transformers
[gemma4] Dissociate kv states sharing from the Cache
#45312
Merged
Overview
Commits
13
Changes
View On GitHub

[gemma4] Dissociate kv states sharing from the Cache #45312

Cyrilvallez merged 13 commits into main from no-cache-gemma4
Cyrilvallez
Cyrilvallez force at least cache sharing always
42aa69f8
Cyrilvallez Cyrilvallez changed the title Fix Gemma4 cache sharing with no_cache Fix Gemma4 layer cache sharing with no_cache 217 days ago
Cyrilvallez Cyrilvallez changed the title Fix Gemma4 layer cache sharing with no_cache Fix Gemma4 layer cache sharing with use_cache=False 217 days ago
HuggingFaceDocBuilderDev
zucchini-nlp
zucchini-nlp approved these changes on 2026-04-08
Cyrilvallez Merge branch 'main' into no-cache-gemma4
47bd0a2e
Cyrilvallez better
beb3d15a
Cyrilvallez adjust tests gradients
2ab838e3
Cyrilvallez Merge branch 'main' into no-cache-gemma4
36666b7c
Cyrilvallez fix mask
b124525d
Cyrilvallez fix
00dc23e2
Cyrilvallez fix
f9d70a25
Cyrilvallez simplify a lot
fbe52163
Cyrilvallez add slow test
a7af04fb
github-actions
Cyrilvallez fix
de5d0d15
Cyrilvallez doc
9c64da7c
Cyrilvallez improve type hints
0a833faf
Cyrilvallez Cyrilvallez changed the title Fix Gemma4 layer cache sharing with use_cache=False [gemma4] Dissociate kv states sharing from the Cache 217 days ago
Cyrilvallez Cyrilvallez merged d42f8ba5 into main 216 days ago
Cyrilvallez Cyrilvallez deleted the no-cache-gemma4 branch 216 days ago

Login to write a write a comment.

Login via GitHub

Reviewers
Assignees
No one assigned
Labels
None yet
Milestone
No milestone
Terms Of Use Privacy Policy Refund Policy Impressum