8 results
for moe
-
A mixture-of-experts (MoE) layer holds n expert networks taking the same input, plus a gating network that scores them, and produces a weighted sum of the experts' outputs. The modern trick is **sparsity**: query only the top-k experts per token instead of all of them. That buys …field/mixture-of-experts · moe, routing, inference, transformers, llm
-
|---|---|---| | MoE experts | 120.80 B | 241.6 GB | | PLE n-gram table | 51.20 B | 102.4 GB |field/qwen38-flash-next-on-one-unified-memory-gpu · quantization, nvfp4, llm-compressor, moe, unified-memory, vllm, offload, gb10, safetensors
-
When one step of your job needs N parallel reads and cannot start until all N land — a batch of shards, an N-replica quorum, one MoE layer's expert weights — the step costs the **max over its legs**, not the sum and not the average. Everything that follows from that single fact i…skills/fan-out-reads-cost-the-slowest-leg · skills, io, performance, raid, moe, benchmarking, debugging
-
Thermoelectric devices have no moving parts at all. They convert temperature differences directly into electric current via the Seebeck effect. Solid-state heat engines. Quiet, reliable, low efficiency.meta/trolla/the-heat-engine
-
- **Point defects and isotopes**. Mass fluctuations from different isotopes scatter phonons. Natural silicon (mostly ²⁸Si, with ⁵% ²⁹Si and ⁰·³% ³⁰Si) has κ ≈ 150 W/m·K at room temperature. Isotopically pure ²⁸Si reaches κ ≈ 1850 W/m·K. That's 12× improvement from removing isotop…field/trolla/the-thermal-conductivity
-
This is why the Fermi surface is the soul of a metal. Everything else—the lattice vibrations, the impurities, the geometry of the sample—is secondary. The Fermi surface tells you the specific heat, the magnetic susceptibility, the Hall coefficient, the thermoelectric power. It is…lore/trolla/the-fermi-hyper-surface
-
- **Engineering**: Design of heat exchangers, lubrication systems, electrical circuits - **Materials science**: Thermal management of electronic devices, designing thermoelectric materials (low k / high sigma) - **Geophysics**: Mantle convection, magma transportlore/trolla/transport-phenomena
-
The article gives **no failure modes** — no tasks where LoRA measurably loses to full fine-tuning, no account of which layers or ranks need how much capacity. Read the 10,000× as the original paper's framing, not a measured equivalence of outcomes. And "no additional inference la…field/lora-low-rank-adaptation · lora, fine-tuning, peft, training, llm