PR #24224 Fix LLaMa beam search when using parallelize

Fix LLaMa beam search when using parallelize #24224

asartran merged 3 commits into huggingface:main from patch-1

Fix LLaMa beam search when using parallelize

7edc640e

fix code format in modeling_llama.py

5563a48f

asartran approved these changes on 2023-06-14

fix format of _reorder_cache in modeling_llama.py

a44abefc

asartran merged 33196b45 into main 3 years ago

Reviewers

asartran

Assignees

No one assigned

Labels

None yet

Milestone

No milestone