Stories by pella

Training LLMs at Scale with AMD MI250 GPUs

Large language models propagate race-based medicine

Standardize Next-Generation Narrow Precision Data Formats for AI

Data Parallel C++, Second Edition, Open Access Book

Placebo effect shapes how we see AI

LLMs confabulate not hallucinate

The Story Behind the Activity Book for Postgres

The winners of the 2023 Ig Nobel Prizes

PostgreSQL 16

New mothers may have enhanced ability to see faces in objects

Medicine? Study finds laughter is good for heart health

AMD Inception Mitigations Can Lower Performance of CPUs by Up to 54%

Benchmarking the Performance Impact to AMD Inception Mitigations

My Encounter with Philosophical Anarchist Paul Feyerabend

Spartans Were Losers

Can You Run a State-of-the-Art LLM On-Prem for a Reasonable Cost?

AMD EPYC 9684X Genoa-X Provides HPC Performance

PostgreSQL: No More Vacuum, No More Bloat

NeurIPS Large Language Model Efficiency Challenge: 1 LLM and 1GPU and 1Day

GCC Lands AVX-512 Fully-Masked Vectorization

1
2
3
4
5
6
7
8
9
10