Hacker News new | ask | show | jobs
user: perone
created: 2009-10-27
karma: 1657

submissions:

Gemma3n architecture: a short guide [slides]
2 points | 0 comments
Diffusion Elites: surprisingly good, simple and embarrassingly parallel
9 points | 0 comments
0 points | 0 comments
0 points | 0 comments
0 points | 0 comments
0 points | 0 comments
0 points | 0 comments
0 points | 0 comments
0 points | 0 comments
0 points | 0 comments
0 points | 0 comments
Show HN: VectorVFS, your filesystem as a vector database
279 points | 138 comments
Notes on Gilbert Simondon's "On the Mode of Existence of Technical Objects"
2 points | 0 comments
0 points | 0 comments
The geometry of data: the missing metric tensor and the Stein score [Part II]
64 points | 7 comments
Memory-Mapped CPU Tensor Between Torch, NumPy, Jax and TensorFlow
1 points | 0 comments
0 points | 0 comments
Large language model data pipelines and Common Crawl (WARC/WAT/WET) formats
2 points | 0 comments
Appreciating the complexity of LLMs data pipelines
1 points | 0 comments
0 points | 0 comments
PyTorch 2 Internals
4 points | 0 comments
PyTorch 2 Internals [slides]
2 points | 1 comments
0 points | 0 comments
0 points | 0 comments
0 points | 0 comments
Appreciating the complexity of large language models data pipelines
1 points | 0 comments
Appreciating the complexity of large language models data pipelines
2 points | 0 comments
Show HN: Feste, an open-source framework to optimize and parallelize NLP tasks
2 points | 0 comments
Tutorial using LLVM to JIT PyTorch graphs to native code (x86/arm/RISC-V/WASM)
2 points | 0 comments