Skip to content

Add extra inference call after warmup for issue #173 #174

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
tinhto-000 wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Add extra inference call after warmup for issue #173 #174

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 12 additions & 0 deletions aiu_fms_testing_utils/scripts/drive_paged_programs.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -421,6 +421,18 @@ def __load_validation_info(
**extra_kwargs,
)

# do an extra inference call to workaround the issue on z/OS where the first inference
# result is always incorrect during multi-AIU (issue 173)
extract_validation_information(
model,
input_ids,
max_new_tokens,
post_iteration_hook=None,
last_n_tokens=64,
prefill_chunk_size=args.prefill_chunk_size,
**extra_kwargs,
)

if USE_DISTRIBUTED:
# wait for rank0 to be finished as it is the only one generating the criteria json
# this is needed since otherwise we may run into a race condition
Expand Down