[NO-TICKET] Add fused_rms_norm config usage in Mistral moreh #43
Merged
codernine-moreh merged 2 commits intov4.42.4-morehfrom Oct 31, 2025
Merged
[NO-TICKET] Add fused_rms_norm config usage in Mistral moreh #43codernine-moreh merged 2 commits intov4.42.4-morehfrom
fused_rms_norm config usage in Mistral moreh #43codernine-moreh merged 2 commits intov4.42.4-morehfrom
Conversation
|
|
||
| if MorehRMSNorm is not None: | ||
| logger.warning( | ||
| "You can't use Masked Structured Growth Training..! You should avoid using rmsnorm in any way. " |
There was a problem hiding this comment.
Masked Structured Growth Training
이게 뭐에요..??
rmsnorm -> RMSNorm 🙏
Author
There was a problem hiding this comment.
남형님 modeling_llama 패치를 그대로 이식한거라 저도 잘...
G선생님의 답변 :
Masked Structural Growth (MSG) Training이란?
**Masked Structural Growth (MSG)**는 대형 언어 모델(Transformer)을 점진적으로 키워가며 학습하는 기법이에요.
즉, 처음부터 거대한 모델을 학습하지 않고, 작은 모델로 시작해서 점점 구조를 확장(레이어 수, hidden size, attention head 수 등 증가)해 나가는 방식입니다.
hyeongjun-jeon
approved these changes
Oct 31, 2025
dongjin-na
approved these changes
Oct 31, 2025
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description