-
Notifications
You must be signed in to change notification settings - Fork 156
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
Train M -> F pronoun interventions on selected Pythia models #52
Comments
I have started working on this, I will start with 19M first and then move onto the bigger models! |
Thanks for being willing to help on this @ankit-bhattarai ! We've now done these items, please let me know if you need any assistance and want to experiment with different interventions. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
(Re)train pythia models with the last 7% of training data adjusted to have all female pronouns.
s3:https://s-eai-neox/pythia/1.3B_dedup/global_step66500
trained for last 5k steps with intervened datas3:https://s-eai-neox/pythia/19M_deduped/global_step133000
trained for last 10k steps with intervened datas3:https://s-eai-neox/pythia/350M_dedup/global_step66500
trained for last 5k steps with intervened datas3:https://s-eai-neox/pythia/6.7B_deduped_new/global_step133000
trained for last 10k steps with intervened dataAll intervened models should be evaluated on the same benchmarks as #16 for all the saved checkpoints post-intervention. All saved intervened checkpoints should also be evaluated on the same benchmarks as chosen in #51 .
If get meaningful numbers from the above and have evaluated all the above models:
s3:https://s-eai-neox/pythia/1.3B_dedup/global_step66500
trained for last 10k steps with intervened dataThe text was updated successfully, but these errors were encountered: