perone | Dark Hacker News

user:	perone
created:	October 27, 2009
karma:	1.7k

1.	Gemma3n architecture: a short guide [slides](drive.google.com) 2 points by perone 194 days ago \| 0 comments
2.	Diffusion Elites: surprisingly good, simple and embarrassingly parallel(blog.christianperone.com) 9 points by perone 315 days ago \| 0 comments
3.	Show HN: VectorVFS, your filesystem as a vector database(vectorvfs.readthedocs.io) 279 points by perone 1 year ago \| 138 comments
4.	Notes on Gilbert Simondon's "On the Mode of Existence of Technical Objects"(blog.christianperone.com) 2 points by perone 1 year ago \| 0 comments
5.	1 points by perone 1 year ago \| discuss
6.	The geometry of data: the missing metric tensor and the Stein score [Part II](blog.christianperone.com) 64 points by perone 1 year ago \| 7 comments
7.	Memory-Mapped CPU Tensor Between Torch, NumPy, Jax and TensorFlow(blog.christianperone.com) 1 points by perone 1 year ago \| 0 comments
8.	Large language model data pipelines and Common Crawl (WARC/WAT/WET) formats(blog.christianperone.com) 2 points by perone 2 years ago \| 0 comments
9.	Appreciating the complexity of LLMs data pipelines(blog.christianperone.com) 1 points by perone 2 years ago \| 0 comments
10.	PyTorch 2 Internals(slideshare.net) 4 points by perone 2 years ago \| 0 comments
11.	PyTorch 2 Internals [slides](drive.google.com) 2 points by perone 2 years ago \| 1 comment
12.	Appreciating the complexity of large language models data pipelines(blog.christianperone.com) 1 points by perone 2 years ago \| 0 comments
13.	Appreciating the complexity of large language models data pipelines(blog.christianperone.com) 2 points by perone 2 years ago \| 0 comments
14.	Show HN: Feste, an open-source framework to optimize and parallelize NLP tasks(feste.readthedocs.io) 2 points by perone 3 years ago \| 0 comments
15.	Tutorial using LLVM to JIT PyTorch graphs to native code (x86/arm/RISC-V/WASM)(blog.christianperone.com) 2 points by perone 3 years ago \| 0 comments