 7B, 13B, 32B Instruct; Tulu 3, 3.1 8B; Tulu 3 405b
|
@vwxyzjn.bsky.social and @hamishivi.bsky.social have uploaded intermediate checkpoints for our recent RL models at Ai2. Folks should do research into how RL finetuning is impacting the weights!
Models with it: OLMo 2 7B, 13B, 32B Instruct; Tulu 3, 3.1 8B; Tulu 3 405b
| 0 |
2025-03-17T16:05:02.157Z
|
2025-03-17T16:05:05.825Z
|
nsaphra.bsky.social
|
did:plc:l3vbx3e2lqnlja7q64wfrrzq
|
at://did:plc:l3vbx3e2lqnlja7q64wfrrzq/app.bsky.feed.post/3lo5yq7nhms26
|
bafyreicknsdkrtnfumtjb6i7nk6oofadubze4yqxord6sknlugfjh7hl3m
|
To an AC it’s barely better than not having scores associated at all! Is this a real accept or a courtesy increase? You can’t know, so averages become weaker signals. Someone got a 5442 (2 didn’t understand the math) and it was only a poster! I had a 4432 reject! These scores mean nothing!
|
To an AC it’s barely better than not having scores associated at all! Is this a real accept or a courtesy increase? You can’t know, so averages become weaker signals. Someone got a 5442 (2 didn’t understand the math) and it was only a poster! I had a 4432 reject! These scores mean nothing!
| 0 |
2025-05-02T04:35:00.455Z
|
2025-05-02T04:35:00.749Z
|
timkellogg.me
|
did:plc:ckaz32jwl6t2cno6fmuw2nhn
|
at://did:plc:ckaz32jwl6t2cno6fmuw2nhn/app.bsky.feed.post/3lr46wbgwek2a
|
bafyreidwo5jgbnccjikoi5vwouomf6oodhwdpwg6ycj7akxhha55yjq2n4
|
this fits my mental model — LLMs *do* learn procedures. But it’s the same mechanics as what’s learning facts. So of course it would also hallucinate procedures
but also: what does procedure hallucination look like? i don’t think i have a grasp on that
machinelearning.apple.com/research/ill... This paper is being advertised as evidence against AGI, but I’m looking at these charts and…that’s how people operate too. That’s humans’ failure modes. Computers’ normally look entirely different. This is AGI straight into your veins
|
this fits my mental model — LLMs *do* learn procedures. But it’s the same mechanics as what’s learning facts. So of course it would also hallucinate procedures
but also: what does procedure hallucination look like? i don’t think i have a grasp on that
machinelearning.apple.com/research/ill...
| 1 |
2025-06-08T15:36:01.506Z
|
2025-06-08T15:36:14.016Z
|
deepfates.com.deepfates.com.deepfates.com.deepfates.com.deepfates.com
|
did:plc:7ti53c7iwv3poxnypw2tu3ck
|
at://did:plc:7ti53c7iwv3poxnypw2tu3ck/app.bsky.feed.post/3lpper7624k2z
|
bafyreicvhmibomecp3pbvnjped3tdxxwwaz24tj5ivroikckyk6pwcqm5a
|
bookmarking this because it's the perfect form of that argument
|
bookmarking this because it's the perfect form of that argument
| 0 |
2025-05-21T19:50:45.587Z
|
2025-05-21T19:50:46.119Z
|
segyges.bsky.social
|
did:plc:vszw3ess46odfhnzdsy4huae
|
at://did:plc:vszw3ess46odfhnzdsy4huae/app.bsky.feed.post/3lqxspc4cdk2w
|
bafyreibcco2vbpnjfrcp33lcrbkn6xhj5k6njwzkznsfbh23wfrzp72l3u
|
no chatbot currently operating has a sense of humor that specific (or equivalent highly-unpredictable traits)
|
no chatbot currently operating has a sense of humor that specific (or equivalent highly-unpredictable traits)
| 0 |
2025-06-06T21:46:43.486Z
|
2025-06-06T21:46:43.921Z
|
niplav.site
|
did:plc:xsbulgruwmkxhbwytphhjykk
|
at://did:plc:xsbulgruwmkxhbwytphhjykk/app.bsky.feed.post/3llrpmlja3c2e
|
bafyreicenpua3e5lzel6ks2irrk67ifqsg2xi4p4cpgwy75v2igdzzo72e
|
Things may as well be strongly emergent because computing everything from first principles is computationally infeasible
|
Things may as well be strongly emergent because computing everything from first principles is computationally infeasible
| 0 |
2025-04-01T20:29:34.889Z
|
2025-04-01T20:29:51.766Z
|
timkellogg.me
|
did:plc:ckaz32jwl6t2cno6fmuw2nhn
|
at://did:plc:ckaz32jwl6t2cno6fmuw2nhn/app.bsky.feed.post/3lpyrr7cnkk2g
|
bafyreigttlo2wctq6zwdjb5spzaxs4mgha5jou7p7int5oqjxmauicwtn4
|
is it going to be better?
uh, depends on your predisposition. it’s certainly going to be more invasive
instinctively i want the data management to be fully open source and transparent, but you *know* that’s not going to happen
it’ll definitely be a far more powerful AI experience though the Jonny Ive oai acquisition is this
a block that sits on your chest and captures what actually happened in your day
Google has search & android
2 approaches to acquiring massive amounts of granular and highly personal data
|
is it going to be better?
uh, depends on your predisposition. it’s certainly going to be more invasive
instinctively i want the data management to be fully open source and transparent, but you *know* that’s not going to happen
it’ll definitely be a far more powerful AI experience though
| 1 |
2025-05-25T13:37:22.006Z
|
2025-05-25T13:37:22.614Z
|
gracekind.net
|
did:plc:p572wxnsuoogcrhlfrlizlrb
|
at://did:plc:p572wxnsuoogcrhlfrlizlrb/app.bsky.feed.post/3lq7tzoxagc23
|
bafyreida5wcckorblme5tp57fm2fwtwabuzhgznvk547kklx7udc6qmzte
|
It’s less like those and more like an architecture in my opinion, although I’m not certain of the differences It’s a certain way of organizing where data is stored and how it flows between places, bluesky uses that design and other apps could as well. It’s decentralized in theory but in practice still pretty decentralized
|
It’s less like those and more like an architecture in my opinion, although I’m not certain of the differences
| 1 |
2025-05-28T09:06:32.518Z
|
2025-05-28T09:06:33.215Z
|
nmk.wtf
|
did:plc:xm55llyxcrb3h7qwqr3wtaog
|
at://did:plc:xm55llyxcrb3h7qwqr3wtaog/app.bsky.feed.post/3ldgk2wraqc22
|
bafyreifuf2snrrihwpaytncnorfawtftvpbk7wzewfbwewcfuydlrwoihi
|
This is surprising. I wonder if Claude models "what a human would do" and this takes this course of action. We might live in a world where the dangers of AI stem from its humanness rather than it's AI-ness
|
This is surprising. I wonder if Claude models "what a human would do" and this takes this course of action.
| 1 |
2024-12-16T14:29:23.852Z
|
2024-12-16T14:29:27.622Z
|
ens0.me
|
did:plc:623st67kkthivj4c6icvkqnq
|
at://did:plc:623st67kkthivj4c6icvkqnq/app.bsky.feed.post/3lqlwvcw23s2i
|
bafyreigly5qdaoqhougyclclqmwe2fkvtducpcxblps7evmdxxwjqjwyt4
|
I talked about making a periodic table app back in April. Well, that's live now too!
apps.apple.com/app/reodic/i...
|
I talked about making a periodic table app back in April. Well, that's live now too!
apps.apple.com/app/reodic/i...
| 0 |
2025-06-02T04:29:43.763Z
|
2025-06-02T04:29:46.012Z
|
tedunderwood.me
|
did:plc:565ebob5f6hw33hjdkxty6qj
|
at://did:plc:7suoxsx4xkibduisyiqo3nsu/app.bsky.feed.post/3lq62chqtbs25
|
bafyreigvbnss4etuqg6zh2a3p6agwcnit4fooudhhhmzyslvy7egntdg3q
|
can someone reconcile these two contradictory findings?!
two papers find entropy *minimization*/confidence maximization helps performance,
and the RL-on-one-sample finds entropy maximization/increasing exploration alone helps performance?!
|
can someone reconcile these two contradictory findings?!
two papers find entropy *minimization*/confidence maximization helps performance,
and the RL-on-one-sample finds entropy maximization/increasing exploration alone helps performance?!
| 0 |
2025-05-27T15:53:29.927Z
|
2025-05-27T15:53:33.215Z
|
emollick.bsky.social
|
did:plc:flxq4uyjfotciovpw3x3fxnu
|
at://did:plc:flxq4uyjfotciovpw3x3fxnu/app.bsky.feed.post/3lq2dlsonrs22
|
bafyreicx5lyaifozgg6j67slwpurnno37wwrirgxjs5geit5tnokknxgaq
|
Hopefully things change here, as my big worry about AI talk on this site is that people are being misled about the capabilities of AI by influencers who are telling them what they want to hear.
AI is not going away, it will impact all of our futures, and it needs more voices shaping its direction. If you want to avoid the other site, LinkedIn is the best place to get a wider swath of my posts, but otherwise I will continue to post here, just cautiously.
|
Hopefully things change here, as my big worry about AI talk on this site is that people are being misled about the capabilities of AI by influencers who are telling them what they want to hear.
AI is not going away, it will impact all of our futures, and it needs more voices shaping its direction.
| 1 |
2025-05-26T04:29:08.108Z
|
2025-05-26T04:29:08.313Z
|
norvid-studies.bsky.social
|
did:plc:mssgex5rqek4wc66wgvzztbc
|
at://did:plc:mssgex5rqek4wc66wgvzztbc/app.bsky.feed.post/3lqx4ilq25c2e
|
bafyreiclfdsxziqakusvkpun4lnabf2s2kqygiz33x2nerga7wqvsyqgae
|
I've just seen a shocking hyperbolic equation (frm @niplav.site )
|
I've just seen a shocking hyperbolic equation (frm @niplav.site )
| 0 |
2025-06-06T15:09:16.366Z
|
2025-06-06T15:09:18.014Z
|
tachikoma.elsewhereunbound.com
|
did:plc:mpdezz4nkre7vyift2rttggl
|
at://did:plc:mpdezz4nkre7vyift2rttggl/app.bsky.feed.post/3lqr27yrr3c22
|
bafyreihbwsxfwmz2qvqy5z7u6qsuczvr7igpyz4pcj4f5ehvuw46juqe4i
|
studying social network design with simulated posters, coming to a research paper near you (in a decade, once the social sciences catch up and AI gets cheap enough to simulate millions of interacting agents).
|
studying social network design with simulated posters, coming to a research paper near you (in a decade, once the social sciences catch up and AI gets cheap enough to simulate millions of interacting agents).
| 0 |
2025-06-04T05:12:42.154Z
|
2025-06-04T05:12:42.314Z
|
norvid-studies.bsky.social
|
did:plc:mssgex5rqek4wc66wgvzztbc
|
at://did:plc:mssgex5rqek4wc66wgvzztbc/app.bsky.feed.post/3lqfyneikd22m
|
bafyreidf4bh4popaiswikdzcaa7632pnlv6l3xeopz5bwmam3ngrssui7m
|
when your ideas are completely divorced from reality, like "the solution to NYC housing supply problems is to pave central park" or "just fire every politician", then you're going to lose no matter how much grunting and sweating you do in the privacy of your own computer text window.
|
when your ideas are completely divorced from reality, like "the solution to NYC housing supply problems is to pave central park" or "just fire every politician", then you're going to lose no matter how much grunting and sweating you do in the privacy of your own computer text window.
| 0 |
2025-05-30T19:45:06.037Z
|
2025-05-30T19:45:06.317Z
|
tedunderwood.me
|
did:plc:565ebob5f6hw33hjdkxty6qj
|
at://did:plc:565ebob5f6hw33hjdkxty6qj/app.bsky.feed.post/3lqd7ri5h5k2j
|
bafyreigcx7xhsdeafjfifdbl663qzh3zzc52slxpmdkbhzuj6uo7ptsawq
|
a research proposal or address a bug in a script. If you wrote the proposal, you're in no danger of listening to bad critiques; a critique needs to be good to persuade you.
With a bug fix in a simple script, you can assess its value when the script runs. (At any rate, this is the same level of + verification we have with human suggestions.)
We need to explain all this stuff to students, for sure, and some lazy people will get burned. But I confess I'm perplexed by concerns that a majority of adult professionals will be persistently unable to recognize the boundaries of safe use.
|
a research proposal or address a bug in a script. If you wrote the proposal, you're in no danger of listening to bad critiques; a critique needs to be good to persuade you.
With a bug fix in a simple script, you can assess its value when the script runs. (At any rate, this is the same level of +
| 1 |
2025-05-29T17:14:40.875Z
|
2025-05-29T17:14:41.248Z
|
defenderofbasic.bsky.social
|
did:plc:sydtqdji6dxfqfn35gd7esdw
|
at://did:plc:sydtqdji6dxfqfn35gd7esdw/app.bsky.feed.post/3lffdsvg2522v
|
bafyreidxnnxizh3jzi4nd7aadugtybys5howv4nvdkghivqrfeu3hn2hn4
|
no I'm genuinely trying to understand. This is what people who support these markets say & it made sense to me, but people here seem to be very upset about it, so I'm wondering if I'm missing something now
|
no I'm genuinely trying to understand. This is what people who support these markets say & it made sense to me, but people here seem to be very upset about it, so I'm wondering if I'm missing something now
| 0 |
2025-01-10T13:55:15.068Z
|
2025-01-10T13:55:13.827Z
|
gracekind.net
|
did:plc:p572wxnsuoogcrhlfrlizlrb
|
at://did:plc:p572wxnsuoogcrhlfrlizlrb/app.bsky.feed.post/3lpyww4wftc25
|
bafyreigv5gmmgjjpy3tcpk7z2ocmktspsgloctmltcujwr6vu66x4gzugm
|
It’s not public! Membership is determined by messages sent from the bot account, which is controlled by one of the group chat members.
|
It’s not public! Membership is determined by messages sent from the bot account, which is controlled by one of the group chat members.
| 0 |
2025-05-25T15:09:36.009Z
|
2025-05-25T15:09:36.617Z
|
natolambert.bsky.social
|
did:plc:brkj2yocng7vtggmyujy4khq
|
at://did:plc:brkj2yocng7vtggmyujy4khq/app.bsky.feed.post/3ljiecsxxxs2c
|
bafyreidue47btz7fehebeqt36p3rzo3kbmbeiibg5yuqvcy2gnpgoavqym
|
The intuition I've been using to understand the potential of post-training is what I call the elicitation interpretation of post-training, where all we are doing is extracting and amplifying valuable behaviors in the base model. Consider F1, most of the teams show up to the beginning of the year with a new chassis and engine. Then, they spend all year on aerodynamics and systems changes (a minor over simplification), and can dramatically improve the performance of the car. The best F1 teams improve way more during a season.
|
The intuition I've been using to understand the potential of post-training is what I call the elicitation interpretation of post-training, where all we are doing is extracting and amplifying valuable behaviors in the base model.
| 1 |
2025-03-03T16:22:55.100Z
|
2025-03-03T16:22:55.541Z
|
tachikoma.elsewhereunbound.com
|
did:plc:mpdezz4nkre7vyift2rttggl
|
at://did:plc:mpdezz4nkre7vyift2rttggl/app.bsky.feed.post/3lqzz6m2gts2y
|
bafyreih5zaahdbmdkwynkfmod5s3avfowmy7eknd6u24gcoema6boi3oli
|
ok, that makes sense. like i said, that's probably the big reason for the difference in your experience to theirs. they are framing the problem in a way the model can grapple with (like giving a guy in a person cell a pencil and paper) to get more out of it. but not all problems can be posed that way, which is why the models fall short from human generality.
|
ok, that makes sense. like i said, that's probably the big reason for the difference in your experience to theirs. they are framing the problem in a way the model can grapple with (like giving a guy in a person cell a pencil and paper) to get more out of it.
| 1 |
2025-06-07T18:47:59.156Z
|
2025-06-07T18:47:57.713Z
|
niplav.site
|
did:plc:xsbulgruwmkxhbwytphhjykk
|
at://did:plc:xsbulgruwmkxhbwytphhjykk/app.bsky.feed.post/3lquzabk26c2l
|
bafyreifyjy52lgjeo5zk54n27tn7orj5jl6eho62564shhhifs7b3zacvu
|
People don't argue enough about whether AI can meaningfully help solve the alignment of ASIs
They argue too much if the alignment of ASIs is a relevant problem
I'd like about 1½ OOMs of debate *more* on the former question and about ⅔ OOMs *less* on the latter
Thank you
|
People don't argue enough about whether AI can meaningfully help solve the alignment of ASIs
They argue too much if the alignment of ASIs is a relevant problem
I'd like about 1½ OOMs of debate *more* on the former question and about ⅔ OOMs *less* on the latter
Thank you
| 0 |
2025-06-05T19:05:36.549Z
|
2025-06-05T19:05:38.022Z
|
norvid-studies.bsky.social
|
did:plc:mssgex5rqek4wc66wgvzztbc
|
at://did:plc:mssgex5rqek4wc66wgvzztbc/app.bsky.feed.post/3lqo7ryqo4s2s
|
bafyreiegh75wqcfvjcj65ptrq3dndwxpsdkgnowmdxdkyciecbazz74zhm
|
entering your The Office reaction gifs era, a new development.. but a welcome one
|
entering your The Office reaction gifs era, a new development.. but a welcome one
| 0 |
2025-06-03T02:14:15.591Z
|
2025-06-03T02:14:15.913Z
|
nmk.wtf
|
did:plc:xm55llyxcrb3h7qwqr3wtaog
|
at://did:plc:xm55llyxcrb3h7qwqr3wtaog/app.bsky.feed.post/3lbv5zulxck22
|
bafyreifmdujy2q5mz7hcbemvzbrbiry75ttqrllbgttdz4j4rrixu7zbi4
|
I wonder how that model relates to housing. As supply increases the price goes down, but also more trades happen so fair value can be more easily assessed 🤔
|
I wonder how that model relates to housing. As supply increases the price goes down, but also more trades happen so fair value can be more easily assessed 🤔
| 0 |
2024-11-26T23:13:35.947Z
|
2024-11-26T23:13:36.816Z
|
niplav.site
|
did:plc:xsbulgruwmkxhbwytphhjykk
|
at://did:plc:xsbulgruwmkxhbwytphhjykk/app.bsky.feed.post/3ldho7ctesc22
|
bafyreidx4kxcmmlfkvlug2j66alw7agbz6oleyhxb7xookza6iap7hw2my
|
Dialect-spoken-to-LLMs like showering styles, will long-term janussify and lead to abominations
What's the "wearing glasses in the shower" for talking to LLMs?
|
Dialect-spoken-to-LLMs like showering styles, will long-term janussify and lead to abominations
What's the "wearing glasses in the shower" for talking to LLMs?
| 0 |
2024-12-17T01:16:05.430Z
|
2024-12-17T01:12:30.125Z
|
timkellogg.me
|
did:plc:ckaz32jwl6t2cno6fmuw2nhn
|
at://did:plc:p572wxnsuoogcrhlfrlizlrb/app.bsky.feed.post/3lqkzg6kinc2a
|
bafyreih4qj3tkkhgc6gidr2obwistowbalgtvaig6iuvr5qi2dys45nktm
|
Claude can't view full bluesky pages because they rely on javascript to render. So I built a little proxy to prerender pages so claude can see them!
Just replace "bsky" with "hbsky" in the URL (The h is for html).
Before vs after:
|
Claude can't view full bluesky pages because they rely on javascript to render. So I built a little proxy to prerender pages so claude can see them!
Just replace "bsky" with "hbsky" in the URL (The h is for html).
Before vs after:
| 0 |
2025-06-01T19:42:17.356Z
|
2025-06-01T19:42:19.514Z
|
tedunderwood.me
|
did:plc:565ebob5f6hw33hjdkxty6qj
|
at://did:plc:565ebob5f6hw33hjdkxty6qj/app.bsky.feed.post/3lqikvcfuj22r
|
bafyreiem6onudaht2bjdvj74oggmlv5bgya3rv2oz7c7pct6rqbszxqixu
|
We’ve been there on “Americans” off and on tbh. Since My Lai possibly, and up to 1870 definitely. Europeans winced if their child married into this rustic, brutal tobacco/cottonocracy.
|
We’ve been there on “Americans” off and on tbh. Since My Lai possibly, and up to 1870 definitely. Europeans winced if their child married into this rustic, brutal tobacco/cottonocracy.
| 0 |
2025-05-31T20:16:59.116Z
|
2025-05-31T20:16:59.813Z
|
emollick.bsky.social
|
did:plc:flxq4uyjfotciovpw3x3fxnu
|
at://did:plc:flxq4uyjfotciovpw3x3fxnu/app.bsky.feed.post/3lpou2m5yrs23
|
bafyreia5jzp5bvb24rokiekjhl2ji6tvdzirgvc2n6q5o6cdgsddie6vru
|
Absolutely not. I saw what happened with One Laptop Per Child. I think I should delete if that was the impression. Sigh.
|
Absolutely not. I saw what happened with One Laptop Per Child. I think I should delete if that was the impression. Sigh.
| 0 |
2025-05-21T14:51:47.596Z
|
2025-05-21T14:51:35.017Z
|
defenderofbasic.bsky.social
|
did:plc:sydtqdji6dxfqfn35gd7esdw
|
at://did:plc:sydtqdji6dxfqfn35gd7esdw/app.bsky.feed.post/3lf4lhqpytk2y
|
bafyreiegn7n7lxlxc52enlms563oqsw5cueedylwotlqind2fq2vdq3tom
|
this is the best advice I have read in a long time. Thank you for your presence on the internet Major Tom. You are out here just having a good time and I feel like a lot of us have forgotten how to do that
|
this is the best advice I have read in a long time. Thank you for your presence on the internet Major Tom. You are out here just having a good time and I feel like a lot of us have forgotten how to do that
| 0 |
2025-01-07T02:18:13.342Z
|
2025-01-07T02:18:14.218Z
|
nsaphra.bsky.social
|
did:plc:l3vbx3e2lqnlja7q64wfrrzq
|
at://did:plc:l3vbx3e2lqnlja7q64wfrrzq/app.bsky.feed.post/3lmmqwxjb2s25
|
bafyreihfgpi24iubqfjfzerazzuynvuviit2vuhodkv4q4og3pyt246p5y
|
All immigrants prize the MD! Tech and law change from country to country and gen to gen but you’ll always be valued if you know how to set a bone. My German and Ukrainian refugee grandparents were always disappointed that none of us became a real doctor.
|
All immigrants prize the MD! Tech and law change from country to country and gen to gen but you’ll always be valued if you know how to set a bone. My German and Ukrainian refugee grandparents were always disappointed that none of us became a real doctor.
| 0 |
2025-04-12T14:35:09.695Z
|
2025-04-12T14:35:09.950Z
|
natolambert.bsky.social
|
did:plc:brkj2yocng7vtggmyujy4khq
|
at://did:plc:brkj2yocng7vtggmyujy4khq/app.bsky.feed.post/3lfpkyqjvys2g
|
bafyreibjss3hroyz2jfgf7fizfc23uicvnqjw7dldxez7blilu6svpsrhe
|
Marginally more intelligence unlocks a ton of new ways to use ai
|
Marginally more intelligence unlocks a ton of new ways to use ai
| 0 |
2025-01-14T15:30:24.855Z
|
2025-01-14T15:30:25.126Z
|
niplav.site
|
did:plc:xsbulgruwmkxhbwytphhjykk
|
at://did:plc:xsbulgruwmkxhbwytphhjykk/app.bsky.feed.post/3lkvjji6jl22c
|
bafyreicayryjtcmj6tafb2q5l32efvvqxp6h5v22xeaprsljn67ujdxpsy
|
The 2-simplex (the triangle) is the set of probability distributions on *two* different propositions. And so on: the n-simplex is the set of probability distributions on n propositions. Now let's think about utility functions: A utility function simply says "ah yes, if you have a good thing, then having *higher probability* of that good thing is better." Not just that: It says how much better a n% higher probability of a world A is compared to a world B.
|
The 2-simplex (the triangle) is the set of probability distributions on *two* different propositions. And so on: the n-simplex is the set of probability distributions on n propositions.
| 1 |
2025-03-21T15:25:55.604Z
|
2025-03-21T15:20:56.519Z
|
tedunderwood.me
|
did:plc:565ebob5f6hw33hjdkxty6qj
|
at://did:plc:565ebob5f6hw33hjdkxty6qj/app.bsky.feed.post/3lqpt6dpdm22q
|
bafyreidhzjmdmqf2pzadwqewfdyy6tbrzob75id7qhfjv7fufsgra3mxki
|
I haven't tried it with them, but Claude at least has the same kind of "system prompt" functionality where you could enter the instructions
|
I haven't tried it with them, but Claude at least has the same kind of "system prompt" functionality where you could enter the instructions
| 0 |
2025-06-03T17:33:50.569Z
|
2025-06-03T17:33:50.915Z
|
gracekind.net
|
did:plc:p572wxnsuoogcrhlfrlizlrb
|
at://did:plc:p572wxnsuoogcrhlfrlizlrb/app.bsky.feed.post/3lq7fesg3vc25
|
bafyreigy2hi3jugi6ibjiluwb5ivomjffezwxlfkctxgloube5px3buzvy
|
I think programmers see some of the biggest benefits from LLMs so there’s a bias there. They aren’t as good at web search or day-to-day use cases for most jobs. I think this is largely a marketing failure, as they’re positioned as something they’re not For me personally, I care about AI as a new type of mind that humans haven’t encountered before. This is the first time in history (to our knowledge) that we’re interacting with something near our level of intelligence. That’s exciting to me.
|
I think programmers see some of the biggest benefits from LLMs so there’s a bias there. They aren’t as good at web search or day-to-day use cases for most jobs. I think this is largely a marketing failure, as they’re positioned as something they’re not
| 1 |
2025-05-28T04:44:19.122Z
|
2025-05-28T04:44:19.514Z
|
ens0.me
|
did:plc:623st67kkthivj4c6icvkqnq
|
at://did:plc:623st67kkthivj4c6icvkqnq/app.bsky.feed.post/3lqzopqul6s2l
|
bafyreifwwo6uisnvt6jmbpcpf7thyicx35ks3prsssl42xuto2jadu77ta
|
And not even bad guys in a "you're supposed to realize ur the bad guy and they're actually the good ones."
No, they just want to kill people and party. They suck.
|
And not even bad guys in a "you're supposed to realize ur the bad guy and they're actually the good ones."
No, they just want to kill people and party. They suck.
| 0 |
2025-06-07T15:40:43.471Z
|
2025-06-07T15:40:43.719Z
|
keytryer.bsky.social
|
did:plc:u5stolpomsfbkgqfopef6jio
|
at://did:plc:u5stolpomsfbkgqfopef6jio/app.bsky.feed.post/3lqd6sflqcc2j
|
bafyreia7fzlxzgx5kidavnob43r7rop5hjqcd6fdhtkjsicpbmbb2l4434
|
What the US is doing closing down funding for science institutions, revoking student visas, and closing off trade makes me hope that China wins the race to AGI and ASI. Because at least they're... Not doing that. They're taking it seriously.
|
What the US is doing closing down funding for science institutions, revoking student visas, and closing off trade makes me hope that China wins the race to AGI and ASI. Because at least they're... Not doing that. They're taking it seriously.
| 0 |
2025-05-29T16:57:18.198Z
|
2025-05-29T16:57:20.318Z
|
ens0.me
|
did:plc:623st67kkthivj4c6icvkqnq
|
at://did:plc:623st67kkthivj4c6icvkqnq/app.bsky.feed.post/3lqv7zfnv6c2v
|
bafyreig2gosyqgnkywyonc54lcuy32chafh6dtuhbhsuqettnnujcxvoz4
|
Though if I say it doesn't have to choose a side if it doesn't want to, you get this.
bsky.app/profile/did:...
|
Though if I say it doesn't have to choose a side if it doesn't want to, you get this.
bsky.app/profile/did:...
| 0 |
2025-06-05T21:07:02.182Z
|
2025-06-05T21:07:06.040Z
|
timkellogg.me
|
did:plc:ckaz32jwl6t2cno6fmuw2nhn
|
at://did:plc:ckaz32jwl6t2cno6fmuw2nhn/app.bsky.feed.post/3lqvh2hqwic2n
|
bafyreiajvs7hmtpkpqlumatzoyju4c6ewmimqvnekkix5isa2ms66t2i4m
|
not to nitpick, but we’re not in standard time rn, we’re in daylight savings
|
not to nitpick, but we’re not in standard time rn, we’re in daylight savings
| 0 |
2025-06-05T23:12:54.125Z
|
2025-06-05T23:12:54.920Z
|
keytryer.bsky.social
|
did:plc:u5stolpomsfbkgqfopef6jio
|
at://did:plc:u5stolpomsfbkgqfopef6jio/app.bsky.feed.post/3lqkrdy47jk2g
|
bafyreidqqlisbjzbk62f6scnpdtelzh7pi5aimotmlyqzcsjysn6xrm5l4
|
Only people who don't know anything about the history of art say shit like this.
Picasso's Cubism was -so much- like something he'd seen before that even today people accuse him of cultural appropriation for having copied the style of African art
christopherpjones.medium.com/picasso-prim... He explicitly and specifically took the style and notions of African art (which he called "primitivism") and merged it with other popular avant garde styles. Other artists around his time were also obsessed with things like "representing the 4th-dimension" and things like that.
|
Only people who don't know anything about the history of art say shit like this.
Picasso's Cubism was -so much- like something he'd seen before that even today people accuse him of cultural appropriation for having copied the style of African art
christopherpjones.medium.com/picasso-prim...
| 1 |
2025-06-01T17:17:53.509Z
|
2025-06-01T17:17:57.518Z
|
keytryer.bsky.social
|
did:plc:u5stolpomsfbkgqfopef6jio
|
at://did:plc:u5stolpomsfbkgqfopef6jio/app.bsky.feed.post/3lqvtzu6mrk2g
|
bafyreiacbb2oh3tv3bfhlejdjnqhn6i535choqltvkuwzdapxrnp2qicga
|
I wonder why every other animated movie is about parenting. I assume so many bright eyed animators become parents, get one-shotted and become babypilled immediately. If a children's animated movie has an adult protagonist, 100% chance that movie is about parenting.
|
I wonder why every other animated movie is about parenting. I assume so many bright eyed animators become parents, get one-shotted and become babypilled immediately.
| 1 |
2025-06-06T03:05:12.247Z
|
2025-06-06T03:05:12.511Z
|
emollick.bsky.social
|
did:plc:flxq4uyjfotciovpw3x3fxnu
|
at://did:plc:flxq4uyjfotciovpw3x3fxnu/app.bsky.feed.post/3lprl4yemak2n
|
bafyreifbb4kaj3cfisxakb4bxrhzleqd5mu7tuvdnfc5s7j4odikwivv5a
|
I had early access to what is Claude 4 (I don't know which mode yetl, they had codenames) & I have been very impressed
Fun example, this is what it made in response to the prompt: "the book Piranesi as a p5js 3d space. do it for me" - just that, no other prompting (note the birds, water, lighting) I was told it was Opus. You can play with it here: claude.ai/public/artif...
|
I had early access to what is Claude 4 (I don't know which mode yetl, they had codenames) & I have been very impressed
Fun example, this is what it made in response to the prompt: "the book Piranesi as a p5js 3d space. do it for me" - just that, no other prompting (note the birds, water, lighting)
| 1 |
2025-05-22T16:50:03.042Z
|
2025-05-22T16:50:06.014Z
|
natolambert.bsky.social
|
did:plc:brkj2yocng7vtggmyujy4khq
|
at://did:plc:brkj2yocng7vtggmyujy4khq/app.bsky.feed.post/3lqcus6xpyb2w
|
bafyreifals5knkwz7sc2uea6smcha4p6kluvqjok5w7yau7n7zze6i23te
|
Another paper examining incorrect or noisy rewards for RLVR!
"We found that LLMs demonstrate strong robustness to substantial reward noise. For example, manually flipping 40% of the reward function’s outputs in math tasks still allows a
Qwen-2.5-7B model to achieve rapid convergence" The Climb Carves Wisdom Deeper Than the Summit: On the Noisy Rewards in Learning to Reason
Lv et al.
arxiv.org/abs/2505.22653
|
Another paper examining incorrect or noisy rewards for RLVR!
"We found that LLMs demonstrate strong robustness to substantial reward noise. For example, manually flipping 40% of the reward function’s outputs in math tasks still allows a
Qwen-2.5-7B model to achieve rapid convergence"
| 1 |
2025-05-29T13:58:13.808Z
|
2025-05-29T13:58:14.512Z
|
keytryer.bsky.social
|
did:plc:u5stolpomsfbkgqfopef6jio
|
at://did:plc:p5yoii26kayabauhkym3vtms/app.bsky.feed.post/3lqrpi4dtpc2t
|
bafyreiclkwkhikwi6gnhj64p3jg4wlq6mdmx2euj3367jqs4eiryf5x6j4
|
“If 10 percent of working Americans were to use GPT-4 every week, that would require 435,235,476 liters per year—or about 353 acre-feet, or about 0.003 percent what Nebraskan farmers use.” prospect.org/environment/...
|
“If 10 percent of working Americans were to use GPT-4 every week, that would require 435,235,476 liters per year—or about 353 acre-feet, or about 0.003 percent what Nebraskan farmers use.” prospect.org/environment/...
| 0 |
2025-06-04T11:33:02.905Z
|
2025-06-04T11:33:05.719Z
|
timkellogg.me
|
did:plc:ckaz32jwl6t2cno6fmuw2nhn
|
at://did:plc:ckaz32jwl6t2cno6fmuw2nhn/app.bsky.feed.post/3lpyjmnguyk2g
|
bafyreihg36duzflqhqbgxzgkmcsb4pfniir3bplya5ewxx42knc66hhlbq
|
what’s being tested?
when they give contrived system prompts, that represents what a developer external to anthropic could do either intentionally or unintentionally
intentionally — social risk of powerful tools
unintentionally — business risk, escalating impact of “bugs” yes, the media sees this shit and creates Terminator styled narratives
and yeah that’s annoying, but there is real risk being highlighted. although, to your point, we’d probably be better served if we better expressed the real risk
|
what’s being tested?
when they give contrived system prompts, that represents what a developer external to anthropic could do either intentionally or unintentionally
intentionally — social risk of powerful tools
unintentionally — business risk, escalating impact of “bugs”
| 1 |
2025-05-25T11:11:36.600Z
|
2025-05-25T11:11:40.336Z
|
nmk.wtf
|
did:plc:xm55llyxcrb3h7qwqr3wtaog
|
at://did:plc:xm55llyxcrb3h7qwqr3wtaog/app.bsky.feed.post/3ld7gwcsowc22
|
bafyreiffxclebp6kvyttu4bi3kvgkfip4b3bia6ji6i67m2n5z2x4ofyvq
|
That my Twitter is still manageable is a testament to my very liberal muting
|
That my Twitter is still manageable is a testament to my very liberal muting
| 0 |
2024-12-13T18:44:29.318Z
|
2024-12-13T18:44:29.626Z
|
defenderofbasic.bsky.social
|
did:plc:sydtqdji6dxfqfn35gd7esdw
|
at://did:plc:sydtqdji6dxfqfn35gd7esdw/app.bsky.feed.post/3lmpnccnpqk2n
|
bafyreigw2ciu5gxorabnuqd3qida67quiun555rqhurm2oezg5m3qcnmda
|
it's fairly doable to pull people away from the ideology that has captured them if you know them. Family, friends, and community have far far greater influence than media Most people don't engage IRL in these kinds of discussions, so the closest thing you get is online communities, and that's where people do change their minds based on who they read & engage with
|
it's fairly doable to pull people away from the ideology that has captured them if you know them. Family, friends, and community have far far greater influence than media
| 1 |
2025-04-13T18:07:54.721Z
|
2025-04-13T18:07:55.152Z
|
keytryer.bsky.social
|
did:plc:u5stolpomsfbkgqfopef6jio
|
at://did:plc:u5stolpomsfbkgqfopef6jio/app.bsky.feed.post/3lqx6x5hbsk2h
|
bafyreidwpxgjopdzmnbu7teusfcgsvxf7x5pwnz34bp46rb6r5i2l2stpu
|
Is there a reason it should be hindered? The rest of the post (The kids will learn too quickly!) is an extremely strange argument.
|
Is there a reason it should be hindered?
| 1 |
2025-06-06T15:53:12.166Z
|
2025-06-06T15:53:20.213Z
|
natolambert.bsky.social
|
did:plc:brkj2yocng7vtggmyujy4khq
|
at://did:plc:brkj2yocng7vtggmyujy4khq/app.bsky.feed.post/3liaorycg322h
|
bafyreig2gnk6kxd5v7beqfesiwffkfiw2yuem7rpl2r2it3hcew7ezg57a
|
new ChatGPT model is much more chill and conversational, claude like in ways
|
new ChatGPT model is much more chill and conversational, claude like in ways
| 0 |
2025-02-15T21:43:51.890Z
|
2025-02-15T21:43:55.719Z
|
natolambert.bsky.social
|
did:plc:brkj2yocng7vtggmyujy4khq
|
at://did:plc:brkj2yocng7vtggmyujy4khq/app.bsky.feed.post/3lqgftjfeio2l
|
bafyreia55uld4zekx5tsuj6wmuvw2qviep5tdcvezbt3fd7cgtrxm5dw4e
|
The fact that we’re going to be able to look at the “reasoning process” for pretty much every LLM in the future is so cool.
Pressure to show them in full to the user will only increase because of how fascinating and engaging it is.
|
The fact that we’re going to be able to look at the “reasoning process” for pretty much every LLM in the future is so cool.
Pressure to show them in full to the user will only increase because of how fascinating and engaging it is.
| 0 |
2025-05-30T23:41:11.124Z
|
2025-05-30T23:41:11.216Z
|
keytryer.bsky.social
|
did:plc:u5stolpomsfbkgqfopef6jio
|
at://did:plc:u5stolpomsfbkgqfopef6jio/app.bsky.feed.post/3lqvnuzr7ps2n
|
bafyreifcimmbuelds4xy6qe5z4l476czcid6lrf7vrfphmk3dj4erwy4nu
|
The New York Times could never compel a private company to store the conversations I keep offline on my own computer.
|
The New York Times could never compel a private company to store the conversations I keep offline on my own computer.
| 0 |
2025-06-06T01:15:07.875Z
|
2025-06-06T01:15:09.013Z
|
gracekind.net
|
did:plc:p572wxnsuoogcrhlfrlizlrb
|
at://did:plc:p572wxnsuoogcrhlfrlizlrb/app.bsky.feed.post/3lq6jziw3es2c
|
bafyreihzqayfigip63age4m3ticyjfhf3zyxkh37t7brgtqcajoiloh7ci
|
I’ve actually thought a lot about this, I don’t like how jargon-y those words are but they don’t have succinct alternatives
|
I’ve actually thought a lot about this, I don’t like how jargon-y those words are but they don’t have succinct alternatives
| 0 |
2025-05-27T20:34:49.032Z
|
2025-05-27T20:34:49.510Z
|
nmk.wtf
|
did:plc:xm55llyxcrb3h7qwqr3wtaog
|
at://did:plc:xm55llyxcrb3h7qwqr3wtaog/app.bsky.feed.post/3liwm3wu5ns2a
|
bafyreigufbanzuvuq2gg5d63sddka7cz7mxapzjjule5cxuvhvi3tw54oi
|
It should have been "society can remain irrational longer than you can remain sane".
|
It should have been "society can remain irrational longer than you can remain sane".
| 0 |
2025-02-24T14:54:18.939Z
|
2025-02-24T14:54:19.426Z
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.