@andrej@trees.social avatar

andrej

@andrej@trees.social

Hi, I am Andrej. I am the admin on trees.social.

I care about equity, the life on earth and people.

I enjoy gardening, hamradio (@dl5psy) and tinkering. My account may contain #catcontent.

I am a RZL member and go by the nick niom there.

Trans rights are human rights.

This profile is from a federated server and may be incomplete. For a complete list of posts, browse on the original instance.

rahmstorf , to random German
@rahmstorf@fediscience.org avatar

Just attending a 3-day workshop on the danger of abrupt ocean circulation changes.
Discussions are quite worrying. E.g. in 35 to 45 % of high quality models, convection in the open North Atlantic collapses in the 2030s due to . -> Major climate disruption. Not good.

andrej ,
@andrej@trees.social avatar

@rahmstorf Eine Bilfbeschreibung waere ganz toll.

rahmstorf , to random German
@rahmstorf@fediscience.org avatar

Warum nehmen die Regenmengen gerade bei Gewitterschauern besonders stark zu?
Das erklärt verständlich mein immer wieder aktueller Artikel von 2016 bei Spektrum.

https://scilogs.spektrum.de/klimalounge/warum-die-globale-erwaermung-mehr-extremregen-bringt/


andrej ,
@andrej@trees.social avatar

@rahmstorf Bildbeschreibung hinzufügen wäre toll.

rahmstorf , to random German
@rahmstorf@fediscience.org avatar

Freiheit, Wohlstand & Sicherheit wird es in den kommenden Jahrzehnten nur geben, wenn es gelingt, die durch fossile Emissionen verursachte #Erderwärmung und Destabilisierung des Klimas zu stoppen.
#Hochwasser
#Starkregen

andrej ,
@andrej@trees.social avatar

@rahmstorf Eine Bildbeschreibung waere ganz toll.

andrej , to random German
@andrej@trees.social avatar

@mnl Which AI/LLM services are you using regularly?

mnl , to random
@mnl@hachyderm.io avatar

will cure cancer and solve climate change!

andrej ,
@andrej@trees.social avatar

@mnl :trollface:

andrej , to random German
@andrej@trees.social avatar

@mnl Not sure if you saw this already: https://arxiv.org/pdf/2211.03622.pdf

"Do Users Write More Insecure Code with AI Assistants?"

mnl , to random
@mnl@hachyderm.io avatar

I'm seriously impressed by the quality of the MoE and even 7B models I've run on my 4 year old M1 over the weekend. This is absolutely solid for most of the really useful ways of LLMs I actually use them for day to day.

Already most of these tools allow me to use local models or cheaply hosted versions of them. I can see the total power used in daily inference be about 10 minutes of running 4 cores.

These models and community finetunes are already much more fun than GPT or claude.

1/

andrej ,
@andrej@trees.social avatar

@mnl While I kind of agree, I think most companies will stay with the big LLMs for convenience and legal reasons.

andrej ,
@andrej@trees.social avatar

@mnl Costs might tip the scale.

andrej ,
@andrej@trees.social avatar

@mnl Hard agree :D

andrej , to random
@andrej@trees.social avatar

@mnl https://www.golem.de/news/nvidia-ace-ausprobiert-endlich-npcs-die-was-zu-sagen-haben-2401-181010.html

  • All
  • Subscribed
  • Moderated
  • Favorites
  • •
  • random
  • test
  • worldmews
  • mews
  • All magazines
    •