Sabitlenmiş Tweet
ricky
21.9K posts
The retelling of dune should be a symposium of kick streams that cover 10,000 years of history and push the propaganda of each competing great house and faction
Tenobrus@tenobrus
agreed on chalamet. beyond that, the dune movies are absolutely fucking crippled by the fact that they're *movies* instead of an HBO show. they're forced into this shitty liminal space where they're simultaneously much too long and much too short to effectively tell the story.
English
agreed on chalamet. beyond that, the dune movies are absolutely fucking crippled by the fact that they're *movies* instead of an HBO show. they're forced into this shitty liminal space where they're simultaneously much too long and much too short to effectively tell the story.
roon@tszzl
the dune movies were doomed from the start to be good and not great due to the casting of chalamet as paul. he does not have the gravitas for a child-god and is much better suited for kind of silly coming of age movies
English
Does this mean I have to stop using `uv` now? 😧
Jamison 🦆@jmelahman
I don't really see myself using an OpenAI product going forward.
English
@keysmashbandit @homieomorphism i was trying so hard to avoid computers at 18
English
We've entered into an agreement to join OpenAI as part of the Codex team.
I'm incredibly proud of the work we've done so far, incredibly grateful to everyone that's supported us, and incredibly excited to keep building tools that make programming feel different.
English
@rickyflows @lisatomic5 how long until we automate women and give REAL romance a shot at propagation
English
the response to this post is fascinating, nearly all women reacting uniformly negatively
Zoya🕊️@Zoya_ki_batein
A nanobot helping a sperm with motility issues along towards an egg. These metal helixes are so small they can completely wrap around the tail of a single sperm and assist it along its journey
English
@tautologer i wonder if it would still be human readable, probably not right?
English
you can trivially (but expensively) fine-tune a compaction model by using "similarity of weight activation pattern on the uncompacted context vs compaction summary" as a reward signal, right?
English
I feel like LLMs are a local optimum we're stuck at because they're easy to train from scratch because training data already abounds
English
@MarkovMagnifico @tautologer I was thinking of his appearance on dwarkesh
English
@rickyflows @tautologer by why is the attention transformer the ideal vessel for scaling up compute?
English
