LoreonLabsPlatform

Overview

Intelligence

Markets
Builders
Research
Narratives
Ecosystems
Launchpads

Discover

Search
Sources

Other

nebullvm

Plug and play modules to optimize the performances of your AI systems 🚀

OtherEmerging

Stars

—

Forks

—

Contributors

8

Last push

39mo ago

Recent commits

Latest commits.

Merge pull request #316 from EikeKohl/fix_deepspeed_multi_gpu
87c6d13Pierpaolo Sorbellini39mo ago
Disable conversation log saving for now to enable deepspeed multi gpu training
007194bEike Kohlmeyer39mo ago
Enable deepspeed distribution for reward model
cc7ebc1Eike Kohlmeyer39mo ago
Move device allocation to learn method to avoid CUDA Allocation Error using deepspeed
179444eEike Kohlmeyer39mo ago
Change device to local rank if deepspeed is enabled
c9e4b9fEike Kohlmeyer39mo ago

fix engine access in train()

2a6f20cEike Kohlmeyer39mo ago

Add local rank argument to main.py

21d891dEike Kohlmeyer39mo ago

Do not use unsupported lr_scheduler if deepspeed is enabled

7935d1eEike Kohlmeyer39mo ago

Top contributors

Builders behind this project.

PierpaoloSorbellini