- Math behind Attention- Q, K, and V
- Math behind √dₖ Scaling Factor in Attention
- Math Behind Backpropagation
- Math Behind Gradient Descent
- Math Behind Cross-Entropy Loss
- Math Behind RoPE (Rotary Position Embedding)
- RMSNorm (Root Mean Square Layer Normalization)
Hardware is the bridge between AI and the physical world
Atoms and bits must work together to create future systems embedded with physical intelligence
We wrote a guide for those curious about the atoms.
the world of advanced packaging is getting crazier and crazier and I am here for it, and we are still so early
(glass panel stuff no earlier than 2029)
(10 Tbps/mm clearly needs DWDM, or I guess a big array of VCSELs)
At #VLSI2026, Intel Foundry showcased innovations powering the future—from Intel 18A-P advancements like Power Boost and backside power delivery to next-generation technologies.
Read more: https://t.co/g93LmwOkOI
#IntelFoundry
New logic Vt offering for #Intel18A-P:
- ULVTLL (ultra low voltage low leakage) between LVT and ULVT.
Via a thinner handling wafers, new materials and optimized EDA tool flow the thermal conductivity is improved.
#Intel18A-P is currently in risk production
Starlink UL signal reverse engineering. Never saw such a concrete and detailed and verified by real setup before (Maybe I am wrong). There are many DL signal reverse engineering.