transformers
Llama et al. / FSDP : Fix breaking change in 4.40 for FSDP
#31161
Merged

Llama et al. / FSDP : Fix breaking change in 4.40 for FSDP #31161

amyeroberts merged 14 commits into main from fix-llama-fsdp
younesbelkada
younesbelkada fix llama fsdp
a2ba7053
younesbelkada younesbelkada requested a review from amyeroberts amyeroberts 2 years ago
younesbelkada fixup
51881061
younesbelkada younesbelkada requested a review from LysandreJik LysandreJik 2 years ago
HuggingFaceDocBuilderDev
amyeroberts
amyeroberts commented on 2024-06-03
younesbelkada
amyeroberts
younesbelkada Merge remote-tracking branch 'origin/main' into HEAD
51e05137
younesbelkada adding FSDP tests for CPU offloading
91c7b0d3
younesbelkada fixes
d04ae5bb
younesbelkada fix tests
ac7977cd
younesbelkada fix tests
8e65359e
younesbelkada add it for mixtral
705afe43
younesbelkada propagate the changes on other models
3ce51785
younesbelkada younesbelkada requested a review from amyeroberts amyeroberts 2 years ago
amyeroberts
amyeroberts Merge branch 'main' into fix-llama-fsdp
bb4c47a4
amyeroberts
amyeroberts commented on 2024-06-26
amyeroberts Update src/transformers/models/phi/modeling_phi.py
bd3fe928
amyeroberts Delete utils/testing_scripts/fsdp_cpu_offloading.py
35981b67
amyeroberts Delete utils/testing_scripts/dummy_fsdp_config.yml
b5e31457
amyeroberts
amyeroberts commented on 2024-06-26
amyeroberts
amyeroberts commented on 2024-06-26
amyeroberts
amyeroberts commented on 2024-06-26
amyeroberts
amyeroberts commented on 2024-06-26
amyeroberts
amyeroberts commented on 2024-06-26
amyeroberts Update + add cache_positions docstring
4c9831ef
amyeroberts
amyeroberts approved these changes on 2024-06-26
amyeroberts amyeroberts merged 3f93fd06 into main 1 year ago
amyeroberts amyeroberts deleted the fix-llama-fsdp branch 1 year ago

Login to write a write a comment.

Login via GitHub

Assignees
No one assigned
Labels
Milestone