Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

LLM: fix mpt load_low_bit issue #10075

Merged
merged 3 commits into from
Feb 5, 2024
Merged

Conversation

JinBridger
Copy link
Contributor

@JinBridger JinBridger commented Feb 2, 2024

https://github.com/analytics-zoo/nano/issues/948

Enable tie_word_embeddings when save mpt model. Otherwise, mpt that saved with save_low_bit cannot run normally.

This may because MPT model requires tie_word_embeddings=True while we manually save this variable as False when saving low bit models.

Copy link
Contributor

@leonardozcm leonardozcm left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM, the last layer of mpt is a customized normal layer instead of linear which is usually seen in other llm models.

@Oscilloscope98 Oscilloscope98 merged commit a9da1a5 into intel:main Feb 5, 2024
19 checks passed
Jasonzzt pushed a commit to Jasonzzt/BigDL that referenced this pull request Feb 19, 2024
liu-shaojun pushed a commit that referenced this pull request Mar 25, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants