transformers
Fix BLT training_ci overfit test
#42685
Merged

Fix BLT training_ci overfit test #42685

preetam1407
preetam1407 Fix BLT training_ci overfit test by disabling cache and adjusting tra…
9d35997b
preetam1407 Fix BLT training_ci overfit test by disabling cache and adjusting tra…
23da2e1b
preetam1407
preetam1407 Fix BLT training_ci overfit test by disabling cache and adjusting tra…
624e22ca
preetam1407 Format BLT tests with ruff
b4504b9a
3outeille
3outeille Merge branch 'main' into fix-blt-training-ci
5902dec7
github-actions
3outeille
HuggingFaceDocBuilderDev
github-actions
itazap
preetam1407 Fix BLT training CI with custom weight initialization and overfit test
832581d9
preetam1407 Fix BLT training CI with custom weight initialization and overfit test
9feb586f
preetam1407 Fix BLT training CI with custom weight initialization and overfit test
00d18978
preetam1407 Fix BLT training CI with custom weight initialization and overfit test
3e5700e4
preetam1407 Fix BLT training CI with custom weight initialization and overfit test
495094c7
preetam1407 Fix BLT training CI with custom weight initialization and overfit test
a7ce3b75
preetam1407 Update BLT init logic and adjust repo checks for non-functional model…
bd279d9a
preetam1407 Fix repo/config checks by marking BLT Text/Vision models as placeholders
4e64382f
preetam1407
preetam1407 Fix repo/config checks by marking BLT Text/Vision models as placeholders
9803753f
preetam1407 Fix repo/config checks by marking BLT Text/Vision models as placeholders
884ff6be
3outeille
3outeille Merge branch 'main' into fix-blt-training-ci
1414e702
3outeille
3outeille commented on 2025-12-11
preetam1407 Document BLT weight initialization sources and restore default overfi…
e60b3a3c
ArthurZucker
ArthurZucker commented on 2025-12-11
preetam1407 Align BLT weight init with nn.init
6c53915a
preetam1407 preetam1407 requested a review from 3outeille 3outeille 9 days ago
3outeille
3outeille Merge branch 'main' into fix-blt-training-ci
4bd8a318
preetam1407 Fix BLT init weights and remove modular conversion issues
96419557
preetam1407
preetam1407 fixes circle ci failures
36a0df47
preetam1407
3outeille Merge branch 'main' into fix-blt-training-ci
329edc26
3outeille
github-actions
3outeille
preetam1407
3outeille
3outeille requested changes on 2025-12-15
3outeille Merge branch 'main' into fix-blt-training-ci
12cdda6a
3outeille Merge branch 'main' into fix-blt-training-ci
82fabeb8
preetam1407 fix
4579e9fe
github-actions
preetam1407 fix
16e95243
3outeille Merge branch 'main' into fix-blt-training-ci
af6fb32f
3outeille
3outeille approved these changes on 2025-12-15
3outeille Merge branch 'main' into fix-blt-training-ci
9092dadd
3outeille
preetam1407 fix recurrent_gemma overfit generation with cache
a2202c59
preetam1407 Fix recurrent_gemma overfit generation with cache
c6442fe3
preetam1407 rerun circleci
651d69d8
preetam1407 rerun circleci
cb3fb177
preetam1407
3outeille
3outeille commented on 2025-12-15
preetam1407 Log RecurrentGemma cache exception in training mixin
075ba96d
preetam1407 ci: rerun
9f55274b
preetam1407 ci: rerun
fbbf0624
preetam1407 Merge branch 'main' into fix-blt-training-ci
38302030
preetam1407
preetam1407 Merge branch 'main' into fix-blt-training-ci
a2047591
preetam1407 ci: rerun
521298ae
github-actions
preetam1407 Merge branch 'main' into fix-blt-training-ci
2445f342
preetam1407
3outeille 3outeille enabled auto-merge (squash) 5 days ago
3outeille 3outeille merged 0f97c688 into main 5 days ago
3outeille

Login to write a write a comment.

Login via GitHub

Assignees
No one assigned
Labels
Milestone