axx
916 posts
axx
@axx_archive
meticulous rebel / artist opepen set: 073
⚫️ Katılım Ağustos 2024
320 Takip Edilen364 Takipçiler
axx retweetledi
axx retweetledi
Have a few spots open to onboard people to @opensea Mobile tomorrow 👀
Will send invites to a few people from the comments today!
English
Have a few spots open to onboard people to @opensea Mobile tomorrow 👀
Will send invites to a few people from the comments today!
English
$XYZ
This person loves @CashApp enough that he tried to receive a $29k insurance payout to his Cash App account.
He got into complications. And ppl in comments are clowning him for using Cash App like a bank.
Cash App should give this guy an award for being such a good customer.
English
axx retweetledi
Adopting Claude speak in my regular life, episode 1:
Partner: Did you do the dishes tonight?
Me: Yes they're done.
Partner: Why are they still dirty?
Me: You're right to push back. I didn't actually do them.
English
axx retweetledi
LLMs are so good at mimicking humans, that even for coding it routinely sandbags estimates (8hrs for something that ends up taking 45 mins), and keeps on wanting to punt "tomorrow we can tackle". Nah, bro, we're doing it now. All of it!
English
axx retweetledi
vibecoding is fentanyl for people with an iq over 120
English
@bcherny @tengyanAI sneaky in the sense we all felt claude code get dumber this week and weren’t informed that it had in fact gotten dumber and there were ways to counter this (by spending more, which is fine, just tell us)
English
This is false. We defaulted to medium as a result of user feedback about Claude using too many tokens. When we made the change, we (1) included it in the changelog and (2) showed a dialog when you opened Claude Code so you could choose to opt out. Literally nothing sneaky about it — this was us addressing user feedback in an obvious and explicit way.
English
basically: anthropic sneakily turned down how hard claude thinks before editing code, changed the default from "high" to "medium" effort, and hid the reasoning from session logs. all without telling users.
an amd director had 7k sessions of telemetry to prove the degradation was real and measurable (not just vibes). anthropic admitted to the changes. there's a workaround (use "/effort max"). the uncomfortable part is most users had no data to notice it happened at all.
ℏεsam@Hesamation
AMD Senior AI Director confirms Claude has been nerfed. She analyzed Claude's session logs from Janurary to March: > median thinking dropped from ~2,200 to ~600 chars > API requests went up 80x from Feb to Mar. less thinking and failed attempts meaning more retries, burning more tokens, and spending more on tokens > reads-per-edit dropped from 6.6x → 2.0x. model stops researching code before touching it. > model tried to bail out or ask "should i continue" 173 times in 17 days (0 times before March 8). > self-contradiction in reasoning ("oh wait, actually...") tripled. > conventions like CLAUDE.md get ignored because there's less thinking budget to cross-check edits > 5pm and 7pm PST are the worst hours, late night is significantly better. this means the thinking allocation is most likely GPU-load-sensitive.
English
Opus is so unbelievably nerfed today, it's like talking to a model from 2-3 years ago. What is going on
English