Skip to content

Pull requests: TransformerLensOrg/TransformerLens

Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Assignee
Filter by who’s assigned
Sort

Pull requests list

Added OLMo(E) v1
#816 opened Dec 15, 2024 by jonasrohw Loading…
5 of 6 tasks
Set prepend_bos to false by default for Qwen models
#815 opened Dec 14, 2024 by degenfabian Loading…
6 of 7 tasks
Throw error when using attn_in with grouped query attention
#810 opened Dec 11, 2024 by degenfabian Loading…
7 tasks done
setup util base matrix class
#803 opened Nov 27, 2024 by bryce13950 Loading…
10 tasks
bumped python min version
#802 opened Nov 27, 2024 by bryce13950 Loading…
10 tasks
Add model upload and load
#779 opened Nov 13, 2024 by mntss Draft
10 tasks
improve model properties table in docs
#769 opened Oct 29, 2024 by mivanit Loading…
7 tasks done
Add py.typed for type hints
#760 opened Oct 18, 2024 by UFO-101 Loading…
7 tasks done
updated dependencies
#682 opened Jul 23, 2024 by bryce13950 Loading…
3 of 10 tasks
Model baichuan
#649 opened Jun 29, 2024 by bryce13950 Draft
4 of 9 tasks
Model config tests
#627 opened Jun 6, 2024 by curt-tigges Loading…
5 of 10 tasks
Add support for model_name-less models.
#603 opened May 19, 2024 by ArthurConmy Draft
10 tasks
Run GitHub CI on MacOS
#598 opened May 16, 2024 by bmillwood Loading…
7 tasks done
Mistral 7b v0.2
#587 opened May 11, 2024 by bryce13950 Loading…
10 tasks
revised demo testing to check all demos
#542 opened Apr 15, 2024 by bryce13950 Draft
1 of 10 tasks
[Draft] Support Flash Attention
#501 opened Jan 30, 2024 by cmathw Draft
6 of 7 tasks
Make tokenize_and_concatenate work with more datasets enhancement New feature or request
#473 opened Dec 28, 2023 by ArthurConmy Loading…
3 tasks
(Draft) Add DLA function to utils
#466 opened Dec 16, 2023 by VasilGeorgiev39 Loading…
3 of 10 tasks
ProTip! Add no:assignee to see everything that’s not assigned.