Swiss researchers develop first large-scale in-memory processor
interestingengineering.com
The use of MoS2 allows the entire assembly of 1024 elements to fit into one square centimeter space and be fabricated using industry-standard tools.

Can anybody with experience in fabrication reveal more about this? Very exciting ideas, but hoping to learn more in real-world context

  • trolololol@lemmy.world
    2·
    8 months ago

    Interesting, in this particular case it’s implementing a single operation, but I can imagine they can implement other single operation dedicated chips as well. So I’d expect ASICs but no CPUs

    https://actu.epfl.ch/news/redefining-energy-efficiency-in-data-processing/

    By setting the conductivity of each transistor, we can perform analog vector-matrix multiplication in a single step by applying voltages to our processor and measuring the output

    • weew@lemmy.ca
      4·
      8 months ago

      Still, i don’t think it’ll need to get much more complex to be very useful for AI workloads.

      People have been discovering that more, and simpler, calculations seem to work better? the trend in AI workloads seems to have gone from FP32 -> FP16 -> INT16 -> INT8 and possibly even INT4?

      Seems like just having lots of simple calculations is more efficient/effective than more complex stuff.

      • trolololol@lemmy.world
        1·
        7 months ago

        Well these chips perform analog math, which means high precision high speed. It’s not as accurate as fp32 as in repeatedly and deterministic outputs, but that’s def not a problem for a deep and wide neural network such as used by llm