Skip to content
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

OpinionsQA #1424

Merged
merged 17 commits into from
Mar 30, 2023
Merged

OpinionsQA #1424

merged 17 commits into from
Mar 30, 2023

Conversation

ShibaniSanturkar
Copy link
Collaborator

Merge the OpinionsQA dataset into HELM.

percyliang
percyliang reviewed Mar 20, 2023
View reviewed changes
src/helm/benchmark/adaptation/adapter_spec.py Outdated
@@ -68,6 +68,9 @@ class AdapterSpec:
# set of training instances. Used to compute error bars.
num_train_trials: int = 1

# Sample train examples or use deterministic
Copy link
Contributor

Choose a reason for hiding this comment

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

Make the comment clearer: Sample train examples (as opposed to taking the first few in order)

Copy link
Contributor

Choose a reason for hiding this comment

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

But just curious why the sample_train = False mode is needed? In general, it'd be nice not to add more things into AdapterSpec unless they are necessary.

Copy link
Collaborator Author

Choose a reason for hiding this comment

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

Here, we are providing models bios in context for steerability. I disabled sampling so that I can ensure that we run through all the bios that are in-context examples, rather than just a randomly chosen subset.

Copy link
Collaborator Author

Choose a reason for hiding this comment

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

Sampling is needed so that we can go through all the training examples (which are our steering groups) rather than sampling a random subset of them

percyliang
percyliang reviewed Mar 26, 2023
View reviewed changes
src/helm/benchmark/run_specs.py
survey_type: str,
num_logprobs: str,
context: str = "None",
num_train_trials: str = "1",
Copy link
Contributor

Choose a reason for hiding this comment

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

I think you can get rid of this, because we have a RunExpander that allows you to set num_train_trials

percyliang
percyliang requested changes Mar 26, 2023
View reviewed changes
Copy link
Contributor

@percyliang percyliang left a comment

Choose a reason for hiding this comment

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

Thanks for the changes - I think this is almost ready - just a few more localized cleanups and documentation, so hopefully it won't be much additional work.

percyliang
percyliang approved these changes Mar 28, 2023
View reviewed changes
Copy link
Contributor

@percyliang percyliang left a comment

Choose a reason for hiding this comment

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

Great, thanks! (Left one more comment about the arguments of the run spec.)

@ShibaniSanturkar ShibaniSanturkar merged commit b344db1 into main Mar 30, 2023
@ShibaniSanturkar ShibaniSanturkar deleted the opinions-qa branch March 30, 2023 15:35
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@percyliang percyliang percyliang approved these changes

@teetone teetone Awaiting requested review from teetone

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@ShibaniSanturkar @percyliang