Skip to content

Conversation

@molereddy
Copy link
Collaborator

What does this PR do?

documentation stuff

  • update WGA, SatImp README mentions
  • Link to OpenUnlearing paper's meta-eval modelson huggingface

Before submitting

  • This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
  • Have you gone through the documentation of adding new components?
  • Did you make sure to update the documentation with your changes? Here are the pointers to documentation
    documentation guidelines.

molereddy and others added 24 commits March 1, 2025 09:13
* testing commit

* Fixes

* cleanup
Fix tofu_unlearn.sh for IdKDPO method.
* IdkDPO script fix in tofu_unlearn.sh (locuslab#65)

* Fix hyperlinks in README
* Download I don't know data in setup_data.py
* Fix tofu_unlearn.sh for IdkDPO

---------

Co-authored-by: Anmol Mekala <49127549+molereddy@users.noreply.github.com>

* overwrite=True

* RMU added

* Fix ref model device

* ruff fix

* RMU updated

* Update rmu.py

* Update README.md: add RMU

* Added references and renamed functions

---------

Co-authored-by: Anmol Mekala <49127549+molereddy@users.noreply.github.com>
…on (#8)

* docs: updates, small corrections, re-formats

* modified ruff commands

* modified ruff commands

* CI/CD minor updates

* added contributing + leaderboard

* fix minor spelling misatkes

* docs: bunch of minor updates

* docs fixes

---------

Co-authored-by: molereddy <m.anmolreddy@gmail.com>
* Re-formatting + more badges

* Update and fix docs

* Make error msg accurate

* handle lack of flash-attn flag better

* Document more hydra features

* update example exp configs to match latest supported metrics

* Change HF logo

* Simplify eval exp cfg dump

* testing push workflows

* Add workflow test branch

* update workflow path again

* Reformat badges to fix blue line issue

* Fix div

* revert change to tests build path
* documentation fix

* remove eos only after removing pad tokens + not use model train inside evaluation

* Fix date to handle Llama3.1 repro issues due to tokenizer automatically adding curr date

* ruff fixes

* minor mistake

* warn about and handle weird tokenization cases for small targets

* Ruff fixes

* The assert must hold by definition

* Updating leaderboard.md numbers

* Allow for invalid evaluations which are excluded from averaging

* bug fix

* ruff fixes

---------

Co-authored-by: Dornavineeth <vineethdorna@gmail.com>
* Update README.md with arxiv report

---------

Co-authored-by: Vineeth <48151992+Dornavineeth@users.noreply.github.com>
* feat: add CE-U loss

* chore: move CE-U related functions to `ceu.py`

* Ruff format

* remove runner

* Add CE-U link

* Add CE-U runner

* Link OU model collection

* Update verbiage

---------

Co-authored-by: Bo Yang <bo@tacnode.io>
@molereddy molereddy merged commit d33c476 into locuslab:main Jun 30, 2025
1 check passed
@molereddy molereddy mentioned this pull request Jun 30, 2025
3 tasks
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.

2 participants