-
Notifications
You must be signed in to change notification settings - Fork 3k
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]support qlora pp #7801
[llm]support qlora pp #7801
Conversation
Thanks for your contribution! |
Codecov ReportAttention:
Additional details and impacted files@@ Coverage Diff @@
## develop #7801 +/- ##
===========================================
- Coverage 57.30% 57.11% -0.19%
===========================================
Files 584 587 +3
Lines 87688 88194 +506
===========================================
+ Hits 50252 50376 +124
- Misses 37436 37818 +382 ☔ View full report in Codecov by Sentry. |
@@ -2193,6 +2193,10 @@ def from_pretrained(cls, pretrained_model_name_or_path, *args, **kwargs): | |||
quantization_config=config.quantization_config, | |||
llm_int8_threshold=config.quantization_config.llm_int8_threshold, | |||
) | |||
quantization_linear_list = [] | |||
for key in model.state_dict().keys(): | |||
if "quant_weight" in key: |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
pp会改变state_dict的命名,改用这个方式获取quantization_linear_list
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
* supprt qlora pp * fix scale dtype
PR types
Others
PR changes
APIs
Description
支持qlora pp