https://github.com/xrsrke/fsdl-megatron

Code for FSDL Breaking down parallelism in Megatron-LM
https://github.com/xrsrke/fsdl-megatron

Last synced: 3 months ago
JSON representation

Code for FSDL Breaking down parallelism in Megatron-LM

• Host: GitHub
• URL: https://github.com/xrsrke/fsdl-megatron
• Owner: xrsrke
• License: mit
• Created: 2023-06-18T04:03:09.000Z (almost 2 years ago)
• Default Branch: main
• Last Pushed: 2023-06-24T20:34:25.000Z (almost 2 years ago)
• Last Synced: 2025-01-21T15:26:21.389Z (4 months ago)
• Language: Python
• Size: 13.7 KB
• Stars: 1
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# fsdl-megatron

Code for FSDL Breaking down parallelism in Megatron-LM