Skip to content

feat: Add on-policy distillation support #57

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.

Sign up for GitHub

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

Jump to bottom
Open
ApGa wants to merge 5 commits into
base: major-update
Choose a base branch
from
Open

feat: Add on-policy distillation support #57

ApGa wants to merge 5 commits into from

Conversation

@ApGa
Copy link
Collaborator

@ApGa ApGa commented Jan 3, 2026 •
edited
Loading

  1. Adds on-policy distillation trainer, addressing Distillation experiments #18.
  2. Fixes a masking issue in the generator (fix by @adityasoni9998).

Example plot of negative reverse KL between Qwen3-4B (student) and Qwen3-8B (teacher).

image

@ApGa ApGa changed the title [WIP] On-policy distillation support feat: Add on-policy distillation support Jan 3, 2026
@ApGa ApGa requested a review from lintangsutawika January 5, 2026 17:43
@ApGa ApGa marked this pull request as ready for review January 5, 2026 17:44
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@lintangsutawika lintangsutawika Awaiting requested review from lintangsutawika

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

3 participants

@ApGa @adityasoni9998