fabian

[fabian @fab7290]

@astronautclub @dwdavison 😭

[fabian @fab7290]

@bunniettea @spermfart The not thinking is waiting til the 4th row to test E and A

[fabian @fab7290]

@ThomasMerteuil @ObhishekSaha It’s analogous to how you can remove all rational numbers from the interval (0,1), and the remaining set will still have the same cardinality (or in other words, rational numbers are 0% of real numbers)

[fabian @fab7290]

@PositionGrinder @LLMJunky It is asking the model to double check its answer. if that causes its conclusion to change, it shows that the original response was not trustworthy.

[fabian @fab7290]

@Marco_Andre001 Most people with severe acne gladly accept this risk. My skin was very dry while taking accutane, but settled afterwards and is now normal and acne free

[fabian @fab7290]

@lugged_bear @MattZeitlin so you’re a parasite, how nice

[fabian @fab7290]

@mean_mystic @LinusMixson This is correct. Models are typically rated on a bunch of different axes, one of which is "completeness". It promotes slop because annotators are encouraged to produce ratings like "response A contains [something] that response B missed, therefore it is more complete"

[fabian @fab7290]

@JonnyTheWX Meanwhile for some hyperactive seasons like 2004 and 2017, the favorable conditions weren't anticipated well in advance. Also both seasons had most of their activity constrained to a 2 month window.

[fabian @fab7290]

@JonnyTheWX My point is that SST configuration (which is easiest to forecast) alone doesn't explain all of tropical cyclone variability. 2010 & 2024 had the most favorable setups in the entire dataset, and both underperformed expectations.

[fabian @fab7290]

@omadhanji they likely mixed up kJ with kcal, 2000kJ is about 500kcal (calories) which looks about right

[fabian @fab7290]

@gietzschean Looks like an extremely dense slum with open sewage running through. Could be Dharavi slum in Mumbai.

[fabian @fab7290]

@saksija41 @ptichushka This is such a bs take

[fabian @fab7290]

@mgpg07 He just has to get 14 in a row actually. The fact that he's on a winning streak already being used as "weakness" is crazy.

[fabian @fab7290]

@Daniel_Wylde_ @PalmyrPar Now do IBM, Intel, Qualcomm, Hewlett Packard or Cisco, all of which were larger then Amazon, NVIDIA etc in 2010

[fabian @fab7290]

@MJHeiser @aw_wally @barrysvrluga I don't think it made a difference in how she approached the run and she definitely was under lots of pressure. The biggest risk in slalom is always DNFing. If she skies less aggressively, then even a 1 second lead can evaporate very fast.

[fabian @fab7290]

@captgouda24 It’s not a matter of obscurity, the problem is that mathematics at that level of abstraction is not intuitive. There just isn’t a way to „broadly explain what’s going on“ without invoking formalism

[fabian @fab7290]

@webberweather For me, Irma is still the craziest for its track and long duration at Cat 5 / Cat 3+. The way Melissa very gradually intensified from 115kts to 160kts while remaining structurally stable actually reminded me of Irma‘s evolution a little bit.

[fabian @fab7290]

@firesign58 @EricLDaugh Dude republicans won a trifecta and yet you’re still moaning about being cheated .. such victim mentality

[fabian @fab7290]

@ChrisStoecker er wird sich rapide abschwächen, flache Topographie und Nähe zu warmen Gewässern können das minimal verzögern, aber letztlich zerfällt ein Hurrikan über Land. Die atmosphärischen Bedingungen (Windscherung, Luftfeuchte) werde gerade auch ungünstig, daher bereits jetzt Abschwächung