-
Notifications
You must be signed in to change notification settings - Fork 4.7k
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
Feature: Basic distilling. #6527
base: main
Are you sure you want to change the base?
Conversation
else: | ||
self.processing_class: "PreTrainedTokenizer" = kwargs.get("tokenizer") | ||
|
||
self.teacher_model = teacher_model |
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.
Does it work on DDP setting?
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.
I'm working on deepspeed
. Default DDP is working.
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.
Anyway I will not gonna using GKDTrainer in trl>=11.0
because this can't use for mllm.
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.
It is working on deepspeed but require deepspeed==0.15.4
What does this PR do?
Fixes # (issue)
Before submitting