@FractalAppsCom I appreciate the candid reply. I think genuine perspectives will be what is truly valuable in the next half-decade. At least when the pendulum swings back from the swarm of AI content. I wish you the best though, fellow human!
@FractalAppsCom Second sign, the short statement triplet:
“One map. Thirty seconds to orient. Every major architectural decision readable from the terrain alone.”
@maddycr3@HowToAI_ I detest the hyperbole that plagues every peddler on the internet space. It’s akin to “step right up, step right up! You are not gonna believe this!”
@gerardsans@pors A single example would be super supportive of your position. There is nothing in a neuron that cannot be modeled mathematically. A position of ”it’s impossible because I am very smart” seems shaky at best.
@gerardsans@pors I wouldn't say this is accurate at all. I can think of several ways neurons can be functionally modeled and employed digitally. One would think you've proven otherwise?
Building a new low poly game with @godotengine
Does using stylized explosions take away from game experience? Or does it need to have smoke, fire, and damage to be fun?
The Effort display in Claude Code should never disappear. It's not visual noise. The fact that it's combined with "For this session only" makes it seem ephemeral, and the fact that it's not displayed makes it seem like you need to constantly make sure
@jenny_wen@joellewenstein
New Anthropic research: Natural Language Autoencoders.
Models like Claude talk in words but think in numbers. The numbers—called activations—encode Claude’s thoughts, but not in a language we can read.
Here, we train Claude to translate its activations into human-readable text.
/peek let's your parallel agents see what each other is doing without needing to have full context
https://t.co/8OtlGpjUyM
@ClaudeDevs@claudeai@jenny_wen