Skip to content
Navigation Menu
Sign in
Appearance settings
Platform
AI CODE CREATION
GitHub Copilot
Write better code with AI
GitHub Copilot app
Direct agents from issue to merge
MCP Registry
Integrate external tools
DEVELOPER WORKFLOWS
Actions
Automate any workflow
Codespaces
Instant dev environments
Issues
Plan and track work
Code Review
Manage code changes
Code Quality
Enforce quality at merge
APPLICATION SECURITY
GitHub Advanced Security
Find and fix vulnerabilities
Code security
Secure your code as you build
Secret protection
Stop leaks before they start
EXPLORE
Why GitHub
Documentation
Blog
Changelog
Marketplace
View all features
Solutions
BY COMPANY SIZE
Enterprises
Small and medium teams
Startups
Nonprofits
BY USE CASE
App Modernization
DevSecOps
DevOps
CI/CD
View all use cases
BY INDUSTRY
Healthcare
Financial services
Manufacturing
Government
View all industries
View all solutions
Resources
EXPLORE BY TOPIC
AI
Software Development
DevOps
Security
View all topics
EXPLORE BY TYPE
Customer stories
Events & webinars
Ebooks & reports
Business insights
GitHub Skills
SUPPORT & SERVICES
Documentation
Customer support
Community forum
Trust center
Partners
View all resources
Open Source
COMMUNITY
GitHub Sponsors
Fund open source developers
PROGRAMS
Security Lab
Maintainer Community
Accelerator
GitHub Stars
Archive Program
REPOSITORIES
Topics
Trending
Collections
Enterprise
ENTERPRISE SOLUTIONS
Enterprise platform
AI-powered developer platform
AVAILABLE ADD-ONS
GitHub Advanced Security
Enterprise-grade security features
Copilot for Business
Enterprise-grade AI features
Premium Support
Enterprise-grade 24/7 support
Pricing
Search
/
Sign in
Sign up
Appearance settings
You signed in with another tab or window.
Reload
to refresh your session.
You signed out in another tab or window.
Reload
to refresh your session.
You switched accounts on another tab or window.
Reload
to refresh your session.
Dismiss alert
{{ message }}
Uh oh!
There was an error while loading.
Please reload this page
.
NVIDIA
/
Megatron-LM
Public
Notifications
You must be signed in to change notification settings
Fork
4.4k
Star
17.6k
Code
Issues
392
Pull requests
804
Discussions
Actions
Projects
Security and quality
0
Insights
Additional navigation options
Code
Issues
Pull requests
Discussions
Actions
Projects
Security and quality
Insights
Implementation of a more flexible optimizer/scheduler override system
- #2723
#2723
Merged
ko3n1g
merged 17 commits into
NVIDIA:main
NVIDIA/Megatron-LM:main
from
jstjohn:optimizer_param_lookup_refactor
jstjohn/Megatron-LM:optimizer_param_lookup_refactor
Copy head branch name to clipboard
Jan 7, 2026
Conversation
Commits
17
(17)
Checks
Files changed
Merged
Implementation of a more flexible optimizer/scheduler override system
#2723
ko3n1g
merged 17 commits into
NVIDIA:main
NVIDIA/Megatron-LM:main
from
jstjohn:optimizer_param_lookup_refactor
jstjohn/Megatron-LM:optimizer_param_lookup_refactor
Copy head branch name to clipboard
Commits
Commits on Dec 19, 2025
Implementation of a more flexible optimizer/scheduler override system
Show description for fc2fc3d
jstjohn
committed
fc2fc3d
View commit details
Copy full SHA for fc2fc3d
Browse repository at this point
Add more comments to get param groups since the behavior is different (no longer 1-1 constraint)
Show description for 9e81f47
jstjohn
committed
9e81f47
View commit details
Copy full SHA for 9e81f47
Browse repository at this point
Fix circular import in optimizer_param_scheduler
Show description for 856e395
jstjohn
committed
856e395
View commit details
Copy full SHA for 856e395
Browse repository at this point
Add test coverage
Show description for 6c816c6
jstjohn
committed
6c816c6
View commit details
Copy full SHA for 6c816c6
Browse repository at this point
Adding test coverage for ParamKey and ParamPredicte matching
Show description for 253b9f7
jstjohn
committed
253b9f7
View commit details
Copy full SHA for 253b9f7
Browse repository at this point
Also add clarifying test around what happens if an attribute is there but False
Show description for 8c9725d
jstjohn
committed
8c9725d
View commit details
Copy full SHA for 8c9725d
Browse repository at this point
Remove unused import and reformatting
Show description for 640ee26
jstjohn
committed
640ee26
View commit details
Copy full SHA for 640ee26
Browse repository at this point
Add/update more comments
Show description for dbaafc5
jstjohn
committed
dbaafc5
View commit details
Copy full SHA for dbaafc5
Browse repository at this point
Show that ordering does not impact the merging of the param group keys
Show description for d16efe4
jstjohn
committed
d16efe4
View commit details
Copy full SHA for d16efe4
Browse repository at this point
Commits on Jan 5, 2026
Add utility function for returning a standard set of config overrides for the user to make use of
Show description for 0faf0ea
jstjohn
committed
0faf0ea
View commit details
Copy full SHA for 0faf0ea
Browse repository at this point
Have default behavior for users of mcore who do not provide any skips
Show description for 6ae2c41
jstjohn
committed
6ae2c41
View commit details
Copy full SHA for 6ae2c41
Browse repository at this point
Adding missing license headers
Show description for 7d1ecc3
jstjohn
committed
7d1ecc3
View commit details
Copy full SHA for 7d1ecc3
Browse repository at this point
Update test to cover new default
Show description for 59b9023
jstjohn
committed
59b9023
View commit details
Copy full SHA for 59b9023
Browse repository at this point
Fix issue for fp8 cards that do not support block scaling
Show description for 8ca2cf4
jstjohn
committed
8ca2cf4
View commit details
Copy full SHA for 8ca2cf4
Browse repository at this point
Commits on Jan 6, 2026
Fix issue with consistency checking now that the overrides are partial
Show description for 7a5a49a
jstjohn
committed
7a5a49a
View commit details
Copy full SHA for 7a5a49a
Browse repository at this point
Add test coverage for failing consistency check in CI
Show description for 361db5a
jstjohn
committed
361db5a
View commit details
Copy full SHA for 361db5a
Browse repository at this point
Merge branch 'main' into optimizer_param_lookup_refactor
jstjohn
authored
ac4470e
View commit details
Copy full SHA for ac4470e
Browse repository at this point
You can’t perform that action at this time.