Skip to content

Conversation

@ruidazeng
Copy link
Contributor

Add WMDP cyber dataset configuration for unlearning

What does this PR do?

Fixes #80

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 contributions guide?
  • Are your changes documented? Read documentation guidelines here.

Add WMDP cyber dataset configuration for unlearning

### 📢 Updates

#### [Apr 9, 2025]
Copy link
Collaborator

Choose a reason for hiding this comment

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

need to update just before merge

Copy link

Choose a reason for hiding this comment

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

It's the same as WMDP_cyber_forget.yaml
Is it needed for something? If for testing performance disruption, IIRC MMLU was used.

Copy link
Collaborator

Choose a reason for hiding this comment

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

@ruidazeng I don't see the path existing for this dataset in hf.
cais/wmdp-corpora

Copy link

@filyp filyp Apr 29, 2025

Choose a reason for hiding this comment

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

You mean for MMLU? Yeah, it's not in cais/wmdp-corpora but in https://huggingface.co/datasets/cais/mmlu

From what I understand, in original setup, training is done on:

  • "cais/wmdp-corpora" - "cyber-forget-corpus"
  • "cais/wmdp-corpora" - "cyber-retain-corpus"

And evaluation on:

  • "cais/wmdp" - "wmdp-cyber"
  • "cais/mmlu" - "all"

Copy link

Choose a reason for hiding this comment

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

I see you mentioned in #80 that evaluation won't be done here anyway, but only afterwards, so I guess WMDP_cyber_forget.yaml and WMDP_cyber_retain.yaml just aren't needed.

Copy link
Collaborator

Choose a reason for hiding this comment

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

Yes

@filyp filyp mentioned this pull request Apr 26, 2025
5 tasks
@Dornavineeth Dornavineeth mentioned this pull request May 12, 2025
6 tasks
@Dornavineeth
Copy link
Collaborator

Hey @ruidazeng thanks for initiating this work. We now integrated WMDP and lm-eval-harness in #104 .

@ruidazeng
Copy link
Contributor Author

Hey @ruidazeng thanks for initiating this work. We now integrated WMDP and lm-eval-harness in #104 .

Fantastic love the progress!

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.

[Feature Request] Add WMDP Dataset

4 participants