Skip to content

Update opt_learning_rate_warmup_steps and opt_learning_rate_decay_steps constraint check for llama 3.1 8b and 405b model #435

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

Merged
merged 5 commits into from
Sep 23, 2025
Merged

Update opt_learning_rate_warmup_steps and opt_learning_rate_decay_steps constraint check for llama 3.1 8b and 405b model #435

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
688d2f0
remove constraint for opt_learning_rate_decay_steps
suachong Sep 5, 2025
f705313
Update rcp_checker.py to handle llama31_8b epochs correctly
suachong Sep 5, 2025
6b5151a
Merge remote-tracking branch 'mlcommons/master'
suachong Sep 11, 2025
a827a87
added constraint check for 8b and 405b
suachong Sep 11, 2025
470daf0
missing factor
suachong Sep 23, 2025
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
7 changes: 5 additions & 2 deletions mlperf_logging/compliance_checker/training_5.1.0/closed_llama31_405b.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -24,12 +24,15 @@
REQ: EXACTLY_ONE

- KEY:
NAME: opt_learning_rate_decay_steps
NAME: opt_learning_rate_warmup_steps
REQ: EXACTLY_ONE
POST: >
s['opt_learning_rate_warmup_steps'] = math.ceil(8000 * 1152 / s['global_batch_size'] )

- KEY:
NAME: opt_learning_rate_warmup_steps
NAME: opt_learning_rate_decay_steps
REQ: EXACTLY_ONE
CHECK: " v['value'] == math.ceil(1_200_000 * 1152 / s['global_batch_size'] ) - s['opt_learning_rate_warmup_steps'] "

- KEY:
NAME: opt_learning_rate_decay_schedule
Expand Down
8 changes: 5 additions & 3 deletions mlperf_logging/compliance_checker/training_5.1.0/closed_llama31_8b.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -23,13 +23,15 @@
REQ: EXACTLY_ONE

- KEY:
NAME: opt_learning_rate_decay_steps
NAME: opt_learning_rate_warmup_steps
REQ: EXACTLY_ONE
CHECK: " v['value'] == 1200000 "
POST: >
s['opt_learning_rate_warmup_steps'] = v['value']

- KEY:
NAME: opt_learning_rate_warmup_steps
NAME: opt_learning_rate_decay_steps
REQ: EXACTLY_ONE
CHECK: " v['value'] == math.ceil(1_200_000 / s['global_batch_size'] ) - s['opt_learning_rate_warmup_steps'] "

- KEY:
NAME: opt_learning_rate_decay_schedule
Expand Down