I've figured out what pisses me off so much about Facebook's Galactica demo.
It's not because people can use to to write bad essays for their homework. There are plenty of large language models that can do that. It's because Facebook is presenting it as something that it most definitely is not.
Facebook is selling it as a knowledge engine, a "new interface to access and manipulate what we know about the universe."
Actually it's just a random bullshit generator.
I won’t be going to #sfn22 this year, but my work will! Please come by board AA.12 on Tuesday morning around 9 am to learn about the anatomical organization of cell types and inhibitory connectivity across visual cortex. Long story short, layer 2 is really different than layer 3, different excitatory neurons have quite different synapse properties, layer 5 has oodles of very specific inhibition, and we found what seems to be a new class of disinhibition specialist interneurons.
Here is my #introduction. I am a biologist interested in #neuroscience and the #evolution of cells and nervous systems. My research group studies #ciliated marine larvae from a whole-organism perspective, combining behaviour, #connectomics, #imaging, #CRISPR, #modelling and other approaches. We love marine #larvae, their #neurons, #cilia, and #neuropeptides.
Time for an #introduction. Nervous systems — yours, mine, those of mice, fish, and insects and worms — are made up of populations of different kinds of neurons that communicate with one another. Current electron microscopy generates millimeter-scale volumes with the morphology and synaptic connectivity of up to 10^5 neurons. As a scientist at the Allen Institute, I am mapping out the cell types and connectivity rules found in cortex and building computational tools to help others do the same.
EM connectomics at the Allen Institute for Brain Science. Figuring out the structure of the brain, one synapse at a time.