Repository navigation
Conversation
|
MLCommons CLA bot All contributors have signed the MLCommons CLA ✍️ ✅ |
uwlei
force-pushed
the
add-qwen35-397b-grpo-rcp-1152
branch
from
October 20, 2026 23:19
81badd3 to
f798681
Compare
uwlei
marked this pull request as ready for review
October 21, 2026 00:48
pavanky
reviewed
Oct 23, 2026
| "gradient_accumulation_steps": 144 | ||
| }, | ||
| "Epochs to converge": [ | ||
| 6912, 6912, 6912, 6912, 6912, 6912 |
Contributor
There was a problem hiding this comment.
I am not too familiar with the benchmark, but the epochs to converge was generally increasing with batch size, but the epochs to converge here goes down from 1024. Is this expected or just some corner case we are hitting?
This branch has not been deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Fixes #482
Supersedes #483
Reference RCP logs PR: mlcommons/training#911
Summary
Adds
qwen35_397b_grpo_ref_1152(BS = 1152,Epochs to converge = [6912, 6912, 6912, 6912, 6912, 6912]) tomlperf_logging/rcp_checker/training_6.1.0/rcps_qwen35_397b_grpo.jsonusing 6 independent seeds of the MLCommons public reference implementation (NVIDIA-NeMo/RLcommit3fca04c9b313d313302923a5bb6b0c8dc0340ed6, BF16 policy training + BF16 vLLM generation,in_flight_weight_updates: false) on 64 NVIDIA GB200 nodes (addressing review feedback on #483).Raw 6-seed reference MLLOG files (
seed_{1..6}.out): mlcommons/training#911Hyperparameters & Convergence Summary (
BS = 1152)BS = 1152(num_prompts_per_step = 72,num_generations_per_prompt = 16,gradient_accumulation_steps = 144)opt_base_learning_rate = 2.1213203436e-06(1e-6 * sqrt(1152 / 256))opt_gradient_clip_norm = 0.0589255651(0.125 * sqrt(256 / 1152))Epochs to converge = [6912, 6912, 6912, 6912, 6912, 6912](6/6 seeds converged at Step 6,eval_accuracyrange0.7291 - 0.7809 > 0.69)