Skip to content

@saforem2 saforem2 Follow

More

@saforem2

saforem2

Follow

View saforem2's full-sized avatar

:octocat:

making rocks think

Sam Foreman saforem2

:octocat:

making rocks think

Follow

training large models for science @argonne-lcf

483 followers · 781 following

@argonne-lcf
Chicago, IL
03:01 (UTC -05:00)
https://samf.sh
https://orcid.org/0000-0002-9981-0876
https://alcf.anl.gov/about/people/sam-foreman
@saforem2
https://scholar.google.com/citations?user=vV_1zDwAAAAJ&hl=en
https://sf.omg.lol

Achievements

Achievement: Pair Extraordinairex2 Achievement: Pull Sharkx3 Achievement: Quickdraw Achievement: YOLO Achievement: Starstruck Achievement: Arctic Code Vault Contributor

Achievements

Achievement: Pair Extraordinairex2 Achievement: Pull Sharkx3 Achievement: Quickdraw Achievement: YOLO Achievement: Starstruck Achievement: Arctic Code Vault Contributor

Highlights

Developer Program Member

Organizations

More

saforem2 /README.md

Sam Foreman

👋 Hi, I'm Sam.

I'm interested generally on large scale distributed training and AI for science¹.

If you're really curious, there's more on my 🏡 website or follow me on twitter !

About Me:
- samf.sh
- Models / Pre-Training Lead in the AI/ML Group @ Argonne National Laboratory (ALCF)
- Interested in:
  - {AI, HPC} for science
  - scaling large models across thousands of GPUs
- 🎶 Now Playing:

Profile Stats

Wakatime

wakatime:

🐍

TakaTime

TakaTime Weekly Report

Time Stats
Languages Projects
Languages Projects
Heatmap Tech Stack

Generated automatically by TakaTime

Footnotes

Mostly trying to get supercomputers to talk to each other. ↩

Pinned Loading

l2hmc-qcd l2hmc-qcd Public

Application of the L2HMC algorithm to simulations in lattice QCD.

Jupyter Notebook 69 9
ezpz ezpz Public

Write once, run anywhere; ezpz 🍋

Python 41 11
torchtitan torchtitan Public

Forked from pytorch/torchtitan

A native PyTorch Library for large model training

Python 4
samf.sh samf.sh Public

my personal website, built with Astro and WebTUI

MDX 1
deepspeedai/DeepSpeed deepspeedai/DeepSpeed Public

DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.

Python 43.1k 5k
argonne-lcf/Megatron-DeepSpeed argonne-lcf/Megatron-DeepSpeed Public

Forked from deepspeedai/Megatron-DeepSpeed

Ongoing research training transformer language models at scale, including: BERT & GPT-2

Python 17 20