Terra Incognita

Diffusion Elites: surprisingly good, simple and embarrassingly parallel

TorchStation Prototype V1 – GPUs panel

VectorVFS: your filesystem as a vector database

Notes on Gilbert Simondon’s “On the Mode of Existence of Technical Objects” and Artificial Intelligence

The geometry of data: the missing metric tensor and the Stein score [Part II]

Torch Titan distributed training code analysis

Memory-mapped CPU tensor between Torch, Numpy, Jax and TensorFlow

Generalisation, Kant’s schematism and Borges’ Funes el memorioso – Part I

PyTorch 2 Internals – Talk

Thoughts on Riemannian metrics and its connection with diffusion/score matching [Part I]

Large language model data pipelines and Common Crawl (WARC/WAT/WET)