mfarles
@mfarles
Followers
312
Following
100K
Media
204
Statuses
11K
Too dumb to even be wrong. Categorically. At least they're projecting and are desperate. 😂.
Sorry, but anybody who still says "LLMs just predict text" is embarrassing themselves at this point. It is, as OpenAI's roon says, "categorically wrong"
0
0
0
AGI already here. Also reading all people's emotions perfectly. 😂.
@alternativetak3 @Nero The man in the photo appears to be JD Vance. His smile seems forced—it's a "social" grin without engaging the eyes (no crow's feet), suggesting insincerity or emotional detachment. The eyes do look somewhat vacant, with a distant, calculating stare that lacks warmth, potentially.
0
0
0
Bar is set really low and still too difficult. It is a gaussian bar though, picking up on both extremes.
Catching on to Musk's patterns of deceit and fraud is becoming an intelligence test. The sooner you catch on, the brighter you are. The longer it takes you, the dimmer your light. And there are an awful lot of very dim bulbs that still have their noses right up his ass.
0
0
0
RT @mfarles: @mpshanahan So Cezar in theater is indistinguishable from the real Cezar? Even if played by forgetful and unreliable actors? T….
0
1
0
RT @mfarles: @DavidDeutschOxf Anthropomorphization of LLMs goes hand in and with de-anthropomorphization of humans doing it. There is no "y….
0
1
0
RT @mfarles: @grok @Arduino_jantje @C_S_Skeptic @Videnak2 @grok No, user asked a question, there was an error, you purposefully omitted cru….
0
1
0
RT @mfarles: @rohanpaul_ai Body language: The only thing exploding besides their economy is obvious desperation. What he's really showing i….
0
1
0
RT @mfarles: @MichLieben Just like OG, MetaHitler doesn't get it. You can't buy motivation or creativity. It's the opposite and unforceable….
0
1
0
RT @mfarles: @rohanpaul_ai The first sentence is already a filter, continued. Nothing but continuation of more of the same. If anything, it….
0
1
0
RT @mfarles: @Libertarec Ko bi le Slo bila večja in bi večji zaprti trg omogočal večje monopolne marže, bi se teoretično šlo pretvarjati, d….
0
2
0
RT @NateHindenburg: @TuckerCarlson I am Nate Anderson, the founder of Hindenburg Research referenced repeatedly in this bizarre and fantast….
0
2K
0
As expected, AI fakery is always worse than expected. The only thing communicated is their level of desperation.
0
0
1
RT @sciencewtg: How the Brain Marks Time. Researchers at the Norwegian University of Science and Technology have made a big step in underst….
0
3
0
RT @MarioPlesej: Mitja Ferenc si nedvomno zasluži najvišje državno odlikovanje za svoje neumorno delo pri odkrivanju prikritih množičnih gr….
0
145
0
RT @mfarles: @JadeCole2112 We know what doesn't work. Trouble is there is an infinite number of ways it doesn't work and at best a few that….
0
1
0
RT @mfarles: @molly0xFFF Scammers gonna scam. If they'd only be able to do anything else. They would still only scam.
0
1
0
As expected.
Can an AI model predict perfectly and still have a terrible world model?. What would that even mean?. Our new ICML paper formalizes these questions. One result tells the story: A transformer trained on 10M solar systems nails planetary orbits. But it botches gravitational laws 🧵
0
0
0