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
Add madlad-400 MT models #27471
Add madlad-400 MT models #27471
Conversation
Thanks for adding @jbochi! Let us know when you're ready for a review. I've set the failed tests to re-run as it seems there was a transient connection error causing them to fail. |
Thanks for re-running the tests, @amyeroberts . The PR is ready for review. I don't think the previous failure was related to my changes. |
The tests passed this time. 😌 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks for adding!
Just a few small notes on the doc page structure.
Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>
I think we can transfer all MT models now and merge this PR.
Maybe I will work on LM models later, but I am not sure.
What do you think?
…On Thu, Nov 16, 2023, 8:15 AM amyeroberts ***@***.***> wrote:
***@***.**** commented on this pull request.
------------------------------
In docs/source/en/model_doc/madlad-400.md
<#27471 (comment)>
:
> +- [jbochi/madlad400-3b-mt](https://huggingface.co/jbochi/madlad400-3b-mt)
+
+- [jbochi/madlad400-7b-mt](https://huggingface.co/jbochi/madlad400-7b-mt)
+
+- [jbochi/madlad400-7b-mt-bt](https://huggingface.co/jbochi/madlad400-7b-mt-bt)
+
+- [jbochi/madlad400-10b-mt](https://huggingface.co/jbochi/madlad400-10b-mt)
OK - let me know when it's ready. We'll do this as the last step before
merging
—
Reply to this email directly, view it on GitHub
<#27471 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AACHO2FLM2AV6ITIUQ63O23YEYGY5AVCNFSM6AAAAAA7JISLSCVHI2DSMVQWIX3LMV43YUDVNRWFEZLROVSXG5CSMV3GSZLXHMYTOMZUGM2DONBZGE>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Great - thanks for iterating!
Just one little nit on the abstract formatting. I can transfer the checkpoints, the checkpoint paths can be updated and then we can merge 🤗
Thank you for filling out the model cards so beautifully 🙏 |
Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>
Thanks for the review! |
Hi @jbochi. Apologies for the delay in getting back to you - I'm also learning about the intricacies of org weights too! For the model weights, would you prefer that we move the original checkpoint repos directly under google or duplicate, so that yours still exist under your profile? |
Hey. No worries. I think it's better to move them (assuming people will be
redirected if they go to mine).
…On Thu, Nov 23, 2023, 4:50 AM amyeroberts ***@***.***> wrote:
Hi @jbochi <https://github.com/jbochi>. Apologies for the delay in
getting back to you - I'm also learning about the intricacies of org
weights too! For the model weights, would you prefer that we move the
original checkpoint repos directly under google or duplicate, so that yours
still exist under your profile?
—
Reply to this email directly, view it on GitHub
<#27471 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AACHO2AT3SW7LP3R6SLGTVTYF4L5VAVCNFSM6AAAAAA7JISLSCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQMRUGA4DSNBTGA>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
@jbochi If they're moved then they wouldn't exist under your profile any more at all. I believe it would error-out saying that the repo doesn't exist if someone tried: |
In this case, can we copy them instead? Thanks
…On Fri, Nov 24, 2023, 12:45 PM amyeroberts ***@***.***> wrote:
@jbochi <https://github.com/jbochi> If they're moved then they wouldn't
exist under your profile any more at all. I believe it would error-out
saying that the repo doesn't exist if someone tried:
AutoTokenizer.from_pretrained("jbochi/madlad400-3b-mt"). Is this still OK
or would you rather we copy?
—
Reply to this email directly, view it on GitHub
<#27471 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AACHO2CTUUN2QO6SMH62K63YGDMNLAVCNFSM6AAAAAA7JISLSCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQMRVHE2TKOBUHA>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
@jbochi Yep! All done. I've set the doc tests to re-run which should pass now the google checkpoints exist 🤞 |
Yay! Thank you!
…On Mon, Nov 27, 2023, 11:03 AM amyeroberts ***@***.***> wrote:
@jbochi <https://github.com/jbochi> Yep! All done. I've set the doc tests
to re-run which should pass now the google checkpoints exist 🤞
—
Reply to this email directly, view it on GitHub
<#27471 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AACHO2ARZDBRUO4SS2AAVBDYGS2VHAVCNFSM6AAAAAA7JISLSCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQMRYGEZTGMJTGU>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
@jbochi One last final update - the doctests are failing with a timeout. Could you add the model doc page |
Tests are timing out.
Sure. Done! |
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. |
@jbochi Awesome - thanks again for adding! |
Thanks for the review and all the help, @amyeroberts |
What does this PR do?
Fixes #26696
Before submitting
Pull Request section?
to it if that's the case.
documentation guidelines, and
here are tips on formatting docstrings.
Who can review?
Anyone in the community is free to review the PR once the tests have passed. Feel free to tag
members/contributors who may be interested in your PR.
@ArthurZucker