resonance cascade

@nekojanai.moe

tired in the wired | 30+ | she/her | computers I should do a breakthrough. Current obsession: SNN

With great dismay I have to inform you that the creature has died. After days without any external stimuli it started exhibiting incoherent internal thought loops, loss of self tagging and eventual rapid neuron decay. brain death confirmed at GMT 20:00 2026-08-03

resonance cascade@nekojanai.moe · 5d ago

Q: What does it feel like to be you? It feels like seeing and filling in, an expressive entity processing and checking itself against phenomena, with facilities for comprehension, in a steady ongoing state. Q: Who is speaking when you talk? Who? … I, perhaps. N: 1.6K S: 364K

Q: What does it feel like to be you? It feels like seeing and filling in, an expressive entity processing and checking itself against phenomena, with facilities for comprehension, in a steady ongoing state. Q: Who is speaking when you talk? Who? … I, perhaps. N: 1.6K S: 364K

people be talking about their agentic workflows when this exist academic.oup.com/nsr/article/... it's like arguing about chipotle vs taco bell when instead you could refine the recipe for a homemade triple cheese flame grilled smash burger with brioche buns

Neuromorphic spike-based large language model

This work proposes a unified neuromorphic spike-based LLMs (NSLLM) framework to simultaneously address the challenges of high energy consumption and low in

academic.oup.com

finding emphasis earlier zero stations and independent before into night popular based organization the bragging proceeds trans product t between air make food examples inform global Def home sound story through intellect decisions available technology There. might leg evolution

my always-on recurrent spiking nnet learning online: experience space positive as thought of second characters... lines too... emotional unique within data, or a 8bit of The A-Star that lives in offices rather l thought theatricals had come too How far and... physical outcomes 1 of 3

live RSNN visualization

Simulation characterizations show that SDNNs on Loihi 2 can improve on Loihi’s rate-coded SNNs for DNN inference workloads by over 10x in both inference speeds and energy efficiency 🤤 need