AbhishekGhosh1998 commited on
Commit
f6ff918
·
verified ·
1 Parent(s): 750ce18

granularity results tiera-rerun-fixed-moonep @ 2026-08-30T07:20:47Z from sp-aiinfra-interactive-aghosh-7jtjl-8xb200-moe-cg-0

Browse files
Files changed (24) hide show
  1. tiera-rerun-fixed-moonep/.completed/ultraep__a1__gbs16 +1 -0
  2. tiera-rerun-fixed-moonep/STATUS.txt +3 -2
  3. tiera-rerun-fixed-moonep/hf_watcher.log +3 -0
  4. tiera-rerun-fixed-moonep/steps.tsv +1 -0
  5. tiera-rerun-fixed-moonep/study.log +62 -0
  6. tiera-rerun-fixed-moonep/ultraep/ponly/manifest.txt +19 -0
  7. tiera-rerun-fixed-moonep/ultraep/ponly/pip_freeze.txt +317 -0
  8. tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq4096__gbs16__cap2p40625/rep1/mem_samples.csv +200 -0
  9. tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq4096__gbs16__cap2p40625/rep1/run.env +70 -0
  10. tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq4096__gbs16__cap2p40625/rep1/train.log +1031 -0
  11. tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq4096__gbs16__cap2p40625/rep2/mem_samples.csv +200 -0
  12. tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq4096__gbs16__cap2p40625/rep2/run.env +70 -0
  13. tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq4096__gbs16__cap2p40625/rep2/train.log +1032 -0
  14. tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq4096__gbs16__cap2p40625/rep3/mem_samples.csv +192 -0
  15. tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq4096__gbs16__cap2p40625/rep3/run.env +70 -0
  16. tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq4096__gbs16__cap2p40625/rep3/train.log +1032 -0
  17. tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq8192__gbs16__cap2p40625/rep1/mem_samples.csv +192 -0
  18. tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq8192__gbs16__cap2p40625/rep1/run.env +70 -0
  19. tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq8192__gbs16__cap2p40625/rep1/train.log +1024 -0
  20. tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq8192__gbs16__cap2p40625/rep2/mem_samples.csv +168 -0
  21. tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq8192__gbs16__cap2p40625/rep2/run.env +70 -0
  22. tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq8192__gbs16__cap2p40625/rep2/train.log +977 -0
  23. tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/results.tsv +5 -0
  24. tiera-rerun-fixed-moonep/ultraep/ponly/provenance.txt +90 -0
tiera-rerun-fixed-moonep/.completed/ultraep__a1__gbs16 ADDED
@@ -0,0 +1 @@
 
 
1
+ 2026-08-30T07:15:17+00:00
tiera-rerun-fixed-moonep/STATUS.txt CHANGED
@@ -1,12 +1,13 @@
1
  study_root: /workspace/tiera-rerun-fixed-moonep
2
  gbs: 16
3
- updated: 2026-08-30T07:00:27+00:00
4
- current: ultraep__a1__gbs16
5
 
6
  completed steps:
7
  echo__a1__gbs16
8
  echo__a2__gbs16
9
  echo__ponly__gbs16
 
10
 
11
  failed steps:
12
  FAIL echo tierbprobe 16 2026-08-30T04:52:38+00:00 2026-08-30T05:06:38+00:00 /workspace/tiera-rerun-fixed-moonep/echo/tierbprobe
 
1
  study_root: /workspace/tiera-rerun-fixed-moonep
2
  gbs: 16
3
+ updated: 2026-08-30T07:15:19+00:00
4
+ current: ultraep__ponly__gbs16
5
 
6
  completed steps:
7
  echo__a1__gbs16
8
  echo__a2__gbs16
9
  echo__ponly__gbs16
10
+ ultraep__a1__gbs16
11
 
12
  failed steps:
13
  FAIL echo tierbprobe 16 2026-08-30T04:52:38+00:00 2026-08-30T05:06:38+00:00 /workspace/tiera-rerun-fixed-moonep/echo/tierbprobe
tiera-rerun-fixed-moonep/hf_watcher.log CHANGED
@@ -38,3 +38,6 @@ https://huggingface.co/MoE-CUDA-Graph-Analysis/tierb-b200-gbs8/tree/main
38
  uploaded: /workspace/tiera-rerun-fixed-moonep -> MoE-CUDA-Graph-Analysis/tierb-b200-gbs8/tiera-rerun-fixed-moonep
39
  https://huggingface.co/MoE-CUDA-Graph-Analysis/tierb-b200-gbs8/tree/main
40
  [2026-08-30T07:10:43+00:00] backup pass
 
 
 
 
38
  uploaded: /workspace/tiera-rerun-fixed-moonep -> MoE-CUDA-Graph-Analysis/tierb-b200-gbs8/tiera-rerun-fixed-moonep
39
  https://huggingface.co/MoE-CUDA-Graph-Analysis/tierb-b200-gbs8/tree/main
40
  [2026-08-30T07:10:43+00:00] backup pass
41
+ uploaded: /workspace/tiera-rerun-fixed-moonep -> MoE-CUDA-Graph-Analysis/tierb-b200-gbs8/tiera-rerun-fixed-moonep
42
+ https://huggingface.co/MoE-CUDA-Graph-Analysis/tierb-b200-gbs8/tree/main
43
+ [2026-08-30T07:20:46+00:00] backup pass
tiera-rerun-fixed-moonep/steps.tsv CHANGED
@@ -4,3 +4,4 @@ FAIL ultraep tierbprobe 16 2026-08-30T05:06:44+00:00 2026-08-30T05:20:40+00:00 /
4
  OK echo a1 16 2026-08-30T06:20:31+00:00 2026-08-30T06:36:28+00:00 /workspace/tiera-rerun-fixed-moonep/echo/a1
5
  OK echo ponly 16 2026-08-30T06:36:31+00:00 2026-08-30T06:44:11+00:00 /workspace/tiera-rerun-fixed-moonep/echo/ponly
6
  OK echo a2 16 2026-08-30T06:44:13+00:00 2026-08-30T07:00:25+00:00 /workspace/tiera-rerun-fixed-moonep/echo/a2
 
 
4
  OK echo a1 16 2026-08-30T06:20:31+00:00 2026-08-30T06:36:28+00:00 /workspace/tiera-rerun-fixed-moonep/echo/a1
5
  OK echo ponly 16 2026-08-30T06:36:31+00:00 2026-08-30T06:44:11+00:00 /workspace/tiera-rerun-fixed-moonep/echo/ponly
6
  OK echo a2 16 2026-08-30T06:44:13+00:00 2026-08-30T07:00:25+00:00 /workspace/tiera-rerun-fixed-moonep/echo/a2
7
+ OK ultraep a1 16 2026-08-30T07:00:27+00:00 2026-08-30T07:15:17+00:00 /workspace/tiera-rerun-fixed-moonep/ultraep/a1
tiera-rerun-fixed-moonep/study.log CHANGED
@@ -659,3 +659,65 @@ https://huggingface.co/MoE-CUDA-Graph-Analysis/tierb-b200-gbs8/tree/main
659
  [guard] iter 7/12 (60s)
660
  [guard] iter 12/12 (70s)
661
  [guard] tearing down process group (outcome=DONE)
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
659
  [guard] iter 7/12 (60s)
660
  [guard] iter 12/12 (70s)
661
  [guard] tearing down process group (outcome=DONE)
662
+ [guard] iter 6/12 (60s)
663
+ [guard] iter 12/12 (70s)
664
+ [guard] tearing down process group (outcome=DONE)
665
+ --------------------------------------------------------------------------
666
+ [07:11:39] CELL G4__ultraep__one__seq4096__gbs16__cap1p095703125 (slots=1)
667
+ [guard] iter 8/12 (60s)
668
+ [guard] iter 12/12 (70s)
669
+ [guard] tearing down process group (outcome=DONE)
670
+ [guard] iter 8/12 (60s)
671
+ [guard] iter 12/12 (70s)
672
+ [guard] tearing down process group (outcome=DONE)
673
+ [guard] iter 6/12 (60s)
674
+ [guard] iter 12/12 (70s)
675
+ [guard] tearing down process group (outcome=DONE)
676
+ --------------------------------------------------------------------------
677
+
678
+ ==============================================================
679
+ SWEEP COMPLETE ok=12 failed=0
680
+ ==============================================================
681
+ config backend policy slots seq gbs cap recompute rep status iter_ms_mean iter_ms_min peak_mib run_dir
682
+ G1 ultraep one 1 4096 16 2.40625 none 1 OK 888.60 869.80 129475 /workspace/tiera-rerun-fixed-moonep/ultraep/a1/a1_oneslot_gbs16/G1__ultraep__one__seq4096__gbs16__cap2p40625/rep1
683
+ G1 ultraep one 1 4096 16 2.40625 none 2 OK 891.61 869.80 129475 /workspace/tiera-rerun-fixed-moonep/ultraep/a1/a1_oneslot_gbs16/G1__ultraep__one__seq4096__gbs16__cap2p40625/rep2
684
+ G1 ultraep one 1 4096 16 2.40625 none 3 OK 891.00 869.80 129475 /workspace/tiera-rerun-fixed-moonep/ultraep/a1/a1_oneslot_gbs16/G1__ultraep__one__seq4096__gbs16__cap2p40625/rep3
685
+ G2 ultraep one 1 4096 16 1.08203125 none 1 OK 466.57 441.70 102733 /workspace/tiera-rerun-fixed-moonep/ultraep/a1/a1_oneslot_gbs16/G2__ultraep__one__seq4096__gbs16__cap1p08203125/rep1
686
+ G2 ultraep one 1 4096 16 1.08203125 none 2 OK 466.68 441.70 102733 /workspace/tiera-rerun-fixed-moonep/ultraep/a1/a1_oneslot_gbs16/G2__ultraep__one__seq4096__gbs16__cap1p08203125/rep2
687
+ G2 ultraep one 1 4096 16 1.08203125 none 3 OK 466.65 441.70 102733 /workspace/tiera-rerun-fixed-moonep/ultraep/a1/a1_oneslot_gbs16/G2__ultraep__one__seq4096__gbs16__cap1p08203125/rep3
688
+ G3 ultraep one 1 4096 16 1.072265625 none 1 OK 448.07 438.50 102043 /workspace/tiera-rerun-fixed-moonep/ultraep/a1/a1_oneslot_gbs16/G3__ultraep__one__seq4096__gbs16__cap1p072265625/rep1
689
+ G3 ultraep one 1 4096 16 1.072265625 none 2 OK 448.14 435.80 102043 /workspace/tiera-rerun-fixed-moonep/ultraep/a1/a1_oneslot_gbs16/G3__ultraep__one__seq4096__gbs16__cap1p072265625/rep2
690
+ G3 ultraep one 1 4096 16 1.072265625 none 3 OK 447.84 432.10 102043 /workspace/tiera-rerun-fixed-moonep/ultraep/a1/a1_oneslot_gbs16/G3__ultraep__one__seq4096__gbs16__cap1p072265625/rep3
691
+ G4 ultraep one 1 4096 16 1.095703125 none 1 OK 467.36 461.70 114075 /workspace/tiera-rerun-fixed-moonep/ultraep/a1/a1_oneslot_gbs16/G4__ultraep__one__seq4096__gbs16__cap1p095703125/rep1
692
+ G4 ultraep one 1 4096 16 1.095703125 none 2 OK 467.78 461.70 114077 /workspace/tiera-rerun-fixed-moonep/ultraep/a1/a1_oneslot_gbs16/G4__ultraep__one__seq4096__gbs16__cap1p095703125/rep2
693
+ G4 ultraep one 1 4096 16 1.095703125 none 3 OK 467.57 461.70 114077 /workspace/tiera-rerun-fixed-moonep/ultraep/a1/a1_oneslot_gbs16/G4__ultraep__one__seq4096__gbs16__cap1p095703125/rep3
694
+
695
+ Results: /workspace/tiera-rerun-fixed-moonep/ultraep/a1/a1_oneslot_gbs16/results.tsv
696
+ Roll up: python3 /workspace/blackwell_run_scripts/granularity/tools/collect.py /workspace/tiera-rerun-fixed-moonep/ultraep/a1/a1_oneslot_gbs16
697
+ ==============================================================
698
+
699
+ Campaign results: /workspace/tiera-rerun-fixed-moonep/ultraep/a1
700
+ Roll up: python3 /workspace/blackwell_run_scripts/granularity/tools/collect.py /workspace/tiera-rerun-fixed-moonep/ultraep/a1 --compare
701
+ Already logged in to Hugging Face as: AbhishekGhosh1998
702
+ uploaded: /workspace/tiera-rerun-fixed-moonep/ultraep/a1 -> MoE-CUDA-Graph-Analysis/tierb-b200-gbs8/tiera-rerun-fixed-moonep/ultraep/a1
703
+ https://huggingface.co/MoE-CUDA-Graph-Analysis/tierb-b200-gbs8/tree/main
704
+
705
+ === STEP ultraep__ponly__gbs16
706
+ ==============================================================
707
+ CAMPAIGN: ultraep backend=ultraep phase=ponly
708
+ Megatron: /workspace/Megatron-UltraEP @ ultraep-bracing-regions-through-kernels
709
+ GBS=16 seq=4096 repeats=3 nsys=0
710
+ results -> /workspace/tiera-rerun-fixed-moonep/ultraep/ponly
711
+ ==============================================================
712
+ --------------------------------------------------------------------------
713
+ [07:15:34] CELL PONLY__ultraep__one__seq4096__gbs16__cap2p40625 (slots=1)
714
+ [guard] iter 12/12 (60s)
715
+ [guard] tearing down process group (outcome=DONE)
716
+ [guard] iter 12/12 (60s)
717
+ [guard] tearing down process group (outcome=DONE)
718
+ [guard] iter 12/12 (60s)
719
+ [guard] tearing down process group (outcome=DONE)
720
+ --------------------------------------------------------------------------
721
+ [07:18:47] CELL PONLY__ultraep__one__seq8192__gbs16__cap2p40625 (slots=1)
722
+ [guard] iter 12/12 (60s)
723
+ [guard] tearing down process group (outcome=DONE)
tiera-rerun-fixed-moonep/ultraep/ponly/manifest.txt ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ timestamp: 2026-08-30T07:15:34+00:00
2
+ host: sp-aiinfra-interactive-aghosh-7jtjl-8xb200-moe-cg-0
3
+ gpu: NVIDIA B200 x8
4
+ driver: 580.95.05
5
+ study_scripts_branch: granularity-ultraep
6
+ study_scripts_commit: 5e3993b
7
+ campaign: ultraep
8
+ backend: ultraep
9
+ phase: ponly
10
+ gbs: 16
11
+ seq: 4096
12
+ configs: G1 G2 G3 G4
13
+ capacity: 2.40625
14
+ repeats: 3
15
+ nsys: 0
16
+ megatron_repo: /workspace/Megatron-UltraEP
17
+ megatron_branch_required: ultraep-bracing-regions-through-kernels
18
+ megatron_branch_actual: ultraep-bracing-regions-through-kernels
19
+ megatron_commit: 03d57ef7
tiera-rerun-fixed-moonep/ultraep/ponly/pip_freeze.txt ADDED
@@ -0,0 +1,317 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ absl-py==2.3.0
2
+ aiohappyeyeballs @ file:///rapids/aiohappyeyeballs-2.6.1-py3-none-any.whl#sha256=f349ba8f4b75cb25c99c5c2d84e997e485204d2902a9597802b0371f09331fb8
3
+ aiohttp @ file:///rapids/aiohttp-3.12.7-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl#sha256=94f98e0e5a49f89b252e115844f756c04fc8050f38252a32a3dd994ce8121f10
4
+ aiosignal @ file:///rapids/aiosignal-1.3.2-py2.py3-none-any.whl#sha256=45cde58e409a301715980c2b01d0c28bdde3770d8290b5eb2173759d9acb31a5
5
+ annotated-types==0.7.0
6
+ anyio==4.9.0
7
+ apex @ file:///opt/pytorch/apex/dist/apex-0.1-cp312-cp312-linux_x86_64.whl#sha256=a1508e4a386362e9c8f46c92ccada9110e421212b964c694235d0fd5a224b170
8
+ argon2-cffi==25.1.0
9
+ argon2-cffi-bindings==21.2.0
10
+ arrow==1.3.0
11
+ asciitree @ file:///rapids/asciitree-0.3.3-py3-none-any.whl#sha256=3e507ccfaeb417823158091fdb9d26b0205f6a59ba680b479a71db65c73d01d4
12
+ asttokens==3.0.0
13
+ astunparse==1.6.3
14
+ async-lru==2.0.5
15
+ attrs==25.3.0
16
+ audioread==3.0.1
17
+ babel==2.17.0
18
+ beautifulsoup4==4.13.4
19
+ black==25.1.0
20
+ bleach==6.2.0
21
+ blis==0.7.11
22
+ cachetools @ file:///rapids/cachetools-6.0.0-py3-none-any.whl#sha256=82e73ba88f7b30228b5507dce1a1f878498fc669d972aef2dde4f3a3c24f103e
23
+ catalogue==2.0.10
24
+ certifi==2025.4.26
25
+ cffi==1.17.1
26
+ charset-normalizer==3.4.2
27
+ click==8.5.0
28
+ cloudpathlib==0.21.1
29
+ cloudpickle @ file:///rapids/cloudpickle-3.1.1-py3-none-any.whl#sha256=c8c5a44295039331ee9dad40ba100a9c7297b6f988e50e87ccdf3765a668350e
30
+ cmake==3.31.6
31
+ comm==0.2.2
32
+ confection==0.1.5
33
+ contourpy==1.3.2
34
+ cuda-bindings @ file:///rapids/cuda_bindings-12.9.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl#sha256=ff0e28d1e34758654b9c961e1f55e4786e49aee6a4dbceaf3cc24c46c672df7e
35
+ cuda-python @ file:///rapids/cuda_python-12.9.0-py3-none-any.whl#sha256=926acba49b2c0a0374c61b7c98f337c085199cf51cdfe4d6423c4129c20547a7
36
+ cudf @ file:///rapids/cudf-25.4.0-cp312-cp312-linux_x86_64.whl#sha256=b779a8e8c0e8577b9b560895958496da211bfac49cdc4f11a49e8d18cb82e348
37
+ cudf-polars @ file:///rapids/cudf_polars-25.4.0-py3-none-any.whl#sha256=227b6c5147e93ea86282c8f7219118f916769fa2ca0cdd931e97a148aa66b869
38
+ cugraph @ file:///rapids/cugraph-25.4.0-cp312-cp312-linux_x86_64.whl#sha256=b837c2dc2f7b235c702e4f8df64479b8fcbb342f848516d5faf0cdc4b5f31f29
39
+ cugraph-service-client @ file:///rapids/cugraph_service_client-25.4.0-py3-none-any.whl#sha256=7e568c3e41154b9752f5173555ffcc282ac52d72343feeb79653b2e2b1232470
40
+ cugraph-service-server @ file:///rapids/cugraph_service_server-25.4.0-py3-none-any.whl#sha256=a2830dcffafcccb16e12ebf85b98f8f247869d38c01ab4dd8664a1e7439d31be
41
+ cuml @ file:///rapids/cuml-25.4.0-cp312-cp312-linux_x86_64.whl#sha256=1e90a464a18b094e24797f1676688c5d7696c4cf97c934504214a53d5aa09aa5
42
+ cupy-cuda12x @ file:///rapids/cupy_cuda12x-13.3.0-cp312-cp312-linux_x86_64.whl#sha256=6a2dfa0aa1e8189a10d1534276c08f1d2f73b747fd28bfbe7b9f4ef66d9af73f
43
+ cuvs @ file:///rapids/cuvs-25.4.0-cp312-cp312-linux_x86_64.whl#sha256=b12f6b9f80e053a3b38542c2947d2516e468c02cad54a13d8b1a4a61d5aa7914
44
+ cycler==0.12.1
45
+ cymem==2.0.11
46
+ Cython==3.1.1
47
+ dask @ file:///rapids/dask-2025.2.0-py3-none-any.whl#sha256=f0fdeef6ceb0a06569d456c9e704f220f7f54e80f3a6ea42ab98cea6bc642b6e
48
+ dask-cuda @ file:///rapids/dask_cuda-25.4.0-py3-none-any.whl#sha256=442d87da0a95d00ca2a262b64c76e785a32ad905c82000fa2227932f81390e19
49
+ dask-cudf @ file:///rapids/dask_cudf-25.4.0-py3-none-any.whl#sha256=5e7cea0b26daea17da6e0bb457435e24460f76ba740358a3a18e239d6015c95c
50
+ datasets==5.0.1
51
+ debugpy==1.8.14
52
+ decorator==5.2.1
53
+ -e git+https://github.com/deepseek-ai/DeepEP.git@34152ae28f80bcc3ee38d7a12cb2ad87cfd4ea72#egg=deep_ep
54
+ defusedxml==0.7.1
55
+ dill==0.4.1
56
+ distributed @ file:///rapids/distributed-2025.2.0-py3-none-any.whl#sha256=368462084be47092d0a835f6c64c9b852da64dc88302b76d5f93bd5131228c67
57
+ distributed-ucxx @ file:///rapids/distributed_ucxx-0.43.0-py3-none-any.whl#sha256=94f60d0ccb0c94ba8dbfa1eaf455874d18aaab17e2ae218c38590e9fdc2d721d
58
+ dm-tree==0.1.9
59
+ einops==0.8.1
60
+ execnet==2.1.1
61
+ executing==2.2.0
62
+ expecttest==0.3.0
63
+ fasteners @ file:///rapids/fasteners-0.19-py3-none-any.whl#sha256=758819cb5d94cdedf4e836988b74de396ceacb8e2794d21f82d131fd9ee77237
64
+ fastjsonschema==2.21.1
65
+ fastrlock @ file:///rapids/fastrlock-0.8.3-cp312-cp312-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_28_x86_64.whl#sha256=5f13ec08f1adb1aa916c384b05ecb7dbebb8df9ea81abd045f60941c6283a670
66
+ filelock==3.18.0
67
+ flash_attn @ file:///opt/pytorch/flash_attn-2.7.4.post1-cp312-cp312-linux_x86_64.whl#sha256=a10455d6c56b387785b7413fd0bbcb392b42b3d4f4b88fda5d86cc8a9ce9182a
68
+ fonttools==4.58.1
69
+ fqdn==1.5.1
70
+ frozenlist @ file:///rapids/frozenlist-1.6.0-cp312-cp312-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl#sha256=4e1be9111cb6756868ac242b3c2bd1f09d9aea09846e4f5c23715e7afb647103
71
+ fsspec==2025.5.1
72
+ gast==0.6.0
73
+ grpcio==1.62.1
74
+ h11==0.16.0
75
+ hf-xet==1.6.0
76
+ httpcore==1.0.9
77
+ httpx==0.28.1
78
+ huggingface_hub==1.29.0
79
+ hypothesis==6.130.8
80
+ idna==3.10
81
+ importlib_metadata==8.7.0
82
+ iniconfig==2.1.0
83
+ intel-openmp==2021.4.0
84
+ ipykernel==6.29.5
85
+ ipython==9.3.0
86
+ ipython_pygments_lexers==1.1.1
87
+ isoduration==20.11.0
88
+ isort==6.0.1
89
+ jedi==0.19.2
90
+ Jinja2==3.1.6
91
+ joblib==1.5.1
92
+ json5==0.12.0
93
+ jsonpointer==3.0.0
94
+ jsonschema==4.24.0
95
+ jsonschema-specifications==2025.4.1
96
+ jupyter-events==0.12.0
97
+ jupyter-lsp==2.2.5
98
+ jupyter_client==8.6.3
99
+ jupyter_core==5.8.1
100
+ jupyter_server==2.16.0
101
+ jupyter_server_terminals==0.5.3
102
+ jupyterlab==4.4.3
103
+ jupyterlab_code_formatter==3.0.2
104
+ jupyterlab_pygments==0.3.0
105
+ jupyterlab_server==2.27.3
106
+ jupyterlab_tensorboard_pro @ file:///builder/jupyterlab_tensorboard_pro-4.0.0-py2.py3-none-any.whl#sha256=6bfcb5ced3358eb72150f4b2ab91312faef2f0ca14d845e8379d8fa52df6902e
107
+ jupytext==1.17.2
108
+ kiwisolver==1.4.8
109
+ kvikio @ file:///rapids/kvikio-25.4.0-cp312-cp312-linux_x86_64.whl#sha256=8b40fd8ea9e4a2974c04b50b7fc26b06094ac807f9a233281d0bafbaa71b1d85
110
+ langcodes==3.5.0
111
+ language_data==1.3.0
112
+ lazy_loader==0.4
113
+ libcudf @ file:///rapids/libcudf-25.4.0-py3-none-linux_x86_64.whl#sha256=238320b4958c1737bba2a609673e2539a78216b811b5498a68e00395d9b021a4
114
+ libcugraph @ file:///rapids/libcugraph-25.4.0-py3-none-linux_x86_64.whl#sha256=29085e901b2f7569ff2653bb1355fe6fad20f2fcaa585f5649da9867d8df766d
115
+ libcuml @ file:///rapids/libcuml-25.4.0-py3-none-linux_x86_64.whl#sha256=8e2e444c72b0e995e15a48da37ff12fb410a47a2baa69ba86ccc9b1a3db91e8c
116
+ libcuvs @ file:///rapids/libcuvs-25.4.0-py3-none-linux_x86_64.whl#sha256=e2f8db61c92b769c590dd752f43439ddd62b2f6592a3c337362b6c1ef220c3b9
117
+ libkvikio @ file:///rapids/libkvikio-25.4.0-py3-none-linux_x86_64.whl#sha256=da8a5713b05fc478bf40529a3e0fb61a2d605bb5a70ca4d461f96a6005dbe8b7
118
+ libraft @ file:///rapids/libraft-25.4.0-py3-none-linux_x86_64.whl#sha256=b921dec005aa4a321dd156776a18376feb47c96a5b67eec125c8a8e19edd1e24
119
+ librmm @ file:///rapids/librmm-25.4.0-py3-none-linux_x86_64.whl#sha256=ff951811bda729d4fa1109ab906225ae1f489a6ec2eeab5d1b674cc34bd4422f
120
+ librmm-cu12 @ file:///rapids/librmm_cu12-25.4.0-py3-none-any.whl#sha256=29296ef2b0e90679a0db280e86015be12d6ab9f4e62a181aa085f62dbadd06b3
121
+ librosa==0.11.0
122
+ libucxx @ file:///rapids/libucxx-0.43.0-py3-none-linux_x86_64.whl#sha256=f15849d7ac2fb68f0fd8998838a13321ddcfb0c594f7e6cbc401a3bcc4750566
123
+ lightning-thunder @ file:///opt/pytorch/lightning-thunder
124
+ lightning-utilities==0.14.3
125
+ lintrunner==0.12.7
126
+ llvmlite==0.42.0
127
+ locket @ file:///rapids/locket-1.0.0-py2.py3-none-any.whl#sha256=b6c819a722f7b6bd955b80781788e4a66a55628b858d347536b7e81325a3a5e3
128
+ looseversion==1.3.0
129
+ marisa-trie==1.2.1
130
+ Markdown==3.8
131
+ markdown-it-py==3.0.0
132
+ MarkupSafe==3.0.2
133
+ matplotlib==3.10.3
134
+ matplotlib-inline==0.1.7
135
+ mdit-py-plugins==0.4.2
136
+ mdurl==0.1.2
137
+ -e git+ssh://git@github.com/sriram-17-17/my-Megatron-MoE-Echo.git@9a23679390d7ab1fe462bc61fa31d90d393db49c#egg=megatron_core
138
+ mistune==3.1.3
139
+ mkl==2021.1.1
140
+ mkl-devel==2021.1.1
141
+ mkl-include==2021.1.1
142
+ ml_dtypes==0.6.0
143
+ mock==5.2.0
144
+ -e git+ssh://git@github.com/sriram-17-17/MoonEP.git@d8bc6667fad7177e6e09b637d773fa67f08bae93#egg=moonep
145
+ mpmath==1.3.0
146
+ msgpack==1.1.0
147
+ multidict @ file:///rapids/multidict-6.4.4-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl#sha256=a920f9cf2abdf6e493c519492d892c362007f113c94da4c239ae88429835bad1
148
+ multiprocess==0.70.19
149
+ murmurhash==1.0.13
150
+ mypy_extensions==1.1.0
151
+ nbclient==0.10.2
152
+ nbconvert==7.16.6
153
+ nbformat==5.10.4
154
+ nest-asyncio==1.6.0
155
+ networkx==3.5
156
+ ninja==1.11.1.4
157
+ notebook==7.4.3
158
+ notebook_shim==0.2.4
159
+ numba==0.59.1
160
+ numba-cuda @ file:///rapids/numba_cuda-0.4.0-py3-none-any.whl#sha256=5b95a5f42068771e37142944d44fe844fa87f1b2f89051461a4cb0af91bcc28a
161
+ numcodecs @ file:///rapids/numcodecs-0.13.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl#sha256=da2230484e6102e5fa3cc1a5dd37ca1f92dfbd183d91662074d6f7574e3e8f53
162
+ numpy==1.26.4
163
+ nvdlfw_inspect @ git+https://github.com/NVIDIA/nvidia-dlfw-inspect.git@6b60eb0e675606fb2fbbfcb12667ecbec75eaf87
164
+ nvfuser==0.2.27a0+9bf5aca
165
+ nvidia-cuda-cccl-cu12==12.9.27
166
+ nvidia-cudnn-frontend @ file:///opt/pytorch/lightning-thunder/tmp/cudnn_frontend
167
+ nvidia-cutlass-dsl==4.4.2
168
+ nvidia-cutlass-dsl-libs-base==4.4.2
169
+ nvidia-dali-cuda120==1.50.0
170
+ nvidia-ml-py @ file:///rapids/nvidia_ml_py-12.575.51-py3-none-any.whl#sha256=eb8641800d98ce40a22f479873f34b482e214a7e80349c63be51c3919845446e
171
+ nvidia-modelopt==0.29.0
172
+ nvidia-modelopt-core==0.29.0
173
+ nvidia-nvcomp-cu12==4.2.0.14
174
+ nvidia-nvimgcodec-cu12==0.5.0.13
175
+ nvidia-nvjpeg-cu12==12.4.0.16
176
+ nvidia-nvjpeg2k-cu12==0.8.1.40
177
+ nvidia-nvshmem-cu12==3.7.2
178
+ nvidia-nvtiff-cu12==0.5.0.67
179
+ nvidia-resiliency-ext==0.4.0
180
+ nvtx @ file:///rapids/nvtx-0.2.11-cp312-cp312-linux_x86_64.whl#sha256=73521e37cba42b4db0b3cd68fc5eb814e55c500a61eeed6edcdf6746315ac53e
181
+ nx-cugraph @ file:///rapids/nx_cugraph-25.4.0-py3-none-any.whl#sha256=e625444486cc899a77bd48e7da76a2edd95c0006dd95fbf323745feae7f33111
182
+ onnx @ file:///opt/pytorch/pytorch/third_party/onnx
183
+ onnx-ir==1.0.0
184
+ onnxscript==0.7.1
185
+ opt_einsum==3.4.0
186
+ optree==0.16.0
187
+ overrides==7.7.0
188
+ packaging==23.2
189
+ pandas @ file:///rapids/pandas-2.2.3-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl#sha256=fffb8ae78d8af97f849404f21411c95062db1496aeb3e56f146f0355c9989319
190
+ pandocfilters==1.5.1
191
+ parso==0.8.4
192
+ partd @ file:///rapids/partd-1.4.2-py3-none-any.whl#sha256=978e4ac767ec4ba5b86c6eaa52e5a2a3bc748a2ca839e8cc798f1cc6ce6efb0f
193
+ pathspec==0.12.1
194
+ pexpect==4.9.0
195
+ pillow==11.2.1
196
+ platformdirs==4.3.8
197
+ pluggy==1.6.0
198
+ ply @ file:///rapids/ply-3.11-py2.py3-none-any.whl#sha256=096f9b8350b65ebd2fd1346b12452efe5b9607f7482813ffca50c22722a807ce
199
+ polars @ file:///rapids/polars-1.25.2-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl#sha256=f7fcbb4f476784384ccda48757fca4e8c2e2c5a0a3aef3717aaf56aee4e30e09
200
+ polygraphy==0.49.20
201
+ pooch==1.8.2
202
+ preshed==3.0.10
203
+ prometheus_client==0.22.1
204
+ prompt_toolkit==3.0.51
205
+ propcache @ file:///rapids/propcache-0.3.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl#sha256=563f9d8c03ad645597b8d010ef4e9eab359faeb11a0a2ac9f7b4bc8c28ebef25
206
+ protobuf==4.24.4
207
+ psutil==7.0.0
208
+ ptyprocess==0.7.0
209
+ PuLP==3.2.1
210
+ pure_eval==0.2.3
211
+ pyarrow==25.0.1
212
+ pybind11==2.13.6
213
+ pybind11_global==2.13.6
214
+ pycocotools @ git+https://github.com/nvidia/cocoapi.git@08350a157b41f0755a536441b5abec89ab934c5b#subdirectory=PythonAPI
215
+ pycparser==2.22
216
+ pydantic==2.11.5
217
+ pydantic_core==2.33.2
218
+ Pygments==2.19.1
219
+ pylibcudf @ file:///rapids/pylibcudf-25.4.0-cp312-cp312-linux_x86_64.whl#sha256=401b8bfd022d06f84e120779ca7c3d01a28b68762aaa8f7873b10d6b1975cf6a
220
+ pylibcugraph @ file:///rapids/pylibcugraph-25.4.0-cp312-cp312-linux_x86_64.whl#sha256=061c7389a88c815b54fd5d3a3584fb87bf693e5602f5be335260d0ba575ba557
221
+ pylibcugraphops @ file:///rapids/pylibcugraphops-25.4.0-cp312-cp312-linux_x86_64.whl#sha256=a25ef4b22166c80718d5c7e5bddb755e81a2fe340fa5becd0ac36ec3edfb208a
222
+ pylibraft @ file:///rapids/pylibraft-25.4.0-cp312-cp312-linux_x86_64.whl#sha256=f2ecd23fe51f9c2caa87bbbe85cb32fcaff4ecfb52ceb9a9991bc50f82782c2e
223
+ pylibwholegraph @ file:///rapids/pylibwholegraph-25.4.0-cp312-cp312-linux_x86_64.whl#sha256=b82913d1e38bda76d8c272fbd8cc9072d987209d3d47cc93a464a37c2925aca2
224
+ pynvjitlink @ file:///rapids/pynvjitlink-0.3.0-cp312-cp312-linux_x86_64.whl#sha256=066caf1314730643d3b48c9847bd65789defdd6a7c78993176eb3d8bbe3f4919
225
+ pynvml @ file:///rapids/pynvml-12.0.0-py3-none-any.whl#sha256=fdff84b62a27dbe98e08e1a647eb77342bef1aebe0878bcd15e99a83fcbecb9e
226
+ pyparsing==3.2.3
227
+ pytest==8.1.1
228
+ pytest-flakefinder==1.1.0
229
+ pytest-rerunfailures==15.1
230
+ pytest-shard==0.1.2
231
+ pytest-xdist==3.7.0
232
+ python-dateutil==2.9.0.post0
233
+ python-hostlist==2.2.1
234
+ python-json-logger==3.3.0
235
+ pytorch-triton @ file:///tmp/dist/pytorch_triton-3.3.0%2Bgit96316ce52.nvinternal-cp312-cp312-linux_x86_64.whl#sha256=6d6921772b9dfd0243c055e0d054de42a113e92cabcfdd1d3e80a5ad9c08cf7d
236
+ pytz @ file:///rapids/pytz-2023.4-py2.py3-none-any.whl#sha256=f90ef520d95e7c46951105338d918664ebfd6f1d995bd7d153127ce90efafa6a
237
+ PyYAML==6.0.2
238
+ pyzmq==26.4.0
239
+ raft-dask @ file:///rapids/raft_dask-25.4.0-cp312-cp312-linux_x86_64.whl#sha256=36cf37ba6ce35454c4729c2cb597a360f22780eefb11d514a16a07137da5eb6c
240
+ rapids-dask-dependency @ file:///rapids/rapids_dask_dependency-25.4.0a0-py3-none-any.whl#sha256=484faca626e7760112a7d8ad3426f5bdd720b609006d8dca87c2607076a348bd
241
+ rapids-logger @ file:///rapids/rapids_logger-0.1.18-py3-none-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl#sha256=3617b2d86291ac6cb957542895f20fc9737ace29773ca90337e6fb04c73bd120
242
+ referencing==0.36.2
243
+ regex==2024.11.6
244
+ requests==2.32.3
245
+ rfc3339-validator==0.1.4
246
+ rfc3986-validator==0.1.1
247
+ rich==14.0.0
248
+ rmm @ file:///rapids/rmm-25.4.0-cp312-cp312-linux_x86_64.whl#sha256=1824bcab3836296666980a997ddcb9c4bca24dfb873012cfa0a7ecea583ca712
249
+ rpds-py==0.25.1
250
+ safetensors==0.5.3
251
+ scikit-learn==1.6.1
252
+ scipy==1.15.3
253
+ Send2Trash==1.8.3
254
+ setuptools==78.1.1
255
+ shellingham==1.5.4
256
+ six==1.16.0
257
+ smart-open==7.1.0
258
+ sniffio==1.3.1
259
+ sortedcontainers==2.4.0
260
+ soundfile==0.13.1
261
+ soupsieve==2.7
262
+ soxr==0.5.0.post1
263
+ spacy==3.7.5
264
+ spacy-legacy==3.0.12
265
+ spacy-loggers==1.0.5
266
+ srsly==2.5.1
267
+ stack-data==0.6.3
268
+ sympy==1.14.0
269
+ tabulate==0.9.0
270
+ tbb==2021.13.1
271
+ tblib @ file:///rapids/tblib-3.1.0-py3-none-any.whl#sha256=670bb4582578134b3d81a84afa1b016128b429f3d48e6cbbaecc9d15675e984e
272
+ tensorboard==2.16.2
273
+ tensorboard-data-server==0.7.2
274
+ tensorrt @ file:///workspace/TensorRT-10.11.0.33/python/tensorrt-10.11.0.33-cp312-none-linux_x86_64.whl#sha256=c2f18f72758a3d54a71901495c8fda9e7094a8bcdd2a5b6b58967fbf05733b3a
275
+ terminado==0.18.1
276
+ thinc==8.2.5
277
+ threadpoolctl==3.6.0
278
+ thriftpy2 @ file:///rapids/thriftpy2-0.5.2-cp312-cp312-linux_x86_64.whl#sha256=c0c7d939bdc5ccf7739154959938445d5fdf72f1d6e12bd563947a36f67e2147
279
+ tinycss2==1.4.0
280
+ toolz @ file:///rapids/toolz-1.0.0-py3-none-any.whl#sha256=292c8f1c4e7516bf9086f8850935c799a874039c8bcf959d47b600e4c44a6236
281
+ torch @ file:///opt/transfer/torch-2.8.0a0%2B5228986c39.nv25.6-29946474-cp312-cp312-linux_x86_64.whl#sha256=7a3c9ff5288d8102a31296b3d62491f3f47c8820a5c9aaefc7dc0fe20e1e8ad5
282
+ torch_tensorrt @ file:///opt/pytorch/torch_tensorrt/dist/torch_tensorrt-2.8.0a0-cp312-cp312-linux_x86_64.whl#sha256=26f4d7a04e7e6e017a5b917b8b98b83720bb2bb0195f48d25f52d567d286bfe1
283
+ torchao @ file:///opt/pytorch/ao
284
+ torchprofile==0.0.4
285
+ torchvision @ file:///opt/transfer/torchvision-0.22.0a0%2B95f10a4e-29946474-cp312-cp312-manylinux_2_28_x86_64.whl#sha256=746a3c674cc61680a86f4a14b374b21af1b5c21f629c42765b40c9494f6b234d
286
+ tornado==6.5.1
287
+ tqdm==4.67.1
288
+ traitlets==5.14.3
289
+ transformer_engine @ file:///pvcdata/aiinfra/b-aghosh/wheels/transformer_engine-2.10.0.dev0%2B10f6128d-cp312-cp312-linux_x86_64.whl#sha256=35cf5f4b83b2baad864e12e512bb422e0929409ac05610812fa521d4942d0189
290
+ treelite @ file:///rapids/treelite-4.4.1-py3-none-linux_x86_64.whl#sha256=174741724205540a1a5b37da81483aeff55d89548f42b3461cf07acac7ae96eb
291
+ typer==0.16.0
292
+ types-dataclasses==0.6.6
293
+ types-python-dateutil==2.9.0.20250516
294
+ typing-inspection==0.4.1
295
+ typing_extensions==4.14.0
296
+ tzdata @ file:///rapids/tzdata-2025.2-py2.py3-none-any.whl#sha256=1a403fada01ff9221ca8044d701868fa132215d84beb92242d9acd2147f667a8
297
+ ucx-py @ file:///rapids/ucx_py-0.43.0-cp312-cp312-linux_x86_64.whl#sha256=4a452fd8d8c983283ada3d49086e3657dc2ee94fbad08efdd708d15ab6e0b73a
298
+ ucxx @ file:///rapids/ucxx-0.43.0-cp312-cp312-linux_x86_64.whl#sha256=f1b4e55cb45c66fe88a2d8f9c2b87751165f3cfec21fbdc9c47c4ddff75e8a5f
299
+ -e git+https://github.com/Dots-Infra/UltraEP.git@94cab099b44fffa99a82fea99e7c12d89cf65e4f#egg=ultra_ep
300
+ uri-template==1.3.0
301
+ urllib3==2.0.7
302
+ wasabi==1.1.3
303
+ wcwidth==0.2.13
304
+ weasel==0.4.1
305
+ webcolors==24.11.1
306
+ webencodings==0.5.1
307
+ websocket-client==1.8.0
308
+ Werkzeug==3.1.3
309
+ wheel==0.45.1
310
+ wrapt==1.17.2
311
+ xdoctest==1.0.2
312
+ xgboost @ file:///rapids/xgboost-2.1.4-py3-none-manylinux_2_39_x86_64.whl#sha256=25eeca27f4048fc491940b3a0b4b5782ee48c33cb3aec059cef71f8da542930b
313
+ xxhash==4.0.1
314
+ yarl @ file:///rapids/yarl-1.20.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl#sha256=bf099e2432131093cc611623e0b0bcc399b8cddd9a91eded8bfb50402ec35018
315
+ zarr @ file:///rapids/zarr-2.18.7-py3-none-any.whl#sha256=ac3dc4033e9ae4e9d7b5e27c97ea3eaf1003cc0a07f010bd83d5134bf8c4b223
316
+ zict @ file:///rapids/zict-3.0.0-py2.py3-none-any.whl#sha256=5796e36bd0e0cc8cf0fbc1ace6a68912611c1dbd74750a3f3026b9b9d6a327ae
317
+ zipp==3.22.0
tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq4096__gbs16__cap2p40625/rep1/mem_samples.csv ADDED
@@ -0,0 +1,200 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 0, 0
2
+ 1, 0
3
+ 2, 0
4
+ 3, 0
5
+ 4, 0
6
+ 5, 0
7
+ 6, 0
8
+ 7, 0
9
+ 0, 0
10
+ 1, 0
11
+ 2, 0
12
+ 3, 0
13
+ 4, 0
14
+ 5, 0
15
+ 6, 0
16
+ 7, 0
17
+ 0, 2
18
+ 1, 1
19
+ 2, 1
20
+ 3, 1
21
+ 4, 1
22
+ 5, 1
23
+ 6, 1
24
+ 7, 1
25
+ 0, 4
26
+ 1, 4
27
+ 2, 4
28
+ 3, 4
29
+ 4, 1
30
+ 5, 1
31
+ 6, 1
32
+ 7, 1
33
+ 0, 4
34
+ 1, 4
35
+ 2, 4
36
+ 3, 4
37
+ 4, 4
38
+ 5, 4
39
+ 6, 4
40
+ 7, 1
41
+ 0, 4
42
+ 1, 4
43
+ 2, 4
44
+ 3, 4
45
+ 4, 4
46
+ 5, 4
47
+ 6, 4
48
+ 7, 4
49
+ 0, 4
50
+ 1, 629
51
+ 2, 629
52
+ 3, 629
53
+ 4, 629
54
+ 5, 629
55
+ 6, 629
56
+ 7, 629
57
+ 0, 919
58
+ 1, 935
59
+ 2, 941
60
+ 3, 919
61
+ 4, 955
62
+ 5, 939
63
+ 6, 947
64
+ 7, 951
65
+ 0, 1895
66
+ 1, 1895
67
+ 2, 1895
68
+ 3, 1895
69
+ 4, 1895
70
+ 5, 1895
71
+ 6, 1895
72
+ 7, 1895
73
+ 0, 2393
74
+ 1, 2401
75
+ 2, 2393
76
+ 3, 2387
77
+ 4, 2389
78
+ 5, 2383
79
+ 6, 2403
80
+ 7, 2391
81
+ 0, 14589
82
+ 1, 14587
83
+ 2, 14587
84
+ 3, 14587
85
+ 4, 14587
86
+ 5, 14587
87
+ 6, 14587
88
+ 7, 14587
89
+ 0, 15027
90
+ 1, 15025
91
+ 2, 15025
92
+ 3, 15025
93
+ 4, 15025
94
+ 5, 15025
95
+ 6, 15025
96
+ 7, 15025
97
+ 0, 16323
98
+ 1, 16321
99
+ 2, 16321
100
+ 3, 16321
101
+ 4, 16321
102
+ 5, 16321
103
+ 6, 16321
104
+ 7, 16321
105
+ 0, 16323
106
+ 1, 16321
107
+ 2, 16321
108
+ 3, 16321
109
+ 4, 16321
110
+ 5, 16321
111
+ 6, 16321
112
+ 7, 16321
113
+ 0, 16323
114
+ 1, 16321
115
+ 2, 16321
116
+ 3, 16321
117
+ 4, 16321
118
+ 5, 16321
119
+ 6, 16321
120
+ 7, 16321
121
+ 0, 16677
122
+ 1, 16675
123
+ 2, 16675
124
+ 3, 16675
125
+ 4, 16675
126
+ 5, 16675
127
+ 6, 16675
128
+ 7, 16675
129
+ 0, 26793
130
+ 1, 26791
131
+ 2, 26791
132
+ 3, 26791
133
+ 4, 26761
134
+ 5, 26791
135
+ 6, 26791
136
+ 7, 26791
137
+ 0, 26793
138
+ 1, 26791
139
+ 2, 26791
140
+ 3, 26791
141
+ 4, 26761
142
+ 5, 26791
143
+ 6, 26791
144
+ 7, 26791
145
+ 0, 26793
146
+ 1, 26791
147
+ 2, 26791
148
+ 3, 26791
149
+ 4, 26761
150
+ 5, 26791
151
+ 6, 26791
152
+ 7, 26791
153
+ 0, 28905
154
+ 1, 28903
155
+ 2, 28921
156
+ 3, 28903
157
+ 4, 28871
158
+ 5, 28937
159
+ 6, 28931
160
+ 7, 28935
161
+ 0, 30001
162
+ 1, 29971
163
+ 2, 30003
164
+ 3, 29989
165
+ 4, 29933
166
+ 5, 30005
167
+ 6, 30005
168
+ 7, 30005
169
+ 0, 31053
170
+ 1, 31025
171
+ 2, 31069
172
+ 3, 31043
173
+ 4, 30955
174
+ 5, 31071
175
+ 6, 31029
176
+ 7, 31043
177
+ 0, 38329
178
+ 1, 38317
179
+ 2, 38317
180
+ 3, 38317
181
+ 4, 38311
182
+ 5, 38315
183
+ 6, 38315
184
+ 7, 38315
185
+ 0, 38067
186
+ 1, 38055
187
+ 2, 38055
188
+ 3, 38055
189
+ 4, 38049
190
+ 5, 38053
191
+ 6, 38053
192
+ 7, 38053
193
+ 0, 3
194
+ 1, 4
195
+ 2, 4
196
+ 3, 4
197
+ 4, 4
198
+ 5, 196
199
+ 6, 35225
200
+ 7, 964
tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq4096__gbs16__cap2p40625/rep1/run.env ADDED
@@ -0,0 +1,70 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ backend=ultraep
2
+ config=PONLY
3
+ family=tierA
4
+ repo=/workspace/Megatron-UltraEP
5
+ branch=ultraep-bracing-regions-through-kernels
6
+ commit=03d57ef7
7
+ num_layers=16
8
+ hidden=4096
9
+ ffn_hidden=14336
10
+ moe_ffn_hidden=1536
11
+ num_experts=8
12
+ topk=2
13
+ experts_per_rank=1
14
+ expert_params=18874368
15
+ heads=32
16
+ kv_groups=8
17
+ kv_channels=
18
+ qk_layernorm=0
19
+ shared_ffn=0
20
+ shared_expert_overlap=0
21
+ moe_token_dispatcher=flex
22
+ mla=0
23
+ dense_first_layers=0
24
+ moe_layers=16
25
+ rotary_base=1000000
26
+ rotary_scaling=
27
+ qkv_bias=0
28
+ router_balancing=aux_loss
29
+ aux_loss_coeff=1e-2
30
+ router_pre_softmax=0
31
+ router_fusion=0
32
+ router_dtype=fp32
33
+ device_initiated_grouped_gemm=1
34
+ norm_epsilon=1e-5
35
+ seq_length=4096
36
+ micro_batch_size=1
37
+ global_batch_size=16
38
+ num_microbatches=2
39
+ assignments_per_rank=8192
40
+ mean_tokens_per_expert=8192
41
+ slot_policy=one
42
+ slots_per_rank=1
43
+ capacity_mode=static
44
+ capacity=2.40625
45
+ cuda_graph=1
46
+ probe_mode=0
47
+ async_grad_reduce=0
48
+ branch_role=timing
49
+ memory_dump_mode=none
50
+ memory_snapshot_path=
51
+ cuda_graph_warmup_steps=3
52
+ recompute=none
53
+ force_balance=0
54
+ train_iters=12
55
+ ep_size=8
56
+ echo_dispatch_overlap=1
57
+ echo_dispatch_recompute=0
58
+ ultraep_quota_min_tokens=1024
59
+ ultraep_weight_sync_join=
60
+ profile_enable=0
61
+ kernel_trace=0
62
+ fp8_mode=mxfp8
63
+ fp8_recipe=mxfp8
64
+ training_seed=1234
65
+ learning_rate=1e-4
66
+ min_learning_rate=1.0e-5
67
+ lr_warmup_iters=5
68
+ state_dump_enable=0
69
+ state_dump_dir=
70
+ moonep_assert_no_pending_reduce=0
tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq4096__gbs16__cap2p40625/rep1/train.log ADDED
@@ -0,0 +1,1031 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ==============================================================
2
+ PONLY / ultraep (slots/rank=1 policy=one)
3
+ E=8 K=2 H'=1536 m=1
4
+ L=16 seq=4096 gbs=16 (2 microbatches)
5
+ cap_mode=static cap=2.40625 cg=1 recompute=none
6
+ memory_dump=none snapshots=/workspace/tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq4096__gbs16__cap2p40625/rep1/memory_snapshots/snapshot.pickle
7
+ repo=/workspace/Megatron-UltraEP @ 03d57ef7
8
+ ==============================================================
9
+ W0830 07:15:36.047000 730290 torch/distributed/run.py:766]
10
+ W0830 07:15:36.047000 730290 torch/distributed/run.py:766] *****************************************
11
+ W0830 07:15:36.047000 730290 torch/distributed/run.py:766] Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed.
12
+ W0830 07:15:36.047000 730290 torch/distributed/run.py:766] *****************************************
13
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
14
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
15
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
16
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
17
+ using world size: 8, data-parallel size: 8, context-parallel size: 1, hierarchical context-parallel sizes: None, tensor-model-parallel size: 1, pipeline-model-parallel size: 1
18
+ WARNING: overriding default arguments for tokenizer_type:GPT2BPETokenizer with tokenizer_type:GPT2BPETokenizer
19
+ Number of virtual stages per pipeline stage: None
20
+ accumulate and all-reduce gradients in fp32 for bfloat16 data type.
21
+ using torch.bfloat16 for parameters ...
22
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
23
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
24
+ ------------------------ arguments ------------------------
25
+ account_for_embedding_in_pipeline_split ......... False
26
+ account_for_loss_in_pipeline_split .............. False
27
+ accumulate_allreduce_grads_in_fp32 .............. True
28
+ adam_beta1 ...................................... 0.9
29
+ adam_beta2 ...................................... 0.999
30
+ adam_eps ........................................ 1e-08
31
+ add_bias_linear ................................. False
32
+ add_position_embedding .......................... False
33
+ add_qkv_bias .................................... False
34
+ adlr_autoresume ................................. False
35
+ adlr_autoresume_interval ........................ 1000
36
+ align_grad_reduce ............................... True
37
+ align_param_gather .............................. False
38
+ app_tag_run_name ................................ None
39
+ app_tag_run_version ............................. 0.0.0
40
+ apply_layernorm_1p .............................. False
41
+ apply_query_key_layer_scaling ................... False
42
+ apply_residual_connection_post_layernorm ........ False
43
+ apply_rope_fusion ............................... True
44
+ async_save ...................................... None
45
+ async_tensor_model_parallel_allreduce ........... True
46
+ attention_backend ............................... AttnBackend.auto
47
+ attention_dropout ............................... 0.0
48
+ attention_softmax_in_fp32 ....................... False
49
+ auto_detect_ckpt_format ......................... False
50
+ barrier_with_L1_time ............................ True
51
+ bert_binary_head ................................ True
52
+ bert_embedder_type .............................. megatron
53
+ bert_load ....................................... None
54
+ bf16 ............................................ True
55
+ bias_dropout_fusion ............................. True
56
+ bias_gelu_fusion ................................ False
57
+ bias_swiglu_fusion .............................. True
58
+ biencoder_projection_dim ........................ 0
59
+ biencoder_shared_query_context_model ............ False
60
+ block_data_path ................................. None
61
+ calc_ft_timeouts ................................ False
62
+ calculate_per_token_loss ........................ False
63
+ check_for_large_grads ........................... False
64
+ check_for_nan_in_loss_and_grad .................. False
65
+ check_for_spiky_loss ............................ False
66
+ check_weight_hash_across_dp_replicas_interval ... None
67
+ ckpt_assume_constant_structure .................. False
68
+ ckpt_convert_format ............................. None
69
+ ckpt_convert_save ............................... None
70
+ ckpt_convert_update_legacy_dist_opt_format ...... False
71
+ ckpt_format ..................................... torch_dist
72
+ ckpt_fully_parallel_load ........................ False
73
+ ckpt_fully_parallel_save ........................ True
74
+ ckpt_fully_parallel_save_deprecated ............. False
75
+ ckpt_step ....................................... None
76
+ classes_fraction ................................ 1.0
77
+ clip_grad ....................................... 1.0
78
+ clone_scatter_output_in_embedding ............... True
79
+ config_logger_dir ...............................
80
+ consumed_train_samples .......................... 0
81
+ consumed_valid_samples .......................... 0
82
+ context_parallel_size ........................... 1
83
+ cp_comm_type .................................... ['p2p']
84
+ create_attention_mask_in_dataloader ............. True
85
+ cross_entropy_fusion_impl ....................... native
86
+ cross_entropy_loss_fusion ....................... False
87
+ cuda_graph_scope ................................ full_iteration
88
+ cuda_graph_warmup_steps ......................... 3
89
+ data_args_path .................................. None
90
+ data_cache_path ................................. None
91
+ data_parallel_random_init ....................... False
92
+ data_parallel_sharding_strategy ................. no_shard
93
+ data_parallel_size .............................. 8
94
+ data_path ....................................... ['/workspace/data/wikitext/wikitext103_gpt2_text_document']
95
+ data_per_class_fraction ......................... 1.0
96
+ data_sharding ................................... True
97
+ dataloader_type ................................. single
98
+ ddp_average_in_collective ....................... False
99
+ ddp_bucket_size ................................. None
100
+ ddp_num_buckets ................................. None
101
+ ddp_pad_buckets_for_high_nccl_busbw ............. False
102
+ decoder_first_pipeline_num_layers ............... None
103
+ decoder_last_pipeline_num_layers ................ None
104
+ decoder_num_layers .............................. None
105
+ decoder_seq_length .............................. None
106
+ decoupled_lr .................................... None
107
+ decoupled_min_lr ................................ None
108
+ decrease_batch_size_if_needed ................... False
109
+ defer_embedding_wgrad_compute ................... False
110
+ delay_wgrad_compute ............................. False
111
+ deprecated_use_mcore_models ..................... True
112
+ deterministic_mode .............................. False
113
+ dino_bottleneck_size ............................ 256
114
+ dino_freeze_last_layer .......................... 1
115
+ dino_head_hidden_size ........................... 2048
116
+ dino_local_crops_number ......................... 10
117
+ dino_local_img_size ............................. 96
118
+ dino_norm_last_layer ............................ False
119
+ dino_teacher_temp ............................... 0.07
120
+ dino_warmup_teacher_temp ........................ 0.04
121
+ dino_warmup_teacher_temp_epochs ................. 30
122
+ disable_bf16_reduced_precision_matmul ........... False
123
+ disable_mamba_mem_eff_path ...................... False
124
+ disable_straggler_on_startup .................... False
125
+ dist_ckpt_format_deprecated ..................... None
126
+ dist_ckpt_strictness ............................ assume_ok_unexpected
127
+ distribute_saved_activations .................... False
128
+ distributed_backend ............................. nccl
129
+ distributed_timeout_minutes ..................... 10
130
+ embedding_init_method_std ....................... None
131
+ embedding_path .................................. None
132
+ empty_unused_memory_level ....................... 0
133
+ enable_cuda_graph ............................... True
134
+ enable_experimental ............................. False
135
+ enable_ft_package ............................... False
136
+ enable_gloo_process_groups ...................... True
137
+ enable_msc ...................................... True
138
+ enable_one_logger ............................... True
139
+ encoder_num_layers .............................. 16
140
+ encoder_seq_length .............................. 4096
141
+ end_weight_decay ................................ 0.1
142
+ eod_mask_loss ................................... False
143
+ error_injection_rate ............................ 0
144
+ error_injection_type ............................ transient_error
145
+ eval_interval ................................... 1000000
146
+ eval_iters ...................................... 0
147
+ evidence_data_path .............................. None
148
+ exit_duration_in_mins ........................... None
149
+ exit_interval ................................... None
150
+ exit_on_missing_checkpoint ...................... False
151
+ exit_signal_handler ............................. False
152
+ exp_avg_dtype ................................... torch.float32
153
+ exp_avg_sq_dtype ................................ torch.float32
154
+ expert_model_parallel_size ...................... 8
155
+ expert_tensor_parallel_size ..................... 1
156
+ export_force_local_attention .................... False
157
+ export_kd_cfg ................................... None
158
+ export_kd_teacher_ckpt_format ................... None
159
+ export_kd_teacher_load .......................... None
160
+ export_kv_cache_quant ........................... False
161
+ export_legacy_megatron .......................... False
162
+ export_model_type ............................... GPTModel
163
+ export_moe_apply_probs_on_input ................. False
164
+ export_qk_l2_norm ............................... False
165
+ export_quant_cfg ................................ None
166
+ export_real_quant_cfg ........................... None
167
+ export_te_mcore_model ........................... False
168
+ external_cuda_graph ............................. False
169
+ ffn_hidden_size ................................. 14336
170
+ finetune ........................................ False
171
+ finetune_data_split ............................. train
172
+ finetune_hf_dataset ............................. None
173
+ first_last_layers_bf16 .......................... False
174
+ flash_decode .................................... False
175
+ fp16 ............................................ False
176
+ fp16_lm_cross_entropy ........................... False
177
+ fp32_residual_connection ........................ False
178
+ fp8 ............................................. e4m3
179
+ fp8_amax_compute_algo ........................... most_recent
180
+ fp8_amax_history_len ............................ 1
181
+ fp8_interval .................................... 1
182
+ fp8_margin ...................................... 0
183
+ fp8_param_gather ................................ False
184
+ fp8_recipe ...................................... mxfp8
185
+ fp8_wgrad ....................................... True
186
+ fsdp_double_buffer .............................. False
187
+ global_batch_size ............................... 16
188
+ grad_reduce_in_bf16 ............................. False
189
+ gradient_accumulation_fusion .................... True
190
+ gradient_reduce_div_fusion ...................... True
191
+ group_query_attention ........................... True
192
+ head_lr_mult .................................... 1.0
193
+ heterogeneous_layers_config_encoded_json ........ None
194
+ heterogeneous_layers_config_path ................ None
195
+ hidden_dropout .................................. 0.0
196
+ hidden_size ..................................... 4096
197
+ hierarchical_context_parallel_sizes ............. None
198
+ high_priority_stream_groups ..................... []
199
+ hybrid_attention_ratio .......................... 0.0
200
+ hybrid_mlp_ratio ................................ 0.0
201
+ hybrid_override_pattern ......................... None
202
+ hysteresis ...................................... 2
203
+ ict_head_size ................................... None
204
+ ict_load ........................................ None
205
+ img_h ........................................... 224
206
+ img_w ........................................... 224
207
+ indexer_batch_size .............................. 128
208
+ indexer_log_interval ............................ 1000
209
+ inference_batch_times_seqlen_threshold .......... -1
210
+ inference_dynamic_batching ...................... False
211
+ inference_dynamic_batching_buffer_guaranteed_fraction 0.2
212
+ inference_dynamic_batching_buffer_overflow_factor None
213
+ inference_dynamic_batching_buffer_size_gb ....... 40.0
214
+ inference_dynamic_batching_chunk_size ........... 256
215
+ inference_dynamic_batching_max_requests_override None
216
+ inference_dynamic_batching_max_tokens_override .. None
217
+ inference_dynamic_batching_num_cuda_graphs ...... 16
218
+ inference_max_batch_size ........................ 8
219
+ inference_max_seq_length ........................ 2560
220
+ inference_rng_tracker ........................... False
221
+ init_method_std ................................. 0.01
222
+ init_method_xavier_uniform ...................... False
223
+ init_model_with_meta_device ..................... False
224
+ initial_loss_scale .............................. 4294967296
225
+ inprocess_active_world_size ..................... 8
226
+ inprocess_barrier_timeout ....................... 120
227
+ inprocess_completion_timeout .................... 120
228
+ inprocess_empty_cuda_cache ...................... False
229
+ inprocess_granularity ........................... node
230
+ inprocess_hard_timeout .......................... 90
231
+ inprocess_heartbeat_interval .................... 30
232
+ inprocess_heartbeat_timeout ..................... 60
233
+ inprocess_last_call_wait ........................ 1
234
+ inprocess_max_iterations ........................ None
235
+ inprocess_monitor_process_interval .............. 1.0
236
+ inprocess_monitor_thread_interval ............... 1.0
237
+ inprocess_progress_watchdog_interval ............ 1.0
238
+ inprocess_restart ............................... False
239
+ inprocess_soft_timeout .......................... 60
240
+ inprocess_termination_grace_time ................ 1
241
+ is_hybrid_model ................................. False
242
+ iter_per_epoch .................................. 1250
243
+ iterations_to_skip .............................. []
244
+ keep_fp8_transpose_cache_when_using_custom_fsdp . False
245
+ kitchen_config_file ............................. None
246
+ kitchen_recipe_number ........................... None
247
+ kv_channels ..................................... 128
248
+ kv_lora_rank .................................... 32
249
+ lazy_mpu_init ................................... None
250
+ load ............................................ None
251
+ load_main_params_from_ckpt ...................... None
252
+ load_model_opt_format ........................... False
253
+ local_rank ...................................... 0
254
+ log_energy ...................................... False
255
+ log_interval .................................... 1
256
+ log_loss_scale_to_tensorboard ................... True
257
+ log_memory_to_tensorboard ....................... False
258
+ log_num_zeros_in_grad ........................... False
259
+ log_params_norm ................................. False
260
+ log_progress .................................... False
261
+ log_straggler ................................... False
262
+ log_throughput .................................. True
263
+ log_timers_to_tensorboard ....................... False
264
+ log_validation_ppl_to_tensorboard ............... False
265
+ log_world_size_to_tensorboard ................... False
266
+ logging_level ................................... None
267
+ loss_scale ...................................... None
268
+ loss_scale_window ............................... 1000
269
+ lr .............................................. 0.0001
270
+ lr_decay_iters .................................. 8
271
+ lr_decay_samples ................................ None
272
+ lr_decay_style .................................. cosine
273
+ lr_warmup_fraction .............................. None
274
+ lr_warmup_init .................................. 0.0
275
+ lr_warmup_iters ................................. 5
276
+ lr_warmup_samples ............................... 0
277
+ lr_wsd_decay_iters .............................. None
278
+ lr_wsd_decay_samples ............................ None
279
+ lr_wsd_decay_style .............................. exponential
280
+ main_grads_dtype ................................ torch.float32
281
+ main_params_dtype ............................... torch.float32
282
+ make_vocab_size_divisible_by .................... 128
283
+ mamba_head_dim .................................. 64
284
+ mamba_num_groups ................................ 8
285
+ mamba_num_heads ................................. None
286
+ mamba_state_dim ................................. 128
287
+ manual_gc ....................................... False
288
+ manual_gc_eval .................................. True
289
+ manual_gc_interval .............................. 0
290
+ mask_factor ..................................... 1.0
291
+ mask_prob ....................................... 0.15
292
+ mask_type ....................................... random
293
+ masked_softmax_fusion ........................... False
294
+ max_position_embeddings ......................... 32768
295
+ max_tokens_to_oom ............................... 12000
296
+ memory_snapshot_path ............................ snapshot.pickle
297
+ merge_file ...................................... /workspace/data/wikitext/gpt2-merges.txt
298
+ micro_batch_size ................................ 1
299
+ microbatch_group_size_per_vp_stage .............. None
300
+ mid_level_dataset_surplus ....................... 0.005
301
+ min_loss_scale .................................. 1.0
302
+ min_lr .......................................... 1e-05
303
+ mlp_chunks_for_prefill .......................... 1
304
+ mmap_bin_files .................................. True
305
+ mock_data ....................................... False
306
+ moe_apply_probs_on_input ........................ False
307
+ moe_aux_loss_coeff .............................. 0.01
308
+ moe_capacity_monitor ............................ True
309
+ moe_deepep_num_sms .............................. 20
310
+ moe_echo_enable_random_offloading ............... False
311
+ moe_echo_expert_dispatch_overlap ................ False
312
+ moe_echo_expert_dispatcher_type ................. hybridep
313
+ moe_echo_recompute_expert_dispatch .............. False
314
+ moe_enable_deepep ............................... False
315
+ moe_enable_echo ................................. False
316
+ moe_enable_hybridep ............................. True
317
+ moe_enable_ultraep .............................. True
318
+ moe_expert_capacity_factor ...................... None
319
+ moe_extended_tp ................................. False
320
+ moe_ffn_hidden_size ............................. 1536
321
+ moe_grouped_gemm ................................ True
322
+ moe_hybridep_num_sms ............................ 16
323
+ moe_input_jitter_eps ............................ None
324
+ moe_layer_freq .................................. 1
325
+ moe_layer_recompute ............................. False
326
+ moe_num_echo_experts ............................ None
327
+ moe_num_redundant_experts_per_rank .............. 1
328
+ moe_pad_expert_input_to_capacity ................ False
329
+ moe_per_layer_logging ........................... False
330
+ moe_permute_fusion .............................. False
331
+ moe_received_token_capacity ..................... 2.40625
332
+ moe_router_bias_update_rate ..................... 0.001
333
+ moe_router_dtype ................................ fp32
334
+ moe_router_enable_expert_bias ................... False
335
+ moe_router_force_load_balancing ................. False
336
+ moe_router_group_topk ........................... None
337
+ moe_router_load_balancing_type .................. aux_loss
338
+ moe_router_num_groups ........................... None
339
+ moe_router_padding_for_fp8 ...................... False
340
+ moe_router_pre_softmax .......................... False
341
+ moe_router_score_function ....................... softmax
342
+ moe_router_topk ................................. 2
343
+ moe_router_topk_scaling_factor .................. None
344
+ moe_shared_expert_intermediate_size ............. None
345
+ moe_shared_expert_overlap ....................... False
346
+ moe_token_dispatcher_type ....................... flex
347
+ moe_token_drop_policy ........................... probs
348
+ moe_ultraep_max_microbatches .................... None
349
+ moe_ultraep_overlap_weight_sync ................. True
350
+ moe_ultraep_weight_sync_join .................... before_dispatch
351
+ moe_upcycling_granularity ....................... 1
352
+ moe_use_device_initiated_grouped_gemm ........... True
353
+ moe_use_legacy_grouped_gemm ..................... False
354
+ moe_use_upcycling ............................... False
355
+ moe_z_loss_coeff ................................ None
356
+ mrope_section ................................... None
357
+ mscale .......................................... 1.0
358
+ mscale_all_dim .................................. 0.0
359
+ mtp_loss_scaling_factor ......................... 0.1
360
+ mtp_num_layers .................................. None
361
+ multi_latent_attention .......................... False
362
+ nccl_all_reduce_for_prefill ..................... False
363
+ nccl_communicator_config_path ................... None
364
+ nccl_ub ......................................... False
365
+ no_load_optim ................................... None
366
+ no_load_rng ..................................... None
367
+ no_persist_layer_norm ........................... False
368
+ no_rope_freq .................................... None
369
+ no_save_optim ................................... None
370
+ no_save_rng ..................................... None
371
+ non_persistent_ckpt_type ........................ None
372
+ non_persistent_global_ckpt_dir .................. None
373
+ non_persistent_local_ckpt_algo .................. fully_parallel
374
+ non_persistent_local_ckpt_dir ................... None
375
+ non_persistent_save_interval .................... None
376
+ norm_epsilon .................................... 1e-05
377
+ normalization ................................... RMSNorm
378
+ num_attention_heads ............................. 32
379
+ num_channels .................................... 3
380
+ num_classes ..................................... 1000
381
+ num_dataset_builder_threads ..................... 1
382
+ num_distributed_optimizer_instances ............. 1
383
+ num_experts ..................................... 8
384
+ num_layers ...................................... 16
385
+ num_layers_at_end_in_bf16 ....................... 1
386
+ num_layers_at_start_in_bf16 ..................... 1
387
+ num_layers_per_virtual_pipeline_stage ........... None
388
+ num_query_groups ................................ 8
389
+ num_virtual_stages_per_pipeline_rank ............ None
390
+ num_workers ..................................... 2
391
+ object_storage_cache_path ....................... None
392
+ one_logger_async ................................ False
393
+ one_logger_project .............................. megatron-lm
394
+ one_logger_run_name ............................. None
395
+ onnx_safe ....................................... None
396
+ openai_gelu ..................................... False
397
+ optimizer ....................................... adam
398
+ optimizer_cpu_offload ........................... False
399
+ optimizer_offload_fraction ...................... 1.0
400
+ output_bert_embeddings .......................... False
401
+ overlap_cpu_optimizer_d2h_h2d ................... False
402
+ overlap_grad_reduce ............................. True
403
+ overlap_moe_expert_parallel_comm ................ False
404
+ overlap_p2p_comm ................................ False
405
+ overlap_p2p_comm_warmup_flush ................... False
406
+ overlap_param_gather ............................ True
407
+ overlap_param_gather_with_optimizer_step ........ False
408
+ override_opt_param_scheduler .................... False
409
+ params_dtype .................................... torch.bfloat16
410
+ patch_dim ....................................... 16
411
+ per_split_data_args_path ........................ None
412
+ perform_initialization .......................... True
413
+ pin_cpu_grads ................................... True
414
+ pin_cpu_params .................................. True
415
+ pipeline_model_parallel_comm_backend ............ None
416
+ pipeline_model_parallel_layout .................. None
417
+ pipeline_model_parallel_size .................... 1
418
+ position_embedding_type ......................... rope
419
+ pretrained_checkpoint ........................... None
420
+ profile ......................................... False
421
+ profile_ranks ................................... [0]
422
+ profile_step_end ................................ 12
423
+ profile_step_start .............................. 10
424
+ q_lora_rank ..................................... None
425
+ qk_head_dim ..................................... 128
426
+ qk_l2_norm ...................................... False
427
+ qk_layernorm .................................... False
428
+ qk_pos_emb_head_dim ............................. 64
429
+ query_in_block_prob ............................. 0.1
430
+ rampup_batch_size ............................... None
431
+ rank ............................................ 0
432
+ recompute_granularity ........................... None
433
+ recompute_method ................................ None
434
+ recompute_modules ............................... None
435
+ recompute_num_layers ............................ None
436
+ record_memory_history ........................... False
437
+ relative_attention_max_distance ................. 128
438
+ relative_attention_num_buckets .................. 32
439
+ replication ..................................... False
440
+ replication_factor .............................. 2
441
+ replication_jump ................................ None
442
+ rerun_mode ...................................... validate_results
443
+ reset_attention_mask ............................ False
444
+ reset_position_ids .............................. False
445
+ result_rejected_tracker_filename ................ None
446
+ retriever_report_topk_accuracies ................ []
447
+ retriever_score_scaling ......................... False
448
+ retriever_seq_length ............................ 256
449
+ retro_add_retriever ............................. False
450
+ retro_attention_gate ............................ 1
451
+ retro_cyclic_train_iters ........................ None
452
+ retro_encoder_attention_dropout ................. 0.1
453
+ retro_encoder_hidden_dropout .................... 0.1
454
+ retro_encoder_layers ............................ 2
455
+ retro_num_neighbors ............................. 2
456
+ retro_num_retrieved_chunks ...................... 2
457
+ retro_project_dir ............................... None
458
+ retro_verify_neighbor_count ..................... True
459
+ reuse_grad_buf_for_mxfp8_param_ag ............... False
460
+ rope_scaling_factor ............................. 8.0
461
+ rope_type ....................................... None
462
+ rotary_base ..................................... 1000000
463
+ rotary_interleaved .............................. False
464
+ rotary_percent .................................. 1.0
465
+ rotary_scaling_factor ........................... 1.0
466
+ rotary_seq_len_interpolation_factor ............. None
467
+ run_workload_inspector_server ................... False
468
+ sample_rate ..................................... 1.0
469
+ save ............................................ None
470
+ save_interval ................................... None
471
+ save_retain_interval ............................ None
472
+ scatter_gather_tensors_in_pipeline .............. True
473
+ seed ............................................ 1234
474
+ seq_length ...................................... 4096
475
+ sequence_parallel ............................... False
476
+ sft ............................................. False
477
+ sft_tokenizer_prompt_format ..................... nemotron-h-aligned
478
+ sgd_momentum .................................... 0.9
479
+ short_seq_prob .................................. 0.1
480
+ skip_train ...................................... False
481
+ skipped_train_samples ........................... 0
482
+ spec ............................................ None
483
+ split ........................................... 100,0,0
484
+ squared_relu .................................... False
485
+ start_weight_decay .............................. 0.1
486
+ straggler_ctrlr_port ............................ 65535
487
+ straggler_minmax_count .......................... 1
488
+ suggested_communication_unit_size ............... None
489
+ swiglu .......................................... True
490
+ swin_backbone_type .............................. tiny
491
+ symmetric_ar_type ............................... None
492
+ te_rng_tracker .................................. True
493
+ tensor_model_parallel_size ...................... 1
494
+ tensorboard_dir ................................. None
495
+ tensorboard_log_interval ........................ 1
496
+ tensorboard_queue_size .......................... 1000
497
+ test_data_path .................................. None
498
+ test_mode ....................................... False
499
+ tiktoken_num_special_tokens ..................... 1000
500
+ tiktoken_pattern ................................ None
501
+ tiktoken_special_tokens ......................... None
502
+ timing_log_level ................................ 0
503
+ timing_log_option ............................... minmax
504
+ titles_data_path ................................ None
505
+ tokenizer_model ................................. None
506
+ tokenizer_type .................................. GPT2BPETokenizer
507
+ torch_fsdp2_reshard_after_forward ............... True
508
+ tp_comm_bootstrap_backend ....................... nccl
509
+ tp_comm_bulk_dgrad .............................. True
510
+ tp_comm_bulk_wgrad .............................. True
511
+ tp_comm_overlap ................................. False
512
+ tp_comm_overlap_ag .............................. True
513
+ tp_comm_overlap_cfg ............................. None
514
+ tp_comm_overlap_rs .............................. True
515
+ tp_comm_overlap_rs_dgrad ........................ False
516
+ tp_comm_split_ag ................................ True
517
+ tp_comm_split_rs ................................ True
518
+ train_data_path ................................. None
519
+ train_iters ..................................... 12
520
+ train_samples ................................... None
521
+ train_sync_interval ............................. None
522
+ transformer_impl ................................ transformer_engine
523
+ transformer_pipeline_model_parallel_size ........ 1
524
+ untie_embeddings_and_output_weights ............. True
525
+ use_checkpoint_args ............................. False
526
+ use_checkpoint_opt_param_scheduler .............. False
527
+ use_cpu_initialization .......................... None
528
+ use_custom_fsdp ................................. False
529
+ use_dist_ckpt ................................... True
530
+ use_dist_ckpt_deprecated ........................ False
531
+ use_distributed_optimizer ....................... True
532
+ use_flash_attn .................................. False
533
+ use_legacy_models ............................... False
534
+ use_mp_args_from_checkpoint_args ................ False
535
+ use_one_sent_docs ............................... False
536
+ use_persistent_ckpt_worker ...................... False
537
+ use_precision_aware_optimizer ................... False
538
+ use_pytorch_profiler ............................ False
539
+ use_ring_exchange_p2p ........................... False
540
+ use_rope_scaling ................................ False
541
+ use_rotary_position_embeddings .................. False
542
+ use_sharp ....................................... False
543
+ use_tokenizer_model_from_checkpoint_args ........ True
544
+ use_torch_fsdp2 ................................. False
545
+ use_torch_optimizer_for_cpu_offload ............. False
546
+ use_tp_pp_dp_mapping ............................ False
547
+ v_head_dim ...................................... 128
548
+ valid_data_path ................................. None
549
+ variable_seq_lengths ............................ False
550
+ virtual_pipeline_model_parallel_size ............ None
551
+ vision_backbone_type ............................ vit
552
+ vision_pretraining .............................. False
553
+ vision_pretraining_type ......................... classify
554
+ vocab_extra_ids ................................. 0
555
+ vocab_file ...................................... /workspace/data/wikitext/gpt2-vocab.json
556
+ vocab_size ...................................... None
557
+ wandb_exp_name ..................................
558
+ wandb_project ...................................
559
+ wandb_save_dir ..................................
560
+ weight_decay .................................... 0.1
561
+ weight_decay_incr_style ......................... constant
562
+ wgrad_deferral_limit ............................ 0
563
+ world_size ...................................... 8
564
+ yaml_cfg ........................................ None
565
+ -------------------- end of arguments ---------------------
566
+ INFO:megatron.core.num_microbatches_calculator:setting number of microbatches to constant 2
567
+ > building GPT2BPETokenizer tokenizer ...
568
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
569
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
570
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
571
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
572
+ > padded vocab (size: 50257) with 47 dummy tokens (new size: 50304)
573
+ WARNING:megatron.core.rerun_state_machine:RerunStateMachine initialized in mode RerunMode.VALIDATE_RESULTS
574
+ > initializing torch distributed ...
575
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
576
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
577
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
578
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
579
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
580
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
581
+ WARNING: one_logger package is required to enable e2e metrics tracking. please go to https://confluence.nvidia.com/display/MLWFO/Package+Repositories for details to install it
582
+ [Gloo] Rank 0 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
583
+ [Gloo] Rank 1 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
584
+ [Gloo] Rank 2 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
585
+ [Gloo] Rank 6 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
586
+ [Gloo] Rank 4 is connected to 7 peer ranks. Expected number of connected peer ranks is : [Gloo] Rank 7
587
+ 3 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
588
+ [Gloo] Rank 5 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
589
+ [Gloo] Rank 7 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
590
+ [Gloo] Rank 1 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
591
+ [Gloo] Rank 0 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
592
+ [Gloo] Rank 2 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
593
+ [Gloo] Rank 3 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
594
+ [Gloo] Rank 4 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
595
+ [Gloo] Rank 5 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
596
+ [Gloo] Rank [Gloo] Rank 76 is connected to is connected to 77 peer ranks. peer ranks. Expected number of connected peer ranks is : Expected number of connected peer ranks is : 77
597
+
598
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
599
+ > initialized tensor model parallel with size 1
600
+ > initialized pipeline model parallel with size 1
601
+ > setting random seeds to 1234 ...
602
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
603
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
604
+ [Gloo] Rank 0 is connected to 0[Gloo] Rank peer ranks. 0Expected number of connected peer ranks is : is connected to [Gloo] Rank 000
605
+ peer ranks. is connected to Expected number of connected peer ranks is : 00 peer ranks.
606
+ Expected number of connected peer ranks is : 0
607
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
608
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
609
+ [rank7]:[W830 07:15:54.463087662 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 7] using GPU 7 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
610
+ [rank1]:[W830 07:15:54.463201588 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 1] using GPU 1 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
611
+ [rank4]:[W830 07:15:54.465606210 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 4] using GPU 4 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
612
+ [rank5]:[W830 07:15:54.465907978 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 5] using GPU 5 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
613
+ [rank3]:[W830 07:15:54.465929999 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 3] using GPU 3 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
614
+ [rank6]:[W830 07:15:54.465978742 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 6] using GPU 6 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
615
+ [rank2]:[W830 07:15:54.467170940 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 2] using GPU 2 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
616
+ > compiling dataset index builder ...
617
+ make: Entering directory '/workspace/Megatron-UltraEP/megatron/core/datasets'
618
+ make: Nothing to be done for 'default'.
619
+ make: Leaving directory '/workspace/Megatron-UltraEP/megatron/core/datasets'
620
+ >>> done with dataset index builder. Compilation time: 0.036 seconds
621
+ WARNING: constraints for invoking optimized fused softmax kernel are not met. We default back to unfused kernel invocations.
622
+ > compiling and loading fused kernels ...
623
+ [rank0]:[W830 07:15:54.641234348 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 0] using GPU 0 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
624
+ >>> done with compiling and loading fused kernels. Compilation time: 3.822 seconds
625
+ time to initialize megatron (seconds): 7.345
626
+ [after megatron is initialized] datetime: 2026-08-30 07:16:00
627
+ building GPT model ...
628
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
629
+ warnings.warn(
630
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
631
+ warnings.warn(
632
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
633
+ warnings.warn(
634
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
635
+ warnings.warn(
636
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
637
+ warnings.warn(
638
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
639
+ warnings.warn(
640
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
641
+ warnings.warn(
642
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
643
+ warnings.warn(
644
+ [INFO] Use MNNVL fabric: False
645
+ [WARN] MAX_NUM_NVL_PEERS is not set. Using detected value 8.
646
+ NVSHMEM v3.7.2
647
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
648
+
649
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
650
+
651
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
652
+
653
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
654
+
655
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
656
+
657
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
658
+
659
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
660
+
661
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
662
+
663
+ UltraEP: NVL domain size <= 8; forcing weight_sync plan mode to direct (was adaptive).
664
+ > number of parameters on (tensor, pipeline) model parallel rank (0, 0): 1687818240
665
+ INFO:megatron.core.distributed.distributed_data_parallel:Setting up DistributedDataParallel with config DistributedDataParallelConfig(grad_reduce_in_fp32=True, overlap_grad_reduce=True, overlap_param_gather=True, align_param_gather=False, use_distributed_optimizer=True, num_distributed_optimizer_instances=1, check_for_nan_in_grad=False, check_for_large_grads=False, bucket_size=40000000, pad_buckets_for_high_nccl_busbw=False, average_in_collective=False, fp8_param_gather=False, reuse_grad_buf_for_mxfp8_param_ag=False, use_custom_fsdp=False, data_parallel_sharding_strategy='no_shard', gradient_reduce_div_fusion=True, suggested_communication_unit_size=None, preserve_fp32_weights=True, keep_fp8_transpose_cache_when_using_custom_fsdp=False, nccl_ub=False, fsdp_double_buffer=False)
666
+ INFO:megatron.core.distributed.param_and_grad_buffer:Number of buckets for gradient all-reduce / reduce-scatter: 18
667
+ Params for bucket 1 (206045184 elements, 206045184 padded size):
668
+ module.output_layer.weight
669
+ Params for bucket 2 (41988096 elements, 41988096 padded size):
670
+ module.decoder.layers.15.self_attention.linear_qkv.weight
671
+ module.decoder.layers.15.pre_mlp_layernorm.weight
672
+ module.decoder.final_layernorm.weight
673
+ module.decoder.layers.15.self_attention.linear_qkv.layer_norm_weight
674
+ module.decoder.layers.15.mlp.router.weight
675
+ module.decoder.layers.15.self_attention.linear_proj.weight
676
+ Params for bucket 3 (41984000 elements, 41984000 padded size):
677
+ module.decoder.layers.14.pre_mlp_layernorm.weight
678
+ module.decoder.layers.14.self_attention.linear_qkv.weight
679
+ module.decoder.layers.14.mlp.router.weight
680
+ module.decoder.layers.14.self_attention.linear_proj.weight
681
+ module.decoder.layers.14.self_attention.linear_qkv.layer_norm_weight
682
+ Params for bucket 4 (41984000 elements, 41984000 padded size):
683
+ module.decoder.layers.13.self_attention.linear_qkv.weight
684
+ module.decoder.layers.13.pre_mlp_layernorm.weight
685
+ module.decoder.layers.13.mlp.router.weight
686
+ module.decoder.layers.13.self_attention.linear_qkv.layer_norm_weight
687
+ module.decoder.layers.13.self_attention.linear_proj.weight
688
+ Params for bucket 5 (41984000 elements, 41984000 padded size):
689
+ module.decoder.layers.12.pre_mlp_layernorm.weight
690
+ module.decoder.layers.12.self_attention.linear_qkv.weight
691
+ module.decoder.layers.12.mlp.router.weight
692
+ module.decoder.layers.12.self_attention.linear_proj.weight
693
+ module.decoder.layers.12.self_attention.linear_qkv.layer_norm_weight
694
+ Params for bucket 6 (41984000 elements, 41984000 padded size):
695
+ module.decoder.layers.11.self_attention.linear_qkv.weight
696
+ module.decoder.layers.11.pre_mlp_layernorm.weight
697
+ module.decoder.layers.11.mlp.router.weight
698
+ module.decoder.layers.11.self_attention.linear_qkv.layer_norm_weight
699
+ module.decoder.layers.11.self_attention.linear_proj.weight
700
+ Params for bucket 7 (41984000 elements, 41984000 padded size):
701
+ module.decoder.layers.10.self_attention.linear_qkv.weight
702
+ module.decoder.layers.10.pre_mlp_layernorm.weight
703
+ module.decoder.layers.10.mlp.router.weight
704
+ module.decoder.layers.10.self_attention.linear_proj.weight
705
+ module.decoder.layers.10.self_attention.linear_qkv.layer_norm_weight
706
+ Params for bucket 8 (41984000 elements, 41984000 padded size):
707
+ module.decoder.layers.9.pre_mlp_layernorm.weight
708
+ module.decoder.layers.9.self_attention.linear_qkv.weight
709
+ module.decoder.layers.9.mlp.router.weight
710
+ module.decoder.layers.9.self_attention.linear_proj.weight
711
+ module.decoder.layers.9.self_attention.linear_qkv.layer_norm_weight
712
+ Params for bucket 9 (41984000 elements, 41984000 padded size):
713
+ module.decoder.layers.8.self_attention.linear_qkv.weight
714
+ module.decoder.layers.8.pre_mlp_layernorm.weight
715
+ module.decoder.layers.8.mlp.router.weight
716
+ module.decoder.layers.8.self_attention.linear_qkv.layer_norm_weight
717
+ module.decoder.layers.8.self_attention.linear_proj.weight
718
+ Params for bucket 10 (41984000 elements, 41984000 padded size):
719
+ module.decoder.layers.7.pre_mlp_layernorm.weight
720
+ module.decoder.layers.7.self_attention.linear_qkv.weight
721
+ module.decoder.layers.7.mlp.router.weight
722
+ module.decoder.layers.7.self_attention.linear_proj.weight
723
+ module.decoder.layers.7.self_attention.linear_qkv.layer_norm_weight
724
+ Params for bucket 11 (41984000 elements, 41984000 padded size):
725
+ module.decoder.layers.6.mlp.router.weight
726
+ module.decoder.layers.6.self_attention.linear_proj.weight
727
+ module.decoder.layers.6.self_attention.linear_qkv.layer_norm_weight
728
+ module.decoder.layers.6.pre_mlp_layernorm.weight
729
+ module.decoder.layers.6.self_attention.linear_qkv.weight
730
+ Params for bucket 12 (41984000 elements, 41984000 padded size):
731
+ module.decoder.layers.5.mlp.router.weight
732
+ module.decoder.layers.5.self_attention.linear_qkv.layer_norm_weight
733
+ module.decoder.layers.5.self_attention.linear_qkv.weight
734
+ module.decoder.layers.5.self_attention.linear_proj.weight
735
+ module.decoder.layers.5.pre_mlp_layernorm.weight
736
+ Params for bucket 13 (41984000 elements, 41984000 padded size):
737
+ module.decoder.layers.4.mlp.router.weight
738
+ module.decoder.layers.4.self_attention.linear_proj.weight
739
+ module.decoder.layers.4.self_attention.linear_qkv.layer_norm_weight
740
+ module.decoder.layers.4.pre_mlp_layernorm.weight
741
+ module.decoder.layers.4.self_attention.linear_qkv.weight
742
+ Params for bucket 14 (41984000 elements, 41984000 padded size):
743
+ module.decoder.layers.3.mlp.router.weight
744
+ module.decoder.layers.3.self_attention.linear_qkv.layer_norm_weight
745
+ module.decoder.layers.3.self_attention.linear_proj.weight
746
+ module.decoder.layers.3.self_attention.linear_qkv.weight
747
+ module.decoder.layers.3.pre_mlp_layernorm.weight
748
+ Params for bucket 15 (41984000 elements, 41984000 padded size):
749
+ module.decoder.layers.2.self_attention.linear_qkv.weight
750
+ module.decoder.layers.2.pre_mlp_layernorm.weight
751
+ module.decoder.layers.2.mlp.router.weight
752
+ module.decoder.layers.2.self_attention.linear_qkv.layer_norm_weight
753
+ module.decoder.layers.2.self_attention.linear_proj.weight
754
+ Params for bucket 16 (41984000 elements, 41984000 padded size):
755
+ module.decoder.layers.1.mlp.router.weight
756
+ module.decoder.layers.1.self_attention.linear_proj.weight
757
+ module.decoder.layers.1.self_attention.linear_qkv.layer_norm_weight
758
+ module.decoder.layers.1.pre_mlp_layernorm.weight
759
+ module.decoder.layers.1.self_attention.linear_qkv.weight
760
+ Params for bucket 17 (41984000 elements, 41984000 padded size):
761
+ module.decoder.layers.0.mlp.router.weight
762
+ module.decoder.layers.0.self_attention.linear_qkv.layer_norm_weight
763
+ module.decoder.layers.0.self_attention.linear_proj.weight
764
+ module.decoder.layers.0.self_attention.linear_qkv.weight
765
+ module.decoder.layers.0.pre_mlp_layernorm.weight
766
+ Params for bucket 18 (206045184 elements, 206045184 padded size):
767
+ module.embedding.word_embeddings.weight
768
+ INFO:megatron.core.distributed.param_and_grad_buffer:Number of buckets for gradient all-reduce / reduce-scatter: 7
769
+ Params for bucket 1 (44040192 elements, 44040192 padded size):
770
+ module.decoder.layers.15.mlp.experts.linear_fc2.weight0
771
+ module.decoder.layers.13.mlp.experts.linear_fc2.weight0
772
+ module.decoder.layers.14.mlp.experts.linear_fc2.weight0
773
+ module.decoder.layers.15.mlp.experts.linear_fc1.weight0
774
+ module.decoder.layers.14.mlp.experts.linear_fc1.weight0
775
+ Params for bucket 2 (50331648 elements, 50331648 padded size):
776
+ module.decoder.layers.12.mlp.experts.linear_fc2.weight0
777
+ module.decoder.layers.11.mlp.experts.linear_fc2.weight0
778
+ module.decoder.layers.13.mlp.experts.linear_fc1.weight0
779
+ module.decoder.layers.11.mlp.experts.linear_fc1.weight0
780
+ module.decoder.layers.12.mlp.experts.linear_fc1.weight0
781
+ Params for bucket 3 (44040192 elements, 44040192 padded size):
782
+ module.decoder.layers.10.mlp.experts.linear_fc2.weight0
783
+ module.decoder.layers.8.mlp.experts.linear_fc2.weight0
784
+ module.decoder.layers.9.mlp.experts.linear_fc2.weight0
785
+ module.decoder.layers.9.mlp.experts.linear_fc1.weight0
786
+ module.decoder.layers.10.mlp.experts.linear_fc1.weight0
787
+ Params for bucket 4 (50331648 elements, 50331648 padded size):
788
+ module.decoder.layers.7.mlp.experts.linear_fc2.weight0
789
+ module.decoder.layers.6.mlp.experts.linear_fc1.weight0
790
+ module.decoder.layers.8.mlp.experts.linear_fc1.weight0
791
+ module.decoder.layers.7.mlp.experts.linear_fc1.weight0
792
+ module.decoder.layers.6.mlp.experts.linear_fc2.weight0
793
+ Params for bucket 5 (44040192 elements, 44040192 padded size):
794
+ module.decoder.layers.5.mlp.experts.linear_fc2.weight0
795
+ module.decoder.layers.4.mlp.experts.linear_fc2.weight0
796
+ module.decoder.layers.5.mlp.experts.linear_fc1.weight0
797
+ module.decoder.layers.3.mlp.experts.linear_fc2.weight0
798
+ module.decoder.layers.4.mlp.experts.linear_fc1.weight0
799
+ Params for bucket 6 (50331648 elements, 50331648 padded size):
800
+ module.decoder.layers.3.mlp.experts.linear_fc1.weight0
801
+ module.decoder.layers.1.mlp.experts.linear_fc2.weight0
802
+ module.decoder.layers.2.mlp.experts.linear_fc1.weight0
803
+ module.decoder.layers.1.mlp.experts.linear_fc1.weight0
804
+ module.decoder.layers.2.mlp.experts.linear_fc2.weight0
805
+ Params for bucket 7 (18874368 elements, 18874368 padded size):
806
+ module.decoder.layers.0.mlp.experts.linear_fc2.weight0
807
+ module.decoder.layers.0.mlp.experts.linear_fc1.weight0
808
+ INFO:megatron.core.optimizer:Setting up optimizer with config OptimizerConfig(optimizer='adam', lr=0.0001, min_lr=1e-05, decoupled_lr=None, decoupled_min_lr=None, weight_decay=0.1, fp8_recipe='mxfp8', fp16=False, bf16=True, reuse_grad_buf_for_mxfp8_param_ag=False, params_dtype=torch.bfloat16, use_precision_aware_optimizer=False, store_param_remainders=True, main_grads_dtype=torch.float32, main_params_dtype=torch.float32, exp_avg_dtype=torch.float32, exp_avg_sq_dtype=torch.float32, loss_scale=None, initial_loss_scale=4294967296, min_loss_scale=1.0, loss_scale_window=1000, hysteresis=2, adam_beta1=0.9, adam_beta2=0.999, adam_eps=1e-08, sgd_momentum=0.9, use_distributed_optimizer=True, overlap_param_gather_with_optimizer_step=False, optimizer_cpu_offload=False, optimizer_offload_fraction=1.0, use_torch_optimizer_for_cpu_offload=False, overlap_cpu_optimizer_d2h_h2d=False, pin_cpu_grads=True, pin_cpu_params=True, clip_grad=1.0, log_num_zeros_in_grad=False, barrier_with_L1_time=True, timers=<megatron.core.timers.Timers object at 0x765b13af8da0>, config_logger_dir='')
809
+ INFO:megatron.core.optimizer_param_scheduler:> learning rate decay style: cosine
810
+ [after model, optimizer, and learning rate scheduler are built] datetime: 2026-08-30 07:16:03
811
+ > building train, validation, and test datasets ...
812
+ > datasets target sizes (minimum size):
813
+ train: 192
814
+ validation: 0
815
+ test: 0
816
+ INFO:megatron.core.datasets.blended_megatron_dataset_config:Let split_matrix = [(0, 1.0), None, None]
817
+ > building train, validation, and test datasets for GPT ...
818
+ INFO:megatron.core.datasets.blended_megatron_dataset_builder:Building GPTDataset splits with sizes=(192, 0, 0) and config=GPTDatasetConfig(random_seed=1234, sequence_length=4096, blend=(['/workspace/data/wikitext/wikitext103_gpt2_text_document'], None), blend_per_split=None, split='100,0,0', split_matrix=[(0, 1.0), None, None], num_dataset_builder_threads=1, path_to_cache=None, mmap_bin_files=True, mock=False, tokenizer=<megatron.training.tokenizer.tokenizer._GPT2BPETokenizer object at 0x765b13c2be60>, mid_level_dataset_surplus=0.005, reset_position_ids=False, reset_attention_mask=False, eod_mask_loss=False, create_attention_mask=True, drop_last_partial_validation_sequence=True, add_extra_token_to_sequence=True, object_storage_cache_path=None)
819
+ INFO:megatron.core.datasets.indexed_dataset:Load the _IndexReader from /workspace/data/wikitext/wikitext103_gpt2_text_document.idx
820
+ INFO:megatron.core.datasets.indexed_dataset: Extract the sequence lengths
821
+ INFO:megatron.core.datasets.indexed_dataset: Extract the sequence pointers
822
+ INFO:megatron.core.datasets.indexed_dataset: Extract the document indices
823
+ INFO:megatron.core.datasets.indexed_dataset:> total number of sequences: 1165029
824
+ INFO:megatron.core.datasets.indexed_dataset:> total number of documents: 1801350
825
+ INFO:megatron.core.datasets.gpt_dataset:Load the GPTDataset train indices
826
+ INFO:megatron.core.datasets.gpt_dataset: Load the document index from 44ed1ee5bcb9e5b030387b8d27a6c406-GPTDataset-train-document_index.npy
827
+ INFO:megatron.core.datasets.gpt_dataset: Load the sample index from 44ed1ee5bcb9e5b030387b8d27a6c406-GPTDataset-train-sample_index.npy
828
+ INFO:megatron.core.datasets.gpt_dataset: Load the shuffle index from 44ed1ee5bcb9e5b030387b8d27a6c406-GPTDataset-train-shuffle_index.npy
829
+ INFO:megatron.core.datasets.gpt_dataset:> total number of samples: 29073
830
+ > finished creating GPT datasets ...
831
+ [after dataloaders are built] datetime: 2026-08-30 07:16:03
832
+ done with setup ...
833
+ (min, max) time across ranks (ms):
834
+ model-and-optimizer-setup ......................: (3267.82, 3267.98)
835
+ train/valid/test-data-iterators-setup ..........: (73.21, 90.51)
836
+ training ...
837
+ Overwriting rerun_state_machine.current_iteration from -1 to 0...
838
+ [before the start of training step] datetime: 2026-08-30 07:16:03
839
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
840
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
841
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
842
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
843
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
844
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
845
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
846
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
847
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
848
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
849
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
850
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
851
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
852
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
853
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
854
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
855
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
856
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
857
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
858
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
859
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
860
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
861
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
862
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
863
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
864
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
865
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
866
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
867
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
868
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
869
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
870
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
871
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
872
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
873
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
874
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
875
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
876
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
877
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
878
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
879
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
880
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
881
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
882
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
883
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
884
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
885
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
886
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
887
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
888
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
889
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
890
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
891
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
892
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
893
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
894
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
895
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
896
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
897
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
898
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
899
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
900
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
901
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
902
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
903
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
904
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
905
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
906
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
907
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
908
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
909
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
910
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
911
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
912
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
913
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
914
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
915
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
916
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
917
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
918
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
919
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
920
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
921
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
922
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
923
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
924
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
925
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
926
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
927
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
928
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
929
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
930
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
931
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
932
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
933
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
934
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
935
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
936
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
937
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
938
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
939
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
940
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
941
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
942
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
943
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
944
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
945
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
946
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
947
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
948
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
949
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
950
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
951
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
952
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
953
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
954
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
955
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
956
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
957
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
958
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
959
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
960
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
961
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
962
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
963
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
964
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
965
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
966
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
967
+ Number of parameters in transformer block in billions: 3.09 [2026-08-30 07:16:29] iteration 1/ 12 | consumed samples: 16 | elapsed time per iteration (ms): 25671.4 | throughput per GPU (TFLOP/s/GPU): 3.3 | learning rate: 2.000000E-05 | global batch size: 16 | lm loss: 1.103286E+01 | loss scale: 1.0 | grad norm: 44.354 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14592 | min safe capacity: 1.7812 |
968
+
969
+ Number of parameters in embedding layers in billions: 0.41
970
+ Total number of parameters in billions: 3.50
971
+ Number of parameters in most loaded shard in billions: 3.4994
972
+ Theoretical memory footprints: weight and optimizer=25029.43 MB
973
+ [Rank 0] (after 1 iterations) memory (MB) | allocated: 14864.61376953125 | max allocated: 22156.96533203125 | reserved: 22904.0 | max reserved: 22904.0
974
+ [2026-08-30 07:16:29] iteration 2/ 12 | consumed samples: 32 | elapsed time per iteration (ms): 271.9 | throughput per GPU (TFLOP/s/GPU): 316.3 | learning rate: 4.000000E-05 | global batch size: 16 | lm loss: 1.102278E+01 | loss scale: 1.0 | grad norm: 46.283 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14848 | min safe capacity: 1.8125 |
975
+ [2026-08-30 07:16:29] iteration 3/ 12 | consumed samples: 48 | elapsed time per iteration (ms): 256.9 | throughput per GPU (TFLOP/s/GPU): 334.7 | learning rate: 6.000000E-05 | global batch size: 16 | lm loss: 1.037333E+01 | loss scale: 1.0 | grad norm: 215.165 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14848 | min safe capacity: 1.8125 |
976
+ Capture CUDA graph for training!!!
977
+ CUDA graph capture done!!!
978
+ [2026-08-30 07:16:30] iteration 4/ 12 | consumed samples: 64 | elapsed time per iteration (ms): 992.4 | throughput per GPU (TFLOP/s/GPU): 86.6 | learning rate: 8.000000E-05 | global batch size: 16 | lm loss: 1.313609E+01 | loss scale: 1.0 | grad norm: 41.178 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14848 | min safe capacity: 1.8125 |
979
+ [2026-08-30 07:16:31] iteration 5/ 12 | consumed samples: 80 | elapsed time per iteration (ms): 177.2 | throughput per GPU (TFLOP/s/GPU): 485.4 | learning rate: 1.000000E-04 | global batch size: 16 | lm loss: 1.037827E+01 | loss scale: 1.0 | grad norm: 15.978 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14848 | min safe capacity: 1.8125 |
980
+ [2026-08-30 07:16:31] iteration 6/ 12 | consumed samples: 96 | elapsed time per iteration (ms): 176.7 | throughput per GPU (TFLOP/s/GPU): 486.7 | learning rate: 7.750000E-05 | global batch size: 16 | lm loss: 1.003041E+01 | loss scale: 1.0 | grad norm: 9.285 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14848 | min safe capacity: 1.8125 |
981
+ [2026-08-30 07:16:31] iteration 7/ 12 | consumed samples: 112 | elapsed time per iteration (ms): 176.7 | throughput per GPU (TFLOP/s/GPU): 486.8 | learning rate: 3.250000E-05 | global batch size: 16 | lm loss: 9.370650E+00 | loss scale: 1.0 | grad norm: 14.716 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14848 | min safe capacity: 1.8125 |
982
+ [2026-08-30 07:16:31] iteration 8/ 12 | consumed samples: 128 | elapsed time per iteration (ms): 176.7 | throughput per GPU (TFLOP/s/GPU): 486.8 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.942919E+00 | loss scale: 1.0 | grad norm: 6.170 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14848 | min safe capacity: 1.8125 |
983
+ [2026-08-30 07:16:31] iteration 9/ 12 | consumed samples: 144 | elapsed time per iteration (ms): 177.2 | throughput per GPU (TFLOP/s/GPU): 485.4 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.783094E+00 | loss scale: 1.0 | grad norm: 5.180 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14848 | min safe capacity: 1.8125 |
984
+ [2026-08-30 07:16:31] iteration 10/ 12 | consumed samples: 160 | elapsed time per iteration (ms): 175.8 | throughput per GPU (TFLOP/s/GPU): 489.2 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.694411E+00 | loss scale: 1.0 | grad norm: 5.615 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14848 | min safe capacity: 1.8125 |
985
+ [2026-08-30 07:16:32] iteration 11/ 12 | consumed samples: 176 | elapsed time per iteration (ms): 177.4 | throughput per GPU (TFLOP/s/GPU): 484.9 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.617825E+00 | loss scale: 1.0 | grad norm: 4.956 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14848 | min safe capacity: 1.8125 |
986
+ [2026-08-30 07:16:32] iteration 12/ 12 | consumed samples: 192 | elapsed time per iteration (ms): 177.1 | throughput per GPU (TFLOP/s/GPU): 485.6 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.517728E+00 | loss scale: 1.0 | grad norm: 4.700 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14848 | min safe capacity: 1.8125 |
987
+ [after training is done] datetime: 2026-08-30 07:16:32
988
+ [rank0]:[W830 07:16:33.716429081 ProcessGroupNCCL.cpp:1505] Warning: WARNING: destroy_process_group() was not called before program exit, which can leak resources. For more info, please see https://pytorch.org/docs/stable/distributed.html#shutdown (function operator())
989
+ [rank6]:[W830 07:16:34.917855225 ProcessGroupNCCL.cpp:1505] Warning: WARNING: destroy_process_group() was not called before program exit, which can leak resources. For more info, please see https://pytorch.org/docs/stable/distributed.html#shutdown (function operator())
990
+ [rank5]:[W830 07:16:34.921394206 ProcessGroupNCCL.cpp:1505] Warning: WARNING: destroy_process_group() was not called before program exit, which can leak resources. For more info, please see https://pytorch.org/docs/stable/distributed.html#shutdown (function operator())
991
+ [rank3]:[W830 07:16:34.923630160 ProcessGroupNCCL.cpp:1505] Warning: WARNING: destroy_process_group() was not called before program exit, which can leak resources. For more info, please see https://pytorch.org/docs/stable/distributed.html#shutdown (function operator())
992
+ [rank4]:[W830 07:16:34.924153413 ProcessGroupNCCL.cpp:1505] Warning: WARNING: destroy_process_group() was not called before program exit, which can leak resources. For more info, please see https://pytorch.org/docs/stable/distributed.html#shutdown (function operator())
993
+ [rank1]:[W830 07:16:34.924377536 ProcessGroupNCCL.cpp:1505] Warning: WARNING: destroy_process_group() was not called before program exit, which can leak resources. For more info, please see https://pytorch.org/docs/stable/distributed.html#shutdown (function operator())
994
+ [rank2]:[W830 07:16:34.924666281 ProcessGroupNCCL.cpp:1505] Warning: WARNING: destroy_process_group() was not called before program exit, which can leak resources. For more info, please see https://pytorch.org/docs/stable/distributed.html#shutdown (function operator())
995
+ W0830 07:16:34.726000 730290 torch/distributed/elastic/agent/server/api.py:719] Received 15 death signal, shutting down workers
996
+ W0830 07:16:34.727000 730290 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 730378 closing signal SIGTERM
997
+ W0830 07:16:34.728000 730290 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 730379 closing signal SIGTERM
998
+ W0830 07:16:34.729000 730290 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 730380 closing signal SIGTERM
999
+ W0830 07:16:34.730000 730290 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 730381 closing signal SIGTERM
1000
+ W0830 07:16:34.732000 730290 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 730382 closing signal SIGTERM
1001
+ W0830 07:16:34.733000 730290 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 730383 closing signal SIGTERM
1002
+ W0830 07:16:34.734000 730290 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 730384 closing signal SIGTERM
1003
+ W0830 07:16:34.735000 730290 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 730385 closing signal SIGTERM
1004
+ Traceback (most recent call last):
1005
+ File "/usr/local/bin/torchrun", line 8, in <module>
1006
+ sys.exit(main())
1007
+ ^^^^^^
1008
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/multiprocessing/errors/__init__.py", line 355, in wrapper
1009
+ return f(*args, **kwargs)
1010
+ ^^^^^^^^^^^^^^^^^^
1011
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/run.py", line 892, in main
1012
+ run(args)
1013
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/run.py", line 883, in run
1014
+ elastic_launch(
1015
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/launcher/api.py", line 139, in __call__
1016
+ return launch_agent(self._config, self._entrypoint, list(args))
1017
+ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1018
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/launcher/api.py", line 261, in launch_agent
1019
+ result = agent.run()
1020
+ ^^^^^^^^^^^
1021
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/metrics/api.py", line 138, in wrapper
1022
+ result = f(*args, **kwargs)
1023
+ ^^^^^^^^^^^^^^^^^^
1024
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/agent/server/api.py", line 711, in run
1025
+ result = self._invoke_run(role)
1026
+ ^^^^^^^^^^^^^^^^^^^^^^
1027
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/agent/server/api.py", line 870, in _invoke_run
1028
+ time.sleep(monitor_interval)
1029
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/multiprocessing/api.py", line 84, in _terminate_process_handler
1030
+ raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval)
1031
+ torch.distributed.elastic.multiprocessing.api.SignalException: Process 730290 got signal: 15
tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq4096__gbs16__cap2p40625/rep2/mem_samples.csv ADDED
@@ -0,0 +1,200 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 0, 0
2
+ 1, 0
3
+ 2, 0
4
+ 3, 0
5
+ 4, 0
6
+ 5, 0
7
+ 6, 0
8
+ 7, 0
9
+ 0, 0
10
+ 1, 0
11
+ 2, 0
12
+ 3, 0
13
+ 4, 0
14
+ 5, 0
15
+ 6, 0
16
+ 7, 0
17
+ 0, 1
18
+ 1, 1
19
+ 2, 1
20
+ 3, 1
21
+ 4, 1
22
+ 5, 1
23
+ 6, 1
24
+ 7, 1
25
+ 0, 4
26
+ 1, 4
27
+ 2, 4
28
+ 3, 4
29
+ 4, 1
30
+ 5, 1
31
+ 6, 1
32
+ 7, 1
33
+ 0, 4
34
+ 1, 4
35
+ 2, 4
36
+ 3, 4
37
+ 4, 4
38
+ 5, 4
39
+ 6, 4
40
+ 7, 1
41
+ 0, 4
42
+ 1, 4
43
+ 2, 4
44
+ 3, 4
45
+ 4, 4
46
+ 5, 4
47
+ 6, 4
48
+ 7, 4
49
+ 0, 4
50
+ 1, 629
51
+ 2, 629
52
+ 3, 629
53
+ 4, 629
54
+ 5, 629
55
+ 6, 629
56
+ 7, 629
57
+ 0, 1025
58
+ 1, 1013
59
+ 2, 1053
60
+ 3, 1033
61
+ 4, 1037
62
+ 5, 1019
63
+ 6, 1029
64
+ 7, 1053
65
+ 0, 1895
66
+ 1, 1895
67
+ 2, 1895
68
+ 3, 1895
69
+ 4, 1895
70
+ 5, 1895
71
+ 6, 1895
72
+ 7, 1895
73
+ 0, 2387
74
+ 1, 2361
75
+ 2, 2397
76
+ 3, 2391
77
+ 4, 2397
78
+ 5, 2391
79
+ 6, 2407
80
+ 7, 2371
81
+ 0, 14623
82
+ 1, 14623
83
+ 2, 14623
84
+ 3, 14605
85
+ 4, 14623
86
+ 5, 14621
87
+ 6, 14623
88
+ 7, 14623
89
+ 0, 15629
90
+ 1, 15649
91
+ 2, 15611
92
+ 3, 15633
93
+ 4, 15615
94
+ 5, 15645
95
+ 6, 15647
96
+ 7, 15625
97
+ 0, 16323
98
+ 1, 16321
99
+ 2, 16321
100
+ 3, 16321
101
+ 4, 16321
102
+ 5, 16321
103
+ 6, 16321
104
+ 7, 16321
105
+ 0, 16323
106
+ 1, 16321
107
+ 2, 16321
108
+ 3, 16321
109
+ 4, 16321
110
+ 5, 16321
111
+ 6, 16321
112
+ 7, 16321
113
+ 0, 16677
114
+ 1, 16675
115
+ 2, 16675
116
+ 3, 16675
117
+ 4, 16675
118
+ 5, 16675
119
+ 6, 16675
120
+ 7, 16675
121
+ 0, 25943
122
+ 1, 25941
123
+ 2, 25941
124
+ 3, 25941
125
+ 4, 25911
126
+ 5, 25941
127
+ 6, 25941
128
+ 7, 25941
129
+ 0, 26793
130
+ 1, 26791
131
+ 2, 26791
132
+ 3, 26791
133
+ 4, 26761
134
+ 5, 26791
135
+ 6, 26791
136
+ 7, 26791
137
+ 0, 26793
138
+ 1, 26791
139
+ 2, 26791
140
+ 3, 26791
141
+ 4, 26761
142
+ 5, 26791
143
+ 6, 26791
144
+ 7, 26791
145
+ 0, 28905
146
+ 1, 28923
147
+ 2, 28931
148
+ 3, 28927
149
+ 4, 28841
150
+ 5, 28915
151
+ 6, 28923
152
+ 7, 28949
153
+ 0, 29891
154
+ 1, 29881
155
+ 2, 29897
156
+ 3, 29919
157
+ 4, 29831
158
+ 5, 29883
159
+ 6, 29893
160
+ 7, 29907
161
+ 0, 31041
162
+ 1, 31037
163
+ 2, 31025
164
+ 3, 31061
165
+ 4, 30987
166
+ 5, 31031
167
+ 6, 31049
168
+ 7, 31041
169
+ 0, 38329
170
+ 1, 38319
171
+ 2, 38315
172
+ 3, 38315
173
+ 4, 38311
174
+ 5, 38317
175
+ 6, 38319
176
+ 7, 38319
177
+ 0, 38067
178
+ 1, 38057
179
+ 2, 38053
180
+ 3, 38053
181
+ 4, 38049
182
+ 5, 38055
183
+ 6, 38057
184
+ 7, 38057
185
+ 0, 34851
186
+ 1, 34777
187
+ 2, 34899
188
+ 3, 34643
189
+ 4, 34833
190
+ 5, 34965
191
+ 6, 34877
192
+ 7, 34967
193
+ 0, 1
194
+ 1, 1
195
+ 2, 1
196
+ 3, 193
197
+ 4, 34830
198
+ 5, 961
199
+ 6, 4
200
+ 7, 4
tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq4096__gbs16__cap2p40625/rep2/run.env ADDED
@@ -0,0 +1,70 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ backend=ultraep
2
+ config=PONLY
3
+ family=tierA
4
+ repo=/workspace/Megatron-UltraEP
5
+ branch=ultraep-bracing-regions-through-kernels
6
+ commit=03d57ef7
7
+ num_layers=16
8
+ hidden=4096
9
+ ffn_hidden=14336
10
+ moe_ffn_hidden=1536
11
+ num_experts=8
12
+ topk=2
13
+ experts_per_rank=1
14
+ expert_params=18874368
15
+ heads=32
16
+ kv_groups=8
17
+ kv_channels=
18
+ qk_layernorm=0
19
+ shared_ffn=0
20
+ shared_expert_overlap=0
21
+ moe_token_dispatcher=flex
22
+ mla=0
23
+ dense_first_layers=0
24
+ moe_layers=16
25
+ rotary_base=1000000
26
+ rotary_scaling=
27
+ qkv_bias=0
28
+ router_balancing=aux_loss
29
+ aux_loss_coeff=1e-2
30
+ router_pre_softmax=0
31
+ router_fusion=0
32
+ router_dtype=fp32
33
+ device_initiated_grouped_gemm=1
34
+ norm_epsilon=1e-5
35
+ seq_length=4096
36
+ micro_batch_size=1
37
+ global_batch_size=16
38
+ num_microbatches=2
39
+ assignments_per_rank=8192
40
+ mean_tokens_per_expert=8192
41
+ slot_policy=one
42
+ slots_per_rank=1
43
+ capacity_mode=static
44
+ capacity=2.40625
45
+ cuda_graph=1
46
+ probe_mode=0
47
+ async_grad_reduce=0
48
+ branch_role=timing
49
+ memory_dump_mode=none
50
+ memory_snapshot_path=
51
+ cuda_graph_warmup_steps=3
52
+ recompute=none
53
+ force_balance=0
54
+ train_iters=12
55
+ ep_size=8
56
+ echo_dispatch_overlap=1
57
+ echo_dispatch_recompute=0
58
+ ultraep_quota_min_tokens=1024
59
+ ultraep_weight_sync_join=
60
+ profile_enable=0
61
+ kernel_trace=0
62
+ fp8_mode=mxfp8
63
+ fp8_recipe=mxfp8
64
+ training_seed=1234
65
+ learning_rate=1e-4
66
+ min_learning_rate=1.0e-5
67
+ lr_warmup_iters=5
68
+ state_dump_enable=0
69
+ state_dump_dir=
70
+ moonep_assert_no_pending_reduce=0
tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq4096__gbs16__cap2p40625/rep2/train.log ADDED
@@ -0,0 +1,1032 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ==============================================================
2
+ PONLY / ultraep (slots/rank=1 policy=one)
3
+ E=8 K=2 H'=1536 m=1
4
+ L=16 seq=4096 gbs=16 (2 microbatches)
5
+ cap_mode=static cap=2.40625 cg=1 recompute=none
6
+ memory_dump=none snapshots=/workspace/tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq4096__gbs16__cap2p40625/rep2/memory_snapshots/snapshot.pickle
7
+ repo=/workspace/Megatron-UltraEP @ 03d57ef7
8
+ ==============================================================
9
+ W0830 07:16:40.222000 733201 torch/distributed/run.py:766]
10
+ W0830 07:16:40.222000 733201 torch/distributed/run.py:766] *****************************************
11
+ W0830 07:16:40.222000 733201 torch/distributed/run.py:766] Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed.
12
+ W0830 07:16:40.222000 733201 torch/distributed/run.py:766] *****************************************
13
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
14
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
15
+ WARNING: one_logger package is required to enable e2e metrics tracking. please go to https://confluence.nvidia.com/display/MLWFO/Package+Repositories for details to install it
16
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
17
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
18
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
19
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
20
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
21
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
22
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
23
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
24
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
25
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
26
+ using world size: 8, data-parallel size: 8, context-parallel size: 1, hierarchical context-parallel sizes: None, tensor-model-parallel size: 1, pipeline-model-parallel size: 1
27
+ WARNING: overriding default arguments for tokenizer_type:GPT2BPETokenizer with tokenizer_type:GPT2BPETokenizer
28
+ Number of virtual stages per pipeline stage: None
29
+ accumulate and all-reduce gradients in fp32 for bfloat16 data type.
30
+ using torch.bfloat16 for parameters ...
31
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
32
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
33
+ ------------------------ arguments ------------------------
34
+ account_for_embedding_in_pipeline_split ......... False
35
+ account_for_loss_in_pipeline_split .............. False
36
+ accumulate_allreduce_grads_in_fp32 .............. True
37
+ adam_beta1 ...................................... 0.9
38
+ adam_beta2 ...................................... 0.999
39
+ adam_eps ........................................ 1e-08
40
+ add_bias_linear ................................. False
41
+ add_position_embedding .......................... False
42
+ add_qkv_bias .................................... False
43
+ adlr_autoresume ................................. False
44
+ adlr_autoresume_interval ........................ 1000
45
+ align_grad_reduce ............................... True
46
+ align_param_gather .............................. False
47
+ app_tag_run_name ................................ None
48
+ app_tag_run_version ............................. 0.0.0
49
+ apply_layernorm_1p .............................. False
50
+ apply_query_key_layer_scaling ................... False
51
+ apply_residual_connection_post_layernorm ........ False
52
+ apply_rope_fusion ............................... True
53
+ async_save ...................................... None
54
+ async_tensor_model_parallel_allreduce ........... True
55
+ attention_backend ............................... AttnBackend.auto
56
+ attention_dropout ............................... 0.0
57
+ attention_softmax_in_fp32 ....................... False
58
+ auto_detect_ckpt_format ......................... False
59
+ barrier_with_L1_time ............................ True
60
+ bert_binary_head ................................ True
61
+ bert_embedder_type .............................. megatron
62
+ bert_load ....................................... None
63
+ bf16 ............................................ True
64
+ bias_dropout_fusion ............................. True
65
+ bias_gelu_fusion ................................ False
66
+ bias_swiglu_fusion .............................. True
67
+ biencoder_projection_dim ........................ 0
68
+ biencoder_shared_query_context_model ............ False
69
+ block_data_path ................................. None
70
+ calc_ft_timeouts ................................ False
71
+ calculate_per_token_loss ........................ False
72
+ check_for_large_grads ........................... False
73
+ check_for_nan_in_loss_and_grad .................. False
74
+ check_for_spiky_loss ............................ False
75
+ check_weight_hash_across_dp_replicas_interval ... None
76
+ ckpt_assume_constant_structure .................. False
77
+ ckpt_convert_format ............................. None
78
+ ckpt_convert_save ............................... None
79
+ ckpt_convert_update_legacy_dist_opt_format ...... False
80
+ ckpt_format ..................................... torch_dist
81
+ ckpt_fully_parallel_load ........................ False
82
+ ckpt_fully_parallel_save ........................ True
83
+ ckpt_fully_parallel_save_deprecated ............. False
84
+ ckpt_step ....................................... None
85
+ classes_fraction ................................ 1.0
86
+ clip_grad ....................................... 1.0
87
+ clone_scatter_output_in_embedding ............... True
88
+ config_logger_dir ...............................
89
+ consumed_train_samples .......................... 0
90
+ consumed_valid_samples .......................... 0
91
+ context_parallel_size ........................... 1
92
+ cp_comm_type .................................... ['p2p']
93
+ create_attention_mask_in_dataloader ............. True
94
+ cross_entropy_fusion_impl ....................... native
95
+ cross_entropy_loss_fusion ....................... False
96
+ cuda_graph_scope ................................ full_iteration
97
+ cuda_graph_warmup_steps ......................... 3
98
+ data_args_path .................................. None
99
+ data_cache_path ................................. None
100
+ data_parallel_random_init ....................... False
101
+ data_parallel_sharding_strategy ................. no_shard
102
+ data_parallel_size .............................. 8
103
+ data_path ....................................... ['/workspace/data/wikitext/wikitext103_gpt2_text_document']
104
+ data_per_class_fraction ......................... 1.0
105
+ data_sharding ................................... True
106
+ dataloader_type ................................. single
107
+ ddp_average_in_collective ....................... False
108
+ ddp_bucket_size ................................. None
109
+ ddp_num_buckets ................................. None
110
+ ddp_pad_buckets_for_high_nccl_busbw ............. False
111
+ decoder_first_pipeline_num_layers ............... None
112
+ decoder_last_pipeline_num_layers ................ None
113
+ decoder_num_layers .............................. None
114
+ decoder_seq_length .............................. None
115
+ decoupled_lr .................................... None
116
+ decoupled_min_lr ................................ None
117
+ decrease_batch_size_if_needed ................... False
118
+ defer_embedding_wgrad_compute ................... False
119
+ delay_wgrad_compute ............................. False
120
+ deprecated_use_mcore_models ..................... True
121
+ deterministic_mode .............................. False
122
+ dino_bottleneck_size ............................ 256
123
+ dino_freeze_last_layer .......................... 1
124
+ dino_head_hidden_size ........................... 2048
125
+ dino_local_crops_number ......................... 10
126
+ dino_local_img_size ............................. 96
127
+ dino_norm_last_layer ............................ False
128
+ dino_teacher_temp ............................... 0.07
129
+ dino_warmup_teacher_temp ........................ 0.04
130
+ dino_warmup_teacher_temp_epochs ................. 30
131
+ disable_bf16_reduced_precision_matmul ........... False
132
+ disable_mamba_mem_eff_path ...................... False
133
+ disable_straggler_on_startup .................... False
134
+ dist_ckpt_format_deprecated ..................... None
135
+ dist_ckpt_strictness ............................ assume_ok_unexpected
136
+ distribute_saved_activations .................... False
137
+ distributed_backend ............................. nccl
138
+ distributed_timeout_minutes ..................... 10
139
+ embedding_init_method_std ....................... None
140
+ embedding_path .................................. None
141
+ empty_unused_memory_level ....................... 0
142
+ enable_cuda_graph ............................... True
143
+ enable_experimental ............................. False
144
+ enable_ft_package ............................... False
145
+ enable_gloo_process_groups ...................... True
146
+ enable_msc ...................................... True
147
+ enable_one_logger ............................... True
148
+ encoder_num_layers .............................. 16
149
+ encoder_seq_length .............................. 4096
150
+ end_weight_decay ................................ 0.1
151
+ eod_mask_loss ................................... False
152
+ error_injection_rate ............................ 0
153
+ error_injection_type ............................ transient_error
154
+ eval_interval ................................... 1000000
155
+ eval_iters ...................................... 0
156
+ evidence_data_path .............................. None
157
+ exit_duration_in_mins ........................... None
158
+ exit_interval ................................... None
159
+ exit_on_missing_checkpoint ...................... False
160
+ exit_signal_handler ............................. False
161
+ exp_avg_dtype ................................... torch.float32
162
+ exp_avg_sq_dtype ................................ torch.float32
163
+ expert_model_parallel_size ...................... 8
164
+ expert_tensor_parallel_size ..................... 1
165
+ export_force_local_attention .................... False
166
+ export_kd_cfg ................................... None
167
+ export_kd_teacher_ckpt_format ................... None
168
+ export_kd_teacher_load .......................... None
169
+ export_kv_cache_quant ........................... False
170
+ export_legacy_megatron .......................... False
171
+ export_model_type ............................... GPTModel
172
+ export_moe_apply_probs_on_input ................. False
173
+ export_qk_l2_norm ............................... False
174
+ export_quant_cfg ................................ None
175
+ export_real_quant_cfg ........................... None
176
+ export_te_mcore_model ........................... False
177
+ external_cuda_graph ............................. False
178
+ ffn_hidden_size ................................. 14336
179
+ finetune ........................................ False
180
+ finetune_data_split ............................. train
181
+ finetune_hf_dataset ............................. None
182
+ first_last_layers_bf16 .......................... False
183
+ flash_decode .................................... False
184
+ fp16 ............................................ False
185
+ fp16_lm_cross_entropy ........................... False
186
+ fp32_residual_connection ........................ False
187
+ fp8 ............................................. e4m3
188
+ fp8_amax_compute_algo ........................... most_recent
189
+ fp8_amax_history_len ............................ 1
190
+ fp8_interval .................................... 1
191
+ fp8_margin ...................................... 0
192
+ fp8_param_gather ................................ False
193
+ fp8_recipe ...................................... mxfp8
194
+ fp8_wgrad ....................................... True
195
+ fsdp_double_buffer .............................. False
196
+ global_batch_size ............................... 16
197
+ grad_reduce_in_bf16 ............................. False
198
+ gradient_accumulation_fusion .................... True
199
+ gradient_reduce_div_fusion ...................... True
200
+ group_query_attention ........................... True
201
+ head_lr_mult .................................... 1.0
202
+ heterogeneous_layers_config_encoded_json ........ None
203
+ heterogeneous_layers_config_path ................ None
204
+ hidden_dropout .................................. 0.0
205
+ hidden_size ..................................... 4096
206
+ hierarchical_context_parallel_sizes ............. None
207
+ high_priority_stream_groups ..................... []
208
+ hybrid_attention_ratio .......................... 0.0
209
+ hybrid_mlp_ratio ................................ 0.0
210
+ hybrid_override_pattern ......................... None
211
+ hysteresis ...................................... 2
212
+ ict_head_size ................................... None
213
+ ict_load ........................................ None
214
+ img_h ........................................... 224
215
+ img_w ........................................... 224
216
+ indexer_batch_size .............................. 128
217
+ indexer_log_interval ............................ 1000
218
+ inference_batch_times_seqlen_threshold .......... -1
219
+ inference_dynamic_batching ...................... False
220
+ inference_dynamic_batching_buffer_guaranteed_fraction 0.2
221
+ inference_dynamic_batching_buffer_overflow_factor None
222
+ inference_dynamic_batching_buffer_size_gb ....... 40.0
223
+ inference_dynamic_batching_chunk_size ........... 256
224
+ inference_dynamic_batching_max_requests_override None
225
+ inference_dynamic_batching_max_tokens_override .. None
226
+ inference_dynamic_batching_num_cuda_graphs ...... 16
227
+ inference_max_batch_size ........................ 8
228
+ inference_max_seq_length ........................ 2560
229
+ inference_rng_tracker ........................... False
230
+ init_method_std ................................. 0.01
231
+ init_method_xavier_uniform ...................... False
232
+ init_model_with_meta_device ..................... False
233
+ initial_loss_scale .............................. 4294967296
234
+ inprocess_active_world_size ..................... 8
235
+ inprocess_barrier_timeout ....................... 120
236
+ inprocess_completion_timeout .................... 120
237
+ inprocess_empty_cuda_cache ...................... False
238
+ inprocess_granularity ........................... node
239
+ inprocess_hard_timeout .......................... 90
240
+ inprocess_heartbeat_interval .................... 30
241
+ inprocess_heartbeat_timeout ..................... 60
242
+ inprocess_last_call_wait ........................ 1
243
+ inprocess_max_iterations ........................ None
244
+ inprocess_monitor_process_interval .............. 1.0
245
+ inprocess_monitor_thread_interval ............... 1.0
246
+ inprocess_progress_watchdog_interval ............ 1.0
247
+ inprocess_restart ............................... False
248
+ inprocess_soft_timeout .......................... 60
249
+ inprocess_termination_grace_time ................ 1
250
+ is_hybrid_model ................................. False
251
+ iter_per_epoch .................................. 1250
252
+ iterations_to_skip .............................. []
253
+ keep_fp8_transpose_cache_when_using_custom_fsdp . False
254
+ kitchen_config_file ............................. None
255
+ kitchen_recipe_number ........................... None
256
+ kv_channels ..................................... 128
257
+ kv_lora_rank .................................... 32
258
+ lazy_mpu_init ................................... None
259
+ load ............................................ None
260
+ load_main_params_from_ckpt ...................... None
261
+ load_model_opt_format ........................... False
262
+ local_rank ...................................... 0
263
+ log_energy ...................................... False
264
+ log_interval .................................... 1
265
+ log_loss_scale_to_tensorboard ................... True
266
+ log_memory_to_tensorboard ....................... False
267
+ log_num_zeros_in_grad ........................... False
268
+ log_params_norm ................................. False
269
+ log_progress .................................... False
270
+ log_straggler ................................... False
271
+ log_throughput .................................. True
272
+ log_timers_to_tensorboard ....................... False
273
+ log_validation_ppl_to_tensorboard ............... False
274
+ log_world_size_to_tensorboard ................... False
275
+ logging_level ................................... None
276
+ loss_scale ...................................... None
277
+ loss_scale_window ............................... 1000
278
+ lr .............................................. 0.0001
279
+ lr_decay_iters .................................. 8
280
+ lr_decay_samples ................................ None
281
+ lr_decay_style .................................. cosine
282
+ lr_warmup_fraction .............................. None
283
+ lr_warmup_init .................................. 0.0
284
+ lr_warmup_iters ................................. 5
285
+ lr_warmup_samples ............................... 0
286
+ lr_wsd_decay_iters .............................. None
287
+ lr_wsd_decay_samples ............................ None
288
+ lr_wsd_decay_style .............................. exponential
289
+ main_grads_dtype ................................ torch.float32
290
+ main_params_dtype ............................... torch.float32
291
+ make_vocab_size_divisible_by .................... 128
292
+ mamba_head_dim .................................. 64
293
+ mamba_num_groups ................................ 8
294
+ mamba_num_heads ................................. None
295
+ mamba_state_dim ................................. 128
296
+ manual_gc ....................................... False
297
+ manual_gc_eval .................................. True
298
+ manual_gc_interval .............................. 0
299
+ mask_factor ..................................... 1.0
300
+ mask_prob ....................................... 0.15
301
+ mask_type ....................................... random
302
+ masked_softmax_fusion ........................... False
303
+ max_position_embeddings ......................... 32768
304
+ max_tokens_to_oom ............................... 12000
305
+ memory_snapshot_path ............................ snapshot.pickle
306
+ merge_file ...................................... /workspace/data/wikitext/gpt2-merges.txt
307
+ micro_batch_size ................................ 1
308
+ microbatch_group_size_per_vp_stage .............. None
309
+ mid_level_dataset_surplus ....................... 0.005
310
+ min_loss_scale .................................. 1.0
311
+ min_lr .......................................... 1e-05
312
+ mlp_chunks_for_prefill .......................... 1
313
+ mmap_bin_files .................................. True
314
+ mock_data ....................................... False
315
+ moe_apply_probs_on_input ........................ False
316
+ moe_aux_loss_coeff .............................. 0.01
317
+ moe_capacity_monitor ............................ True
318
+ moe_deepep_num_sms .............................. 20
319
+ moe_echo_enable_random_offloading ............... False
320
+ moe_echo_expert_dispatch_overlap ................ False
321
+ moe_echo_expert_dispatcher_type ................. hybridep
322
+ moe_echo_recompute_expert_dispatch .............. False
323
+ moe_enable_deepep ............................... False
324
+ moe_enable_echo ................................. False
325
+ moe_enable_hybridep ............................. True
326
+ moe_enable_ultraep .............................. True
327
+ moe_expert_capacity_factor ...................... None
328
+ moe_extended_tp ................................. False
329
+ moe_ffn_hidden_size ............................. 1536
330
+ moe_grouped_gemm ................................ True
331
+ moe_hybridep_num_sms ............................ 16
332
+ moe_input_jitter_eps ............................ None
333
+ moe_layer_freq .................................. 1
334
+ moe_layer_recompute ............................. False
335
+ moe_num_echo_experts ............................ None
336
+ moe_num_redundant_experts_per_rank .............. 1
337
+ moe_pad_expert_input_to_capacity ................ False
338
+ moe_per_layer_logging ........................... False
339
+ moe_permute_fusion .............................. False
340
+ moe_received_token_capacity ..................... 2.40625
341
+ moe_router_bias_update_rate ..................... 0.001
342
+ moe_router_dtype ................................ fp32
343
+ moe_router_enable_expert_bias ................... False
344
+ moe_router_force_load_balancing ................. False
345
+ moe_router_group_topk ........................... None
346
+ moe_router_load_balancing_type .................. aux_loss
347
+ moe_router_num_groups ........................... None
348
+ moe_router_padding_for_fp8 ...................... False
349
+ moe_router_pre_softmax .......................... False
350
+ moe_router_score_function ....................... softmax
351
+ moe_router_topk ................................. 2
352
+ moe_router_topk_scaling_factor .................. None
353
+ moe_shared_expert_intermediate_size ............. None
354
+ moe_shared_expert_overlap ....................... False
355
+ moe_token_dispatcher_type ....................... flex
356
+ moe_token_drop_policy ........................... probs
357
+ moe_ultraep_max_microbatches .................... None
358
+ moe_ultraep_overlap_weight_sync ................. True
359
+ moe_ultraep_weight_sync_join .................... before_dispatch
360
+ moe_upcycling_granularity ....................... 1
361
+ moe_use_device_initiated_grouped_gemm ........... True
362
+ moe_use_legacy_grouped_gemm ..................... False
363
+ moe_use_upcycling ............................... False
364
+ moe_z_loss_coeff ................................ None
365
+ mrope_section ................................... None
366
+ mscale .......................................... 1.0
367
+ mscale_all_dim .................................. 0.0
368
+ mtp_loss_scaling_factor ......................... 0.1
369
+ mtp_num_layers .................................. None
370
+ multi_latent_attention .......................... False
371
+ nccl_all_reduce_for_prefill ..................... False
372
+ nccl_communicator_config_path ................... None
373
+ nccl_ub ......................................... False
374
+ no_load_optim ................................... None
375
+ no_load_rng ..................................... None
376
+ no_persist_layer_norm ........................... False
377
+ no_rope_freq .................................... None
378
+ no_save_optim ................................... None
379
+ no_save_rng ..................................... None
380
+ non_persistent_ckpt_type ........................ None
381
+ non_persistent_global_ckpt_dir .................. None
382
+ non_persistent_local_ckpt_algo .................. fully_parallel
383
+ non_persistent_local_ckpt_dir ................... None
384
+ non_persistent_save_interval .................... None
385
+ norm_epsilon .................................... 1e-05
386
+ normalization ................................... RMSNorm
387
+ num_attention_heads ............................. 32
388
+ num_channels .................................... 3
389
+ num_classes ..................................... 1000
390
+ num_dataset_builder_threads ..................... 1
391
+ num_distributed_optimizer_instances ............. 1
392
+ num_experts ..................................... 8
393
+ num_layers ...................................... 16
394
+ num_layers_at_end_in_bf16 ....................... 1
395
+ num_layers_at_start_in_bf16 ..................... 1
396
+ num_layers_per_virtual_pipeline_stage ........... None
397
+ num_query_groups ................................ 8
398
+ num_virtual_stages_per_pipeline_rank ............ None
399
+ num_workers ..................................... 2
400
+ object_storage_cache_path ....................... None
401
+ one_logger_async ................................ False
402
+ one_logger_project .............................. megatron-lm
403
+ one_logger_run_name ............................. None
404
+ onnx_safe ....................................... None
405
+ openai_gelu ..................................... False
406
+ optimizer ....................................... adam
407
+ optimizer_cpu_offload ........................... False
408
+ optimizer_offload_fraction ...................... 1.0
409
+ output_bert_embeddings .......................... False
410
+ overlap_cpu_optimizer_d2h_h2d ................... False
411
+ overlap_grad_reduce ............................. True
412
+ overlap_moe_expert_parallel_comm ................ False
413
+ overlap_p2p_comm ................................ False
414
+ overlap_p2p_comm_warmup_flush ................... False
415
+ overlap_param_gather ............................ True
416
+ overlap_param_gather_with_optimizer_step ........ False
417
+ override_opt_param_scheduler .................... False
418
+ params_dtype .................................... torch.bfloat16
419
+ patch_dim ....................................... 16
420
+ per_split_data_args_path ........................ None
421
+ perform_initialization .......................... True
422
+ pin_cpu_grads ................................... True
423
+ pin_cpu_params .................................. True
424
+ pipeline_model_parallel_comm_backend ............ None
425
+ pipeline_model_parallel_layout .................. None
426
+ pipeline_model_parallel_size .................... 1
427
+ position_embedding_type ......................... rope
428
+ pretrained_checkpoint ........................... None
429
+ profile ......................................... False
430
+ profile_ranks ................................... [0]
431
+ profile_step_end ................................ 12
432
+ profile_step_start .............................. 10
433
+ q_lora_rank ..................................... None
434
+ qk_head_dim ..................................... 128
435
+ qk_l2_norm ...................................... False
436
+ qk_layernorm .................................... False
437
+ qk_pos_emb_head_dim ............................. 64
438
+ query_in_block_prob ............................. 0.1
439
+ rampup_batch_size ............................... None
440
+ rank ............................................ 0
441
+ recompute_granularity ........................... None
442
+ recompute_method ................................ None
443
+ recompute_modules ............................... None
444
+ recompute_num_layers ............................ None
445
+ record_memory_history ........................... False
446
+ relative_attention_max_distance ................. 128
447
+ relative_attention_num_buckets .................. 32
448
+ replication ..................................... False
449
+ replication_factor .............................. 2
450
+ replication_jump ................................ None
451
+ rerun_mode ...................................... validate_results
452
+ reset_attention_mask ............................ False
453
+ reset_position_ids .............................. False
454
+ result_rejected_tracker_filename ................ None
455
+ retriever_report_topk_accuracies ................ []
456
+ retriever_score_scaling ......................... False
457
+ retriever_seq_length ............................ 256
458
+ retro_add_retriever ............................. False
459
+ retro_attention_gate ............................ 1
460
+ retro_cyclic_train_iters ........................ None
461
+ retro_encoder_attention_dropout ................. 0.1
462
+ retro_encoder_hidden_dropout .................... 0.1
463
+ retro_encoder_layers ............................ 2
464
+ retro_num_neighbors ............................. 2
465
+ retro_num_retrieved_chunks ...................... 2
466
+ retro_project_dir ............................... None
467
+ retro_verify_neighbor_count ..................... True
468
+ reuse_grad_buf_for_mxfp8_param_ag ............... False
469
+ rope_scaling_factor ............................. 8.0
470
+ rope_type ....................................... None
471
+ rotary_base ..................................... 1000000
472
+ rotary_interleaved .............................. False
473
+ rotary_percent .................................. 1.0
474
+ rotary_scaling_factor ........................... 1.0
475
+ rotary_seq_len_interpolation_factor ............. None
476
+ run_workload_inspector_server ................... False
477
+ sample_rate ..................................... 1.0
478
+ save ............................................ None
479
+ save_interval ................................... None
480
+ save_retain_interval ............................ None
481
+ scatter_gather_tensors_in_pipeline .............. True
482
+ seed ............................................ 1234
483
+ seq_length ...................................... 4096
484
+ sequence_parallel ............................... False
485
+ sft ............................................. False
486
+ sft_tokenizer_prompt_format ..................... nemotron-h-aligned
487
+ sgd_momentum .................................... 0.9
488
+ short_seq_prob .................................. 0.1
489
+ skip_train ...................................... False
490
+ skipped_train_samples ........................... 0
491
+ spec ............................................ None
492
+ split ........................................... 100,0,0
493
+ squared_relu .................................... False
494
+ start_weight_decay .............................. 0.1
495
+ straggler_ctrlr_port ............................ 65535
496
+ straggler_minmax_count .......................... 1
497
+ suggested_communication_unit_size ............... None
498
+ swiglu .......................................... True
499
+ swin_backbone_type .............................. tiny
500
+ symmetric_ar_type ............................... None
501
+ te_rng_tracker .................................. True
502
+ tensor_model_parallel_size ...................... 1
503
+ tensorboard_dir ................................. None
504
+ tensorboard_log_interval ........................ 1
505
+ tensorboard_queue_size .......................... 1000
506
+ test_data_path .................................. None
507
+ test_mode ....................................... False
508
+ tiktoken_num_special_tokens ..................... 1000
509
+ tiktoken_pattern ................................ None
510
+ tiktoken_special_tokens ......................... None
511
+ timing_log_level ................................ 0
512
+ timing_log_option ............................... minmax
513
+ titles_data_path ................................ None
514
+ tokenizer_model ................................. None
515
+ tokenizer_type .................................. GPT2BPETokenizer
516
+ torch_fsdp2_reshard_after_forward ............... True
517
+ tp_comm_bootstrap_backend ....................... nccl
518
+ tp_comm_bulk_dgrad .............................. True
519
+ tp_comm_bulk_wgrad .............................. True
520
+ tp_comm_overlap ................................. False
521
+ tp_comm_overlap_ag .............................. True
522
+ tp_comm_overlap_cfg ............................. None
523
+ tp_comm_overlap_rs .............................. True
524
+ tp_comm_overlap_rs_dgrad ........................ False
525
+ tp_comm_split_ag ................................ True
526
+ tp_comm_split_rs ................................ True
527
+ train_data_path ................................. None
528
+ train_iters ..................................... 12
529
+ train_samples ................................... None
530
+ train_sync_interval ............................. None
531
+ transformer_impl ................................ transformer_engine
532
+ transformer_pipeline_model_parallel_size ........ 1
533
+ untie_embeddings_and_output_weights ............. True
534
+ use_checkpoint_args ............................. False
535
+ use_checkpoint_opt_param_scheduler .............. False
536
+ use_cpu_initialization .......................... None
537
+ use_custom_fsdp ................................. False
538
+ use_dist_ckpt ................................... True
539
+ use_dist_ckpt_deprecated ........................ False
540
+ use_distributed_optimizer ....................... True
541
+ use_flash_attn .................................. False
542
+ use_legacy_models ............................... False
543
+ use_mp_args_from_checkpoint_args ................ False
544
+ use_one_sent_docs ............................... False
545
+ use_persistent_ckpt_worker ...................... False
546
+ use_precision_aware_optimizer ................... False
547
+ use_pytorch_profiler ............................ False
548
+ use_ring_exchange_p2p ........................... False
549
+ use_rope_scaling ................................ False
550
+ use_rotary_position_embeddings .................. False
551
+ use_sharp ....................................... False
552
+ use_tokenizer_model_from_checkpoint_args ........ True
553
+ use_torch_fsdp2 ................................. False
554
+ use_torch_optimizer_for_cpu_offload ............. False
555
+ use_tp_pp_dp_mapping ............................ False
556
+ v_head_dim ...................................... 128
557
+ valid_data_path ................................. None
558
+ variable_seq_lengths ............................ False
559
+ virtual_pipeline_model_parallel_size ............ None
560
+ vision_backbone_type ............................ vit
561
+ vision_pretraining .............................. False
562
+ vision_pretraining_type ......................... classify
563
+ vocab_extra_ids ................................. 0
564
+ vocab_file ...................................... /workspace/data/wikitext/gpt2-vocab.json
565
+ vocab_size ...................................... None
566
+ wandb_exp_name ..................................
567
+ wandb_project ...................................
568
+ wandb_save_dir ..................................
569
+ weight_decay .................................... 0.1
570
+ weight_decay_incr_style ......................... constant
571
+ wgrad_deferral_limit ............................ 0
572
+ world_size ...................................... 8
573
+ yaml_cfg ........................................ None
574
+ -------------------- end of arguments ---------------------
575
+ INFO:megatron.core.num_microbatches_calculator:setting number of microbatches to constant 2
576
+ > building GPT2BPETokenizer tokenizer ...
577
+ > padded vocab (size: 50257) with 47 dummy tokens (new size: 50304)
578
+ WARNING:megatron.core.rerun_state_machine:RerunStateMachine initialized in mode RerunMode.VALIDATE_RESULTS
579
+ > initializing torch distributed ...
580
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
581
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
582
+ [Gloo] Rank [Gloo] Rank [Gloo] Rank 12 is connected to 0 is connected to 7 is connected to 7 peer ranks. 7 peer ranks. Expected number of connected peer ranks is : peer ranks. Expected number of connected peer ranks is : 7Expected number of connected peer ranks is : 77
583
+
584
+
585
+ [Gloo] Rank 4 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
586
+ [Gloo] Rank 5 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
587
+ [Gloo] Rank 7 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
588
+ [Gloo] Rank 3 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
589
+ [Gloo] Rank 6 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
590
+ [Gloo] Rank 2 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
591
+ [Gloo] Rank 1 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
592
+ [Gloo] Rank 0 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
593
+ [Gloo] Rank 3 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
594
+ [Gloo] Rank 4 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
595
+ [Gloo] Rank 7 is connected to 7 peer ranks. [Gloo] Rank Expected number of connected peer ranks is : 5[Gloo] Rank 7 is connected to 6
596
+ 7 is connected to peer ranks. 7Expected number of connected peer ranks is : peer ranks. 7Expected number of connected peer ranks is :
597
+ 7
598
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
599
+ > initialized tensor model parallel with size 1
600
+ > initialized pipeline model parallel with size 1
601
+ > setting random seeds to 1234 ...
602
+ [Gloo] Rank [Gloo] Rank 00 is connected to is connected to 00 peer ranks. peer ranks. Expected number of connected peer ranks is : Expected number of connected peer ranks is : 00
603
+
604
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
605
+ [Gloo] Rank [Gloo] Rank 00 is connected to is connected to 00 peer ranks. peer ranks. Expected number of connected peer ranks is : Expected number of connected peer ranks is : 00
606
+
607
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
608
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
609
+ [rank2]:[W830 07:16:58.382028938 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 2] using GPU 2 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
610
+ [rank1]:[W830 07:16:58.382055297 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 1] using GPU 1 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
611
+ [rank3]:[W830 07:16:58.382100618 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 3] using GPU 3 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
612
+ [rank5]:[W830 07:16:58.385683384 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 5] using GPU 5 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
613
+ [rank7]:[W830 07:16:58.385745662 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 7] using GPU 7 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
614
+ [rank6]:[W830 07:16:58.387488833 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 6] using GPU 6 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
615
+ [rank4]:[W830 07:16:58.387506428 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 4] using GPU 4 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
616
+ > compiling dataset index builder ...
617
+ make: Entering directory '/workspace/Megatron-UltraEP/megatron/core/datasets'
618
+ make: Nothing to be done for 'default'.
619
+ make: Leaving directory '/workspace/Megatron-UltraEP/megatron/core/datasets'
620
+ >>> done with dataset index builder. Compilation time: 0.034 seconds
621
+ WARNING: constraints for invoking optimized fused softmax kernel are not met. We default back to unfused kernel invocations.
622
+ > compiling and loading fused kernels ...
623
+ [rank0]:[W830 07:16:58.563298755 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 0] using GPU 0 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
624
+ >>> done with compiling and loading fused kernels. Compilation time: 3.652 seconds
625
+ time to initialize megatron (seconds): 6.876
626
+ [after megatron is initialized] datetime: 2026-08-30 07:17:03
627
+ building GPT model ...
628
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
629
+ warnings.warn(
630
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
631
+ warnings.warn(
632
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
633
+ warnings.warn(
634
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
635
+ warnings.warn(
636
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
637
+ warnings.warn(
638
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
639
+ warnings.warn(
640
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
641
+ warnings.warn(
642
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
643
+ warnings.warn(
644
+ [INFO] Use MNNVL fabric: False
645
+ [WARN] MAX_NUM_NVL_PEERS is not set. Using detected value 8.
646
+ NVSHMEM v3.7.2
647
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
648
+
649
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
650
+
651
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
652
+
653
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
654
+
655
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
656
+
657
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
658
+
659
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
660
+
661
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
662
+
663
+ UltraEP: NVL domain size <= 8; forcing weight_sync plan mode to direct (was adaptive).
664
+ > number of parameters on (tensor, pipeline) model parallel rank (0, 0): 1687818240
665
+ INFO:megatron.core.distributed.distributed_data_parallel:Setting up DistributedDataParallel with config DistributedDataParallelConfig(grad_reduce_in_fp32=True, overlap_grad_reduce=True, overlap_param_gather=True, align_param_gather=False, use_distributed_optimizer=True, num_distributed_optimizer_instances=1, check_for_nan_in_grad=False, check_for_large_grads=False, bucket_size=40000000, pad_buckets_for_high_nccl_busbw=False, average_in_collective=False, fp8_param_gather=False, reuse_grad_buf_for_mxfp8_param_ag=False, use_custom_fsdp=False, data_parallel_sharding_strategy='no_shard', gradient_reduce_div_fusion=True, suggested_communication_unit_size=None, preserve_fp32_weights=True, keep_fp8_transpose_cache_when_using_custom_fsdp=False, nccl_ub=False, fsdp_double_buffer=False)
666
+ INFO:megatron.core.distributed.param_and_grad_buffer:Number of buckets for gradient all-reduce / reduce-scatter: 18
667
+ Params for bucket 1 (206045184 elements, 206045184 padded size):
668
+ module.output_layer.weight
669
+ Params for bucket 2 (41988096 elements, 41988096 padded size):
670
+ module.decoder.final_layernorm.weight
671
+ module.decoder.layers.15.mlp.router.weight
672
+ module.decoder.layers.15.self_attention.linear_qkv.layer_norm_weight
673
+ module.decoder.layers.15.self_attention.linear_proj.weight
674
+ module.decoder.layers.15.pre_mlp_layernorm.weight
675
+ module.decoder.layers.15.self_attention.linear_qkv.weight
676
+ Params for bucket 3 (41984000 elements, 41984000 padded size):
677
+ module.decoder.layers.14.mlp.router.weight
678
+ module.decoder.layers.14.self_attention.linear_proj.weight
679
+ module.decoder.layers.14.self_attention.linear_qkv.layer_norm_weight
680
+ module.decoder.layers.14.pre_mlp_layernorm.weight
681
+ module.decoder.layers.14.self_attention.linear_qkv.weight
682
+ Params for bucket 4 (41984000 elements, 41984000 padded size):
683
+ module.decoder.layers.13.mlp.router.weight
684
+ module.decoder.layers.13.self_attention.linear_qkv.layer_norm_weight
685
+ module.decoder.layers.13.self_attention.linear_proj.weight
686
+ module.decoder.layers.13.self_attention.linear_qkv.weight
687
+ module.decoder.layers.13.pre_mlp_layernorm.weight
688
+ Params for bucket 5 (41984000 elements, 41984000 padded size):
689
+ module.decoder.layers.12.mlp.router.weight
690
+ module.decoder.layers.12.self_attention.linear_proj.weight
691
+ module.decoder.layers.12.self_attention.linear_qkv.layer_norm_weight
692
+ module.decoder.layers.12.self_attention.linear_qkv.weight
693
+ module.decoder.layers.12.pre_mlp_layernorm.weight
694
+ Params for bucket 6 (41984000 elements, 41984000 padded size):
695
+ module.decoder.layers.11.mlp.router.weight
696
+ module.decoder.layers.11.self_attention.linear_qkv.layer_norm_weight
697
+ module.decoder.layers.11.self_attention.linear_proj.weight
698
+ module.decoder.layers.11.self_attention.linear_qkv.weight
699
+ module.decoder.layers.11.pre_mlp_layernorm.weight
700
+ Params for bucket 7 (41984000 elements, 41984000 padded size):
701
+ module.decoder.layers.10.self_attention.linear_qkv.weight
702
+ module.decoder.layers.10.pre_mlp_layernorm.weight
703
+ module.decoder.layers.10.mlp.router.weight
704
+ module.decoder.layers.10.self_attention.linear_qkv.layer_norm_weight
705
+ module.decoder.layers.10.self_attention.linear_proj.weight
706
+ Params for bucket 8 (41984000 elements, 41984000 padded size):
707
+ module.decoder.layers.9.pre_mlp_layernorm.weight
708
+ module.decoder.layers.9.self_attention.linear_qkv.weight
709
+ module.decoder.layers.9.mlp.router.weight
710
+ module.decoder.layers.9.self_attention.linear_proj.weight
711
+ module.decoder.layers.9.self_attention.linear_qkv.layer_norm_weight
712
+ Params for bucket 9 (41984000 elements, 41984000 padded size):
713
+ module.decoder.layers.8.self_attention.linear_qkv.weight
714
+ module.decoder.layers.8.pre_mlp_layernorm.weight
715
+ module.decoder.layers.8.mlp.router.weight
716
+ module.decoder.layers.8.self_attention.linear_qkv.layer_norm_weight
717
+ module.decoder.layers.8.self_attention.linear_proj.weight
718
+ Params for bucket 10 (41984000 elements, 41984000 padded size):
719
+ module.decoder.layers.7.pre_mlp_layernorm.weight
720
+ module.decoder.layers.7.self_attention.linear_qkv.weight
721
+ module.decoder.layers.7.mlp.router.weight
722
+ module.decoder.layers.7.self_attention.linear_proj.weight
723
+ module.decoder.layers.7.self_attention.linear_qkv.layer_norm_weight
724
+ Params for bucket 11 (41984000 elements, 41984000 padded size):
725
+ module.decoder.layers.6.mlp.router.weight
726
+ module.decoder.layers.6.self_attention.linear_proj.weight
727
+ module.decoder.layers.6.self_attention.linear_qkv.layer_norm_weight
728
+ module.decoder.layers.6.pre_mlp_layernorm.weight
729
+ module.decoder.layers.6.self_attention.linear_qkv.weight
730
+ Params for bucket 12 (41984000 elements, 41984000 padded size):
731
+ module.decoder.layers.5.self_attention.linear_qkv.weight
732
+ module.decoder.layers.5.self_attention.linear_proj.weight
733
+ module.decoder.layers.5.pre_mlp_layernorm.weight
734
+ module.decoder.layers.5.mlp.router.weight
735
+ module.decoder.layers.5.self_attention.linear_qkv.layer_norm_weight
736
+ Params for bucket 13 (41984000 elements, 41984000 padded size):
737
+ module.decoder.layers.4.mlp.router.weight
738
+ module.decoder.layers.4.self_attention.linear_qkv.layer_norm_weight
739
+ module.decoder.layers.4.self_attention.linear_proj.weight
740
+ module.decoder.layers.4.self_attention.linear_qkv.weight
741
+ module.decoder.layers.4.pre_mlp_layernorm.weight
742
+ Params for bucket 14 (41984000 elements, 41984000 padded size):
743
+ module.decoder.layers.3.mlp.router.weight
744
+ module.decoder.layers.3.self_attention.linear_proj.weight
745
+ module.decoder.layers.3.self_attention.linear_qkv.layer_norm_weight
746
+ module.decoder.layers.3.pre_mlp_layernorm.weight
747
+ module.decoder.layers.3.self_attention.linear_qkv.weight
748
+ Params for bucket 15 (41984000 elements, 41984000 padded size):
749
+ module.decoder.layers.2.mlp.router.weight
750
+ module.decoder.layers.2.self_attention.linear_qkv.layer_norm_weight
751
+ module.decoder.layers.2.self_attention.linear_proj.weight
752
+ module.decoder.layers.2.self_attention.linear_qkv.weight
753
+ module.decoder.layers.2.pre_mlp_layernorm.weight
754
+ Params for bucket 16 (41984000 elements, 41984000 padded size):
755
+ module.decoder.layers.1.pre_mlp_layernorm.weight
756
+ module.decoder.layers.1.mlp.router.weight
757
+ module.decoder.layers.1.self_attention.linear_qkv.layer_norm_weight
758
+ module.decoder.layers.1.self_attention.linear_qkv.weight
759
+ module.decoder.layers.1.self_attention.linear_proj.weight
760
+ Params for bucket 17 (41984000 elements, 41984000 padded size):
761
+ module.decoder.layers.0.mlp.router.weight
762
+ module.decoder.layers.0.self_attention.linear_qkv.weight
763
+ module.decoder.layers.0.pre_mlp_layernorm.weight
764
+ module.decoder.layers.0.self_attention.linear_qkv.layer_norm_weight
765
+ module.decoder.layers.0.self_attention.linear_proj.weight
766
+ Params for bucket 18 (206045184 elements, 206045184 padded size):
767
+ module.embedding.word_embeddings.weight
768
+ INFO:megatron.core.distributed.param_and_grad_buffer:Number of buckets for gradient all-reduce / reduce-scatter: 7
769
+ Params for bucket 1 (44040192 elements, 44040192 padded size):
770
+ module.decoder.layers.14.mlp.experts.linear_fc1.weight0
771
+ module.decoder.layers.15.mlp.experts.linear_fc1.weight0
772
+ module.decoder.layers.15.mlp.experts.linear_fc2.weight0
773
+ module.decoder.layers.13.mlp.experts.linear_fc2.weight0
774
+ module.decoder.layers.14.mlp.experts.linear_fc2.weight0
775
+ Params for bucket 2 (50331648 elements, 50331648 padded size):
776
+ module.decoder.layers.13.mlp.experts.linear_fc1.weight0
777
+ module.decoder.layers.11.mlp.experts.linear_fc1.weight0
778
+ module.decoder.layers.12.mlp.experts.linear_fc1.weight0
779
+ module.decoder.layers.11.mlp.experts.linear_fc2.weight0
780
+ module.decoder.layers.12.mlp.experts.linear_fc2.weight0
781
+ Params for bucket 3 (44040192 elements, 44040192 padded size):
782
+ module.decoder.layers.8.mlp.experts.linear_fc2.weight0
783
+ module.decoder.layers.9.mlp.experts.linear_fc2.weight0
784
+ module.decoder.layers.10.mlp.experts.linear_fc2.weight0
785
+ module.decoder.layers.10.mlp.experts.linear_fc1.weight0
786
+ module.decoder.layers.9.mlp.experts.linear_fc1.weight0
787
+ Params for bucket 4 (50331648 elements, 50331648 padded size):
788
+ module.decoder.layers.6.mlp.experts.linear_fc1.weight0
789
+ module.decoder.layers.7.mlp.experts.linear_fc2.weight0
790
+ module.decoder.layers.8.mlp.experts.linear_fc1.weight0
791
+ module.decoder.layers.7.mlp.experts.linear_fc1.weight0
792
+ module.decoder.layers.6.mlp.experts.linear_fc2.weight0
793
+ Params for bucket 5 (44040192 elements, 44040192 padded size):
794
+ module.decoder.layers.5.mlp.experts.linear_fc1.weight0
795
+ module.decoder.layers.4.mlp.experts.linear_fc1.weight0
796
+ module.decoder.layers.5.mlp.experts.linear_fc2.weight0
797
+ module.decoder.layers.4.mlp.experts.linear_fc2.weight0
798
+ module.decoder.layers.3.mlp.experts.linear_fc2.weight0
799
+ Params for bucket 6 (50331648 elements, 50331648 padded size):
800
+ module.decoder.layers.3.mlp.experts.linear_fc1.weight0
801
+ module.decoder.layers.1.mlp.experts.linear_fc2.weight0
802
+ module.decoder.layers.2.mlp.experts.linear_fc1.weight0
803
+ module.decoder.layers.2.mlp.experts.linear_fc2.weight0
804
+ module.decoder.layers.1.mlp.experts.linear_fc1.weight0
805
+ Params for bucket 7 (18874368 elements, 18874368 padded size):
806
+ module.decoder.layers.0.mlp.experts.linear_fc2.weight0
807
+ module.decoder.layers.0.mlp.experts.linear_fc1.weight0
808
+ INFO:megatron.core.optimizer:Setting up optimizer with config OptimizerConfig(optimizer='adam', lr=0.0001, min_lr=1e-05, decoupled_lr=None, decoupled_min_lr=None, weight_decay=0.1, fp8_recipe='mxfp8', fp16=False, bf16=True, reuse_grad_buf_for_mxfp8_param_ag=False, params_dtype=torch.bfloat16, use_precision_aware_optimizer=False, store_param_remainders=True, main_grads_dtype=torch.float32, main_params_dtype=torch.float32, exp_avg_dtype=torch.float32, exp_avg_sq_dtype=torch.float32, loss_scale=None, initial_loss_scale=4294967296, min_loss_scale=1.0, loss_scale_window=1000, hysteresis=2, adam_beta1=0.9, adam_beta2=0.999, adam_eps=1e-08, sgd_momentum=0.9, use_distributed_optimizer=True, overlap_param_gather_with_optimizer_step=False, optimizer_cpu_offload=False, optimizer_offload_fraction=1.0, use_torch_optimizer_for_cpu_offload=False, overlap_cpu_optimizer_d2h_h2d=False, pin_cpu_grads=True, pin_cpu_params=True, clip_grad=1.0, log_num_zeros_in_grad=False, barrier_with_L1_time=True, timers=<megatron.core.timers.Timers object at 0x75f2b19ecce0>, config_logger_dir='')
809
+ INFO:megatron.core.optimizer_param_scheduler:> learning rate decay style: cosine
810
+ [after model, optimizer, and learning rate scheduler are built] datetime: 2026-08-30 07:17:07
811
+ > building train, validation, and test datasets ...
812
+ > datasets target sizes (minimum size):
813
+ train: 192
814
+ validation: 0
815
+ test: 0
816
+ INFO:megatron.core.datasets.blended_megatron_dataset_config:Let split_matrix = [(0, 1.0), None, None]
817
+ > building train, validation, and test datasets for GPT ...
818
+ INFO:megatron.core.datasets.blended_megatron_dataset_builder:Building GPTDataset splits with sizes=(192, 0, 0) and config=GPTDatasetConfig(random_seed=1234, sequence_length=4096, blend=(['/workspace/data/wikitext/wikitext103_gpt2_text_document'], None), blend_per_split=None, split='100,0,0', split_matrix=[(0, 1.0), None, None], num_dataset_builder_threads=1, path_to_cache=None, mmap_bin_files=True, mock=False, tokenizer=<megatron.training.tokenizer.tokenizer._GPT2BPETokenizer object at 0x75f2b1b52ae0>, mid_level_dataset_surplus=0.005, reset_position_ids=False, reset_attention_mask=False, eod_mask_loss=False, create_attention_mask=True, drop_last_partial_validation_sequence=True, add_extra_token_to_sequence=True, object_storage_cache_path=None)
819
+ INFO:megatron.core.datasets.indexed_dataset:Load the _IndexReader from /workspace/data/wikitext/wikitext103_gpt2_text_document.idx
820
+ INFO:megatron.core.datasets.indexed_dataset: Extract the sequence lengths
821
+ INFO:megatron.core.datasets.indexed_dataset: Extract the sequence pointers
822
+ INFO:megatron.core.datasets.indexed_dataset: Extract the document indices
823
+ INFO:megatron.core.datasets.indexed_dataset:> total number of sequences: 1165029
824
+ INFO:megatron.core.datasets.indexed_dataset:> total number of documents: 1801350
825
+ INFO:megatron.core.datasets.gpt_dataset:Load the GPTDataset train indices
826
+ INFO:megatron.core.datasets.gpt_dataset: Load the document index from 44ed1ee5bcb9e5b030387b8d27a6c406-GPTDataset-train-document_index.npy
827
+ INFO:megatron.core.datasets.gpt_dataset: Load the sample index from 44ed1ee5bcb9e5b030387b8d27a6c406-GPTDataset-train-sample_index.npy
828
+ INFO:megatron.core.datasets.gpt_dataset: Load the shuffle index from 44ed1ee5bcb9e5b030387b8d27a6c406-GPTDataset-train-shuffle_index.npy
829
+ INFO:megatron.core.datasets.gpt_dataset:> total number of samples: 29073
830
+ > finished creating GPT datasets ...
831
+ [after dataloaders are built] datetime: 2026-08-30 07:17:07
832
+ done with setup ...
833
+ (min, max) time across ranks (ms):
834
+ model-and-optimizer-setup ......................: (3279.61, 3279.82)
835
+ train/valid/test-data-iterators-setup ..........: (70.36, 90.21)
836
+ training ...
837
+ Overwriting rerun_state_machine.current_iteration from -1 to 0...
838
+ [before the start of training step] datetime: 2026-08-30 07:17:07
839
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
840
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
841
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
842
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
843
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
844
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
845
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
846
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
847
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
848
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
849
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
850
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
851
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
852
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
853
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
854
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
855
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
856
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
857
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
858
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
859
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
860
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
861
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
862
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
863
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
864
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
865
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
866
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
867
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
868
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
869
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
870
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
871
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
872
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
873
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
874
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
875
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
876
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
877
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
878
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
879
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
880
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
881
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
882
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
883
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
884
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
885
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
886
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
887
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
888
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
889
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
890
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
891
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
892
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
893
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
894
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
895
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
896
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
897
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
898
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
899
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
900
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
901
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
902
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
903
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
904
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
905
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
906
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
907
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
908
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
909
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
910
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
911
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
912
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
913
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
914
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
915
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
916
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
917
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
918
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
919
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
920
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
921
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
922
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
923
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
924
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
925
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
926
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
927
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
928
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
929
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
930
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
931
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
932
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
933
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
934
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
935
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
936
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
937
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
938
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
939
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
940
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
941
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
942
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
943
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
944
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
945
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
946
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
947
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
948
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
949
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
950
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
951
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
952
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
953
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
954
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
955
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
956
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
957
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
958
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
959
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
960
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
961
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
962
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
963
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
964
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
965
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
966
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
967
+ [2026-08-30 07:17:32] iteration 1/ 12 | consumed samples: 16 | elapsed time per iteration (ms): 25410.3 | throughput per GPU (TFLOP/s/GPU): 3.4 | learning rate: 2.000000E-05 | global batch size: 16 | lm loss: 1.103286E+01 | loss scale: 1.0 | grad norm: 44.355 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14592 | min safe capacity: 1.7812 |Number of parameters in transformer block in billions: 3.09
968
+
969
+ Number of parameters in embedding layers in billions: 0.41
970
+ Total number of parameters in billions: 3.50
971
+ Number of parameters in most loaded shard in billions: 3.4994
972
+ Theoretical memory footprints: weight and optimizer=25029.43 MB
973
+ [Rank 0] (after 1 iterations) memory (MB) | allocated: 14864.61376953125 | max allocated: 22156.96533203125 | reserved: 22904.0 | max reserved: 22904.0
974
+ [2026-08-30 07:17:33] iteration 2/ 12 | consumed samples: 32 | elapsed time per iteration (ms): 269.0 | throughput per GPU (TFLOP/s/GPU): 319.7 | learning rate: 4.000000E-05 | global batch size: 16 | lm loss: 1.102278E+01 | loss scale: 1.0 | grad norm: 46.283 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14848 | min safe capacity: 1.8125 |
975
+ [2026-08-30 07:17:33] iteration 3/ 12 | consumed samples: 48 | elapsed time per iteration (ms): 255.2 | throughput per GPU (TFLOP/s/GPU): 337.0 | learning rate: 6.000000E-05 | global batch size: 16 | lm loss: 1.037312E+01 | loss scale: 1.0 | grad norm: 215.157 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14848 | min safe capacity: 1.8125 |
976
+ Capture CUDA graph for training!!!
977
+ CUDA graph capture done!!!
978
+ [2026-08-30 07:17:34] iteration 4/ 12 | consumed samples: 64 | elapsed time per iteration (ms): 958.4 | throughput per GPU (TFLOP/s/GPU): 89.7 | learning rate: 8.000000E-05 | global batch size: 16 | lm loss: 1.313629E+01 | loss scale: 1.0 | grad norm: 41.160 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14848 | min safe capacity: 1.8125 |
979
+ [2026-08-30 07:17:34] iteration 5/ 12 | consumed samples: 80 | elapsed time per iteration (ms): 177.2 | throughput per GPU (TFLOP/s/GPU): 485.3 | learning rate: 1.000000E-04 | global batch size: 16 | lm loss: 1.037849E+01 | loss scale: 1.0 | grad norm: 15.996 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14848 | min safe capacity: 1.8125 |
980
+ [2026-08-30 07:17:34] iteration 6/ 12 | consumed samples: 96 | elapsed time per iteration (ms): 176.6 | throughput per GPU (TFLOP/s/GPU): 486.9 | learning rate: 7.750000E-05 | global batch size: 16 | lm loss: 1.002978E+01 | loss scale: 1.0 | grad norm: 9.280 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14848 | min safe capacity: 1.8125 |
981
+ [2026-08-30 07:17:34] iteration 7/ 12 | consumed samples: 112 | elapsed time per iteration (ms): 176.4 | throughput per GPU (TFLOP/s/GPU): 487.6 | learning rate: 3.250000E-05 | global batch size: 16 | lm loss: 9.371683E+00 | loss scale: 1.0 | grad norm: 14.666 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14848 | min safe capacity: 1.8125 |
982
+ [2026-08-30 07:17:34] iteration 8/ 12 | consumed samples: 128 | elapsed time per iteration (ms): 176.5 | throughput per GPU (TFLOP/s/GPU): 487.1 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.945133E+00 | loss scale: 1.0 | grad norm: 6.151 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14848 | min safe capacity: 1.8125 |
983
+ [2026-08-30 07:17:35] iteration 9/ 12 | consumed samples: 144 | elapsed time per iteration (ms): 177.1 | throughput per GPU (TFLOP/s/GPU): 485.6 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.781912E+00 | loss scale: 1.0 | grad norm: 5.229 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14848 | min safe capacity: 1.8125 |
984
+ [2026-08-30 07:17:35] iteration 10/ 12 | consumed samples: 160 | elapsed time per iteration (ms): 176.8 | throughput per GPU (TFLOP/s/GPU): 486.3 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.691033E+00 | loss scale: 1.0 | grad norm: 5.449 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14848 | min safe capacity: 1.8125 |
985
+ [2026-08-30 07:17:35] iteration 11/ 12 | consumed samples: 176 | elapsed time per iteration (ms): 176.9 | throughput per GPU (TFLOP/s/GPU): 486.0 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.622260E+00 | loss scale: 1.0 | grad norm: 5.177 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14848 | min safe capacity: 1.8125 |
986
+ [2026-08-30 07:17:35] iteration 12/ 12 | consumed samples: 192 | elapsed time per iteration (ms): 177.3 | throughput per GPU (TFLOP/s/GPU): 485.1 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.521139E+00 | loss scale: 1.0 | grad norm: 4.770 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14848 | min safe capacity: 1.8125 |
987
+ [after training is done] datetime: 2026-08-30 07:17:35
988
+ [rank0]:[W830 07:17:36.122943666 ProcessGroupNCCL.cpp:1505] Warning: WARNING: destroy_process_group() was not called before program exit, which can leak resources. For more info, please see https://pytorch.org/docs/stable/distributed.html#shutdown (function operator())
989
+ [rank7]:[W830 07:17:38.278356363 ProcessGroupNCCL.cpp:1505] Warning: WARNING: destroy_process_group() was not called before program exit, which can leak resources. For more info, please see https://pytorch.org/docs/stable/distributed.html#shutdown (function operator())
990
+ [rank4]:[W830 07:17:38.292970631 ProcessGroupNCCL.cpp:1505] Warning: WARNING: destroy_process_group() was not called before program exit, which can leak resources. For more info, please see https://pytorch.org/docs/stable/distributed.html#shutdown (function operator())
991
+ [rank1]:[W830 07:17:38.294445197 ProcessGroupNCCL.cpp:1505] Warning: WARNING: destroy_process_group() was not called before program exit, which can leak resources. For more info, please see https://pytorch.org/docs/stable/distributed.html#shutdown (function operator())
992
+ [rank2]:[W830 07:17:38.295700220 ProcessGroupNCCL.cpp:1505] Warning: WARNING: destroy_process_group() was not called before program exit, which can leak resources. For more info, please see https://pytorch.org/docs/stable/distributed.html#shutdown (function operator())
993
+ [rank6]:[W830 07:17:38.573436033 ProcessGroupNCCL.cpp:1505] Warning: WARNING: destroy_process_group() was not called before program exit, which can leak resources. For more info, please see https://pytorch.org/docs/stable/distributed.html#shutdown (function operator())
994
+ [rank3]:[W830 07:17:38.753577982 ProcessGroupNCCL.cpp:1505] Warning: WARNING: destroy_process_group() was not called before program exit, which can leak resources. For more info, please see https://pytorch.org/docs/stable/distributed.html#shutdown (function operator())
995
+ [rank5]:[W830 07:17:38.842898039 ProcessGroupNCCL.cpp:1505] Warning: WARNING: destroy_process_group() was not called before program exit, which can leak resources. For more info, please see https://pytorch.org/docs/stable/distributed.html#shutdown (function operator())
996
+ W0830 07:17:38.911000 733201 torch/distributed/elastic/agent/server/api.py:719] Received 15 death signal, shutting down workers
997
+ W0830 07:17:38.911000 733201 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 733289 closing signal SIGTERM
998
+ W0830 07:17:38.912000 733201 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 733290 closing signal SIGTERM
999
+ W0830 07:17:38.913000 733201 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 733291 closing signal SIGTERM
1000
+ W0830 07:17:38.914000 733201 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 733292 closing signal SIGTERM
1001
+ W0830 07:17:38.915000 733201 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 733293 closing signal SIGTERM
1002
+ W0830 07:17:38.916000 733201 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 733294 closing signal SIGTERM
1003
+ W0830 07:17:38.917000 733201 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 733295 closing signal SIGTERM
1004
+ W0830 07:17:38.918000 733201 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 733296 closing signal SIGTERM
1005
+ Traceback (most recent call last):
1006
+ File "/usr/local/bin/torchrun", line 8, in <module>
1007
+ sys.exit(main())
1008
+ ^^^^^^
1009
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/multiprocessing/errors/__init__.py", line 355, in wrapper
1010
+ return f(*args, **kwargs)
1011
+ ^^^^^^^^^^^^^^^^^^
1012
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/run.py", line 892, in main
1013
+ run(args)
1014
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/run.py", line 883, in run
1015
+ elastic_launch(
1016
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/launcher/api.py", line 139, in __call__
1017
+ return launch_agent(self._config, self._entrypoint, list(args))
1018
+ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1019
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/launcher/api.py", line 261, in launch_agent
1020
+ result = agent.run()
1021
+ ^^^^^^^^^^^
1022
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/metrics/api.py", line 138, in wrapper
1023
+ result = f(*args, **kwargs)
1024
+ ^^^^^^^^^^^^^^^^^^
1025
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/agent/server/api.py", line 711, in run
1026
+ result = self._invoke_run(role)
1027
+ ^^^^^^^^^^^^^^^^^^^^^^
1028
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/agent/server/api.py", line 870, in _invoke_run
1029
+ time.sleep(monitor_interval)
1030
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/multiprocessing/api.py", line 84, in _terminate_process_handler
1031
+ raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval)
1032
+ torch.distributed.elastic.multiprocessing.api.SignalException: Process 733201 got signal: 15
tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq4096__gbs16__cap2p40625/rep3/mem_samples.csv ADDED
@@ -0,0 +1,192 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 0, 0
2
+ 1, 0
3
+ 2, 0
4
+ 3, 0
5
+ 4, 0
6
+ 5, 0
7
+ 6, 0
8
+ 7, 0
9
+ 0, 0
10
+ 1, 0
11
+ 2, 0
12
+ 3, 0
13
+ 4, 0
14
+ 5, 0
15
+ 6, 0
16
+ 7, 0
17
+ 0, 2
18
+ 1, 1
19
+ 2, 1
20
+ 3, 1
21
+ 4, 1
22
+ 5, 1
23
+ 6, 1
24
+ 7, 1
25
+ 0, 4
26
+ 1, 4
27
+ 2, 4
28
+ 3, 4
29
+ 4, 1
30
+ 5, 1
31
+ 6, 1
32
+ 7, 1
33
+ 0, 4
34
+ 1, 4
35
+ 2, 4
36
+ 3, 4
37
+ 4, 4
38
+ 5, 4
39
+ 6, 4
40
+ 7, 4
41
+ 0, 4
42
+ 1, 629
43
+ 2, 629
44
+ 3, 4
45
+ 4, 629
46
+ 5, 629
47
+ 6, 629
48
+ 7, 629
49
+ 0, 889
50
+ 1, 881
51
+ 2, 883
52
+ 3, 889
53
+ 4, 889
54
+ 5, 897
55
+ 6, 923
56
+ 7, 899
57
+ 0, 1895
58
+ 1, 1895
59
+ 2, 1895
60
+ 3, 1895
61
+ 4, 1895
62
+ 5, 1895
63
+ 6, 1895
64
+ 7, 1895
65
+ 0, 2415
66
+ 1, 2427
67
+ 2, 2425
68
+ 3, 2419
69
+ 4, 2417
70
+ 5, 2435
71
+ 6, 2461
72
+ 7, 2429
73
+ 0, 14787
74
+ 1, 14823
75
+ 2, 14797
76
+ 3, 14839
77
+ 4, 14807
78
+ 5, 14817
79
+ 6, 14833
80
+ 7, 14859
81
+ 0, 15599
82
+ 1, 15627
83
+ 2, 15591
84
+ 3, 15603
85
+ 4, 15653
86
+ 5, 15631
87
+ 6, 15605
88
+ 7, 15601
89
+ 0, 16323
90
+ 1, 16321
91
+ 2, 16321
92
+ 3, 16321
93
+ 4, 16321
94
+ 5, 16321
95
+ 6, 16321
96
+ 7, 16321
97
+ 0, 16323
98
+ 1, 16321
99
+ 2, 16321
100
+ 3, 16321
101
+ 4, 16321
102
+ 5, 16321
103
+ 6, 16321
104
+ 7, 16321
105
+ 0, 16677
106
+ 1, 16675
107
+ 2, 16675
108
+ 3, 16675
109
+ 4, 16675
110
+ 5, 16675
111
+ 6, 16675
112
+ 7, 16675
113
+ 0, 25943
114
+ 1, 25941
115
+ 2, 25877
116
+ 3, 25941
117
+ 4, 25847
118
+ 5, 25941
119
+ 6, 25877
120
+ 7, 25941
121
+ 0, 26793
122
+ 1, 26791
123
+ 2, 26791
124
+ 3, 26791
125
+ 4, 26761
126
+ 5, 26791
127
+ 6, 26791
128
+ 7, 26791
129
+ 0, 26793
130
+ 1, 26791
131
+ 2, 26791
132
+ 3, 26791
133
+ 4, 26761
134
+ 5, 26791
135
+ 6, 26791
136
+ 7, 26791
137
+ 0, 28851
138
+ 1, 28845
139
+ 2, 28835
140
+ 3, 28853
141
+ 4, 28813
142
+ 5, 28863
143
+ 6, 28851
144
+ 7, 28877
145
+ 0, 29909
146
+ 1, 29909
147
+ 2, 29921
148
+ 3, 29917
149
+ 4, 29833
150
+ 5, 29933
151
+ 6, 29923
152
+ 7, 29943
153
+ 0, 31001
154
+ 1, 30981
155
+ 2, 31035
156
+ 3, 31009
157
+ 4, 30967
158
+ 5, 31013
159
+ 6, 31015
160
+ 7, 31031
161
+ 0, 38331
162
+ 1, 38319
163
+ 2, 38319
164
+ 3, 38319
165
+ 4, 38311
166
+ 5, 38317
167
+ 6, 38317
168
+ 7, 38319
169
+ 0, 37807
170
+ 1, 38057
171
+ 2, 38057
172
+ 3, 38057
173
+ 4, 38049
174
+ 5, 38055
175
+ 6, 38053
176
+ 7, 38057
177
+ 0, 34853
178
+ 1, 34967
179
+ 2, 34841
180
+ 3, 34967
181
+ 4, 34833
182
+ 5, 34577
183
+ 6, 35227
184
+ 7, 34967
185
+ 0, 1
186
+ 1, 193
187
+ 2, 34838
188
+ 3, 961
189
+ 4, 1
190
+ 5, 1
191
+ 6, 1
192
+ 7, 4
tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq4096__gbs16__cap2p40625/rep3/run.env ADDED
@@ -0,0 +1,70 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ backend=ultraep
2
+ config=PONLY
3
+ family=tierA
4
+ repo=/workspace/Megatron-UltraEP
5
+ branch=ultraep-bracing-regions-through-kernels
6
+ commit=03d57ef7
7
+ num_layers=16
8
+ hidden=4096
9
+ ffn_hidden=14336
10
+ moe_ffn_hidden=1536
11
+ num_experts=8
12
+ topk=2
13
+ experts_per_rank=1
14
+ expert_params=18874368
15
+ heads=32
16
+ kv_groups=8
17
+ kv_channels=
18
+ qk_layernorm=0
19
+ shared_ffn=0
20
+ shared_expert_overlap=0
21
+ moe_token_dispatcher=flex
22
+ mla=0
23
+ dense_first_layers=0
24
+ moe_layers=16
25
+ rotary_base=1000000
26
+ rotary_scaling=
27
+ qkv_bias=0
28
+ router_balancing=aux_loss
29
+ aux_loss_coeff=1e-2
30
+ router_pre_softmax=0
31
+ router_fusion=0
32
+ router_dtype=fp32
33
+ device_initiated_grouped_gemm=1
34
+ norm_epsilon=1e-5
35
+ seq_length=4096
36
+ micro_batch_size=1
37
+ global_batch_size=16
38
+ num_microbatches=2
39
+ assignments_per_rank=8192
40
+ mean_tokens_per_expert=8192
41
+ slot_policy=one
42
+ slots_per_rank=1
43
+ capacity_mode=static
44
+ capacity=2.40625
45
+ cuda_graph=1
46
+ probe_mode=0
47
+ async_grad_reduce=0
48
+ branch_role=timing
49
+ memory_dump_mode=none
50
+ memory_snapshot_path=
51
+ cuda_graph_warmup_steps=3
52
+ recompute=none
53
+ force_balance=0
54
+ train_iters=12
55
+ ep_size=8
56
+ echo_dispatch_overlap=1
57
+ echo_dispatch_recompute=0
58
+ ultraep_quota_min_tokens=1024
59
+ ultraep_weight_sync_join=
60
+ profile_enable=0
61
+ kernel_trace=0
62
+ fp8_mode=mxfp8
63
+ fp8_recipe=mxfp8
64
+ training_seed=1234
65
+ learning_rate=1e-4
66
+ min_learning_rate=1.0e-5
67
+ lr_warmup_iters=5
68
+ state_dump_enable=0
69
+ state_dump_dir=
70
+ moonep_assert_no_pending_reduce=0
tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq4096__gbs16__cap2p40625/rep3/train.log ADDED
@@ -0,0 +1,1032 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ==============================================================
2
+ PONLY / ultraep (slots/rank=1 policy=one)
3
+ E=8 K=2 H'=1536 m=1
4
+ L=16 seq=4096 gbs=16 (2 microbatches)
5
+ cap_mode=static cap=2.40625 cg=1 recompute=none
6
+ memory_dump=none snapshots=/workspace/tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq4096__gbs16__cap2p40625/rep3/memory_snapshots/snapshot.pickle
7
+ repo=/workspace/Megatron-UltraEP @ 03d57ef7
8
+ ==============================================================
9
+ W0830 07:17:44.406000 736138 torch/distributed/run.py:766]
10
+ W0830 07:17:44.406000 736138 torch/distributed/run.py:766] *****************************************
11
+ W0830 07:17:44.406000 736138 torch/distributed/run.py:766] Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed.
12
+ W0830 07:17:44.406000 736138 torch/distributed/run.py:766] *****************************************
13
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
14
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
15
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
16
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
17
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
18
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
19
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
20
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
21
+ WARNING: one_logger package is required to enable e2e metrics tracking. please go to https://confluence.nvidia.com/display/MLWFO/Package+Repositories for details to install it
22
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
23
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
24
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
25
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
26
+ using world size: 8, data-parallel size: 8, context-parallel size: 1, hierarchical context-parallel sizes: None, tensor-model-parallel size: 1, pipeline-model-parallel size: 1
27
+ WARNING: overriding default arguments for tokenizer_type:GPT2BPETokenizer with tokenizer_type:GPT2BPETokenizer
28
+ Number of virtual stages per pipeline stage: None
29
+ accumulate and all-reduce gradients in fp32 for bfloat16 data type.
30
+ using torch.bfloat16 for parameters ...
31
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
32
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
33
+ ------------------------ arguments ------------------------
34
+ account_for_embedding_in_pipeline_split ......... False
35
+ account_for_loss_in_pipeline_split .............. False
36
+ accumulate_allreduce_grads_in_fp32 .............. True
37
+ adam_beta1 ...................................... 0.9
38
+ adam_beta2 ...................................... 0.999
39
+ adam_eps ........................................ 1e-08
40
+ add_bias_linear ................................. False
41
+ add_position_embedding .......................... False
42
+ add_qkv_bias .................................... False
43
+ adlr_autoresume ................................. False
44
+ adlr_autoresume_interval ........................ 1000
45
+ align_grad_reduce ............................... True
46
+ align_param_gather .............................. False
47
+ app_tag_run_name ................................ None
48
+ app_tag_run_version ............................. 0.0.0
49
+ apply_layernorm_1p .............................. False
50
+ apply_query_key_layer_scaling ................... False
51
+ apply_residual_connection_post_layernorm ........ False
52
+ apply_rope_fusion ............................... True
53
+ async_save ...................................... None
54
+ async_tensor_model_parallel_allreduce ........... True
55
+ attention_backend ............................... AttnBackend.auto
56
+ attention_dropout ............................... 0.0
57
+ attention_softmax_in_fp32 ....................... False
58
+ auto_detect_ckpt_format ......................... False
59
+ barrier_with_L1_time ............................ True
60
+ bert_binary_head ................................ True
61
+ bert_embedder_type .............................. megatron
62
+ bert_load ....................................... None
63
+ bf16 ............................................ True
64
+ bias_dropout_fusion ............................. True
65
+ bias_gelu_fusion ................................ False
66
+ bias_swiglu_fusion .............................. True
67
+ biencoder_projection_dim ........................ 0
68
+ biencoder_shared_query_context_model ............ False
69
+ block_data_path ................................. None
70
+ calc_ft_timeouts ................................ False
71
+ calculate_per_token_loss ........................ False
72
+ check_for_large_grads ........................... False
73
+ check_for_nan_in_loss_and_grad .................. False
74
+ check_for_spiky_loss ............................ False
75
+ check_weight_hash_across_dp_replicas_interval ... None
76
+ ckpt_assume_constant_structure .................. False
77
+ ckpt_convert_format ............................. None
78
+ ckpt_convert_save ............................... None
79
+ ckpt_convert_update_legacy_dist_opt_format ...... False
80
+ ckpt_format ..................................... torch_dist
81
+ ckpt_fully_parallel_load ........................ False
82
+ ckpt_fully_parallel_save ........................ True
83
+ ckpt_fully_parallel_save_deprecated ............. False
84
+ ckpt_step ....................................... None
85
+ classes_fraction ................................ 1.0
86
+ clip_grad ....................................... 1.0
87
+ clone_scatter_output_in_embedding ............... True
88
+ config_logger_dir ...............................
89
+ consumed_train_samples .......................... 0
90
+ consumed_valid_samples .......................... 0
91
+ context_parallel_size ........................... 1
92
+ cp_comm_type .................................... ['p2p']
93
+ create_attention_mask_in_dataloader ............. True
94
+ cross_entropy_fusion_impl ....................... native
95
+ cross_entropy_loss_fusion ....................... False
96
+ cuda_graph_scope ................................ full_iteration
97
+ cuda_graph_warmup_steps ......................... 3
98
+ data_args_path .................................. None
99
+ data_cache_path ................................. None
100
+ data_parallel_random_init ....................... False
101
+ data_parallel_sharding_strategy ................. no_shard
102
+ data_parallel_size .............................. 8
103
+ data_path ....................................... ['/workspace/data/wikitext/wikitext103_gpt2_text_document']
104
+ data_per_class_fraction ......................... 1.0
105
+ data_sharding ................................... True
106
+ dataloader_type ................................. single
107
+ ddp_average_in_collective ....................... False
108
+ ddp_bucket_size ................................. None
109
+ ddp_num_buckets ................................. None
110
+ ddp_pad_buckets_for_high_nccl_busbw ............. False
111
+ decoder_first_pipeline_num_layers ............... None
112
+ decoder_last_pipeline_num_layers ................ None
113
+ decoder_num_layers .............................. None
114
+ decoder_seq_length .............................. None
115
+ decoupled_lr .................................... None
116
+ decoupled_min_lr ................................ None
117
+ decrease_batch_size_if_needed ................... False
118
+ defer_embedding_wgrad_compute ................... False
119
+ delay_wgrad_compute ............................. False
120
+ deprecated_use_mcore_models ..................... True
121
+ deterministic_mode .............................. False
122
+ dino_bottleneck_size ............................ 256
123
+ dino_freeze_last_layer .......................... 1
124
+ dino_head_hidden_size ........................... 2048
125
+ dino_local_crops_number ......................... 10
126
+ dino_local_img_size ............................. 96
127
+ dino_norm_last_layer ............................ False
128
+ dino_teacher_temp ............................... 0.07
129
+ dino_warmup_teacher_temp ........................ 0.04
130
+ dino_warmup_teacher_temp_epochs ................. 30
131
+ disable_bf16_reduced_precision_matmul ........... False
132
+ disable_mamba_mem_eff_path ...................... False
133
+ disable_straggler_on_startup .................... False
134
+ dist_ckpt_format_deprecated ..................... None
135
+ dist_ckpt_strictness ............................ assume_ok_unexpected
136
+ distribute_saved_activations .................... False
137
+ distributed_backend ............................. nccl
138
+ distributed_timeout_minutes ..................... 10
139
+ embedding_init_method_std ....................... None
140
+ embedding_path .................................. None
141
+ empty_unused_memory_level ....................... 0
142
+ enable_cuda_graph ............................... True
143
+ enable_experimental ............................. False
144
+ enable_ft_package ............................... False
145
+ enable_gloo_process_groups ...................... True
146
+ enable_msc ...................................... True
147
+ enable_one_logger ............................... True
148
+ encoder_num_layers .............................. 16
149
+ encoder_seq_length .............................. 4096
150
+ end_weight_decay ................................ 0.1
151
+ eod_mask_loss ................................... False
152
+ error_injection_rate ............................ 0
153
+ error_injection_type ............................ transient_error
154
+ eval_interval ................................... 1000000
155
+ eval_iters ...................................... 0
156
+ evidence_data_path .............................. None
157
+ exit_duration_in_mins ........................... None
158
+ exit_interval ................................... None
159
+ exit_on_missing_checkpoint ...................... False
160
+ exit_signal_handler ............................. False
161
+ exp_avg_dtype ................................... torch.float32
162
+ exp_avg_sq_dtype ................................ torch.float32
163
+ expert_model_parallel_size ...................... 8
164
+ expert_tensor_parallel_size ..................... 1
165
+ export_force_local_attention .................... False
166
+ export_kd_cfg ................................... None
167
+ export_kd_teacher_ckpt_format ................... None
168
+ export_kd_teacher_load .......................... None
169
+ export_kv_cache_quant ........................... False
170
+ export_legacy_megatron .......................... False
171
+ export_model_type ............................... GPTModel
172
+ export_moe_apply_probs_on_input ................. False
173
+ export_qk_l2_norm ............................... False
174
+ export_quant_cfg ................................ None
175
+ export_real_quant_cfg ........................... None
176
+ export_te_mcore_model ........................... False
177
+ external_cuda_graph ............................. False
178
+ ffn_hidden_size ................................. 14336
179
+ finetune ........................................ False
180
+ finetune_data_split ............................. train
181
+ finetune_hf_dataset ............................. None
182
+ first_last_layers_bf16 .......................... False
183
+ flash_decode .................................... False
184
+ fp16 ............................................ False
185
+ fp16_lm_cross_entropy ........................... False
186
+ fp32_residual_connection ........................ False
187
+ fp8 ............................................. e4m3
188
+ fp8_amax_compute_algo ........................... most_recent
189
+ fp8_amax_history_len ............................ 1
190
+ fp8_interval .................................... 1
191
+ fp8_margin ...................................... 0
192
+ fp8_param_gather ................................ False
193
+ fp8_recipe ...................................... mxfp8
194
+ fp8_wgrad ....................................... True
195
+ fsdp_double_buffer .............................. False
196
+ global_batch_size ............................... 16
197
+ grad_reduce_in_bf16 ............................. False
198
+ gradient_accumulation_fusion .................... True
199
+ gradient_reduce_div_fusion ...................... True
200
+ group_query_attention ........................... True
201
+ head_lr_mult .................................... 1.0
202
+ heterogeneous_layers_config_encoded_json ........ None
203
+ heterogeneous_layers_config_path ................ None
204
+ hidden_dropout .................................. 0.0
205
+ hidden_size ..................................... 4096
206
+ hierarchical_context_parallel_sizes ............. None
207
+ high_priority_stream_groups ..................... []
208
+ hybrid_attention_ratio .......................... 0.0
209
+ hybrid_mlp_ratio ................................ 0.0
210
+ hybrid_override_pattern ......................... None
211
+ hysteresis ...................................... 2
212
+ ict_head_size ................................... None
213
+ ict_load ........................................ None
214
+ img_h ........................................... 224
215
+ img_w ........................................... 224
216
+ indexer_batch_size .............................. 128
217
+ indexer_log_interval ............................ 1000
218
+ inference_batch_times_seqlen_threshold .......... -1
219
+ inference_dynamic_batching ...................... False
220
+ inference_dynamic_batching_buffer_guaranteed_fraction 0.2
221
+ inference_dynamic_batching_buffer_overflow_factor None
222
+ inference_dynamic_batching_buffer_size_gb ....... 40.0
223
+ inference_dynamic_batching_chunk_size ........... 256
224
+ inference_dynamic_batching_max_requests_override None
225
+ inference_dynamic_batching_max_tokens_override .. None
226
+ inference_dynamic_batching_num_cuda_graphs ...... 16
227
+ inference_max_batch_size ........................ 8
228
+ inference_max_seq_length ........................ 2560
229
+ inference_rng_tracker ........................... False
230
+ init_method_std ................................. 0.01
231
+ init_method_xavier_uniform ...................... False
232
+ init_model_with_meta_device ..................... False
233
+ initial_loss_scale .............................. 4294967296
234
+ inprocess_active_world_size ..................... 8
235
+ inprocess_barrier_timeout ....................... 120
236
+ inprocess_completion_timeout .................... 120
237
+ inprocess_empty_cuda_cache ...................... False
238
+ inprocess_granularity ........................... node
239
+ inprocess_hard_timeout .......................... 90
240
+ inprocess_heartbeat_interval .................... 30
241
+ inprocess_heartbeat_timeout ..................... 60
242
+ inprocess_last_call_wait ........................ 1
243
+ inprocess_max_iterations ........................ None
244
+ inprocess_monitor_process_interval .............. 1.0
245
+ inprocess_monitor_thread_interval ............... 1.0
246
+ inprocess_progress_watchdog_interval ............ 1.0
247
+ inprocess_restart ............................... False
248
+ inprocess_soft_timeout .......................... 60
249
+ inprocess_termination_grace_time ................ 1
250
+ is_hybrid_model ................................. False
251
+ iter_per_epoch .................................. 1250
252
+ iterations_to_skip .............................. []
253
+ keep_fp8_transpose_cache_when_using_custom_fsdp . False
254
+ kitchen_config_file ............................. None
255
+ kitchen_recipe_number ........................... None
256
+ kv_channels ..................................... 128
257
+ kv_lora_rank .................................... 32
258
+ lazy_mpu_init ................................... None
259
+ load ............................................ None
260
+ load_main_params_from_ckpt ...................... None
261
+ load_model_opt_format ........................... False
262
+ local_rank ...................................... 0
263
+ log_energy ...................................... False
264
+ log_interval .................................... 1
265
+ log_loss_scale_to_tensorboard ................... True
266
+ log_memory_to_tensorboard ....................... False
267
+ log_num_zeros_in_grad ........................... False
268
+ log_params_norm ................................. False
269
+ log_progress .................................... False
270
+ log_straggler ................................... False
271
+ log_throughput .................................. True
272
+ log_timers_to_tensorboard ....................... False
273
+ log_validation_ppl_to_tensorboard ............... False
274
+ log_world_size_to_tensorboard ................... False
275
+ logging_level ................................... None
276
+ loss_scale ...................................... None
277
+ loss_scale_window ............................... 1000
278
+ lr .............................................. 0.0001
279
+ lr_decay_iters .................................. 8
280
+ lr_decay_samples ................................ None
281
+ lr_decay_style .................................. cosine
282
+ lr_warmup_fraction .............................. None
283
+ lr_warmup_init .................................. 0.0
284
+ lr_warmup_iters ................................. 5
285
+ lr_warmup_samples ............................... 0
286
+ lr_wsd_decay_iters .............................. None
287
+ lr_wsd_decay_samples ............................ None
288
+ lr_wsd_decay_style .............................. exponential
289
+ main_grads_dtype ................................ torch.float32
290
+ main_params_dtype ............................... torch.float32
291
+ make_vocab_size_divisible_by .................... 128
292
+ mamba_head_dim .................................. 64
293
+ mamba_num_groups ................................ 8
294
+ mamba_num_heads ................................. None
295
+ mamba_state_dim ................................. 128
296
+ manual_gc ....................................... False
297
+ manual_gc_eval .................................. True
298
+ manual_gc_interval .............................. 0
299
+ mask_factor ..................................... 1.0
300
+ mask_prob ....................................... 0.15
301
+ mask_type ....................................... random
302
+ masked_softmax_fusion ........................... False
303
+ max_position_embeddings ......................... 32768
304
+ max_tokens_to_oom ............................... 12000
305
+ memory_snapshot_path ............................ snapshot.pickle
306
+ merge_file ...................................... /workspace/data/wikitext/gpt2-merges.txt
307
+ micro_batch_size ................................ 1
308
+ microbatch_group_size_per_vp_stage .............. None
309
+ mid_level_dataset_surplus ....................... 0.005
310
+ min_loss_scale .................................. 1.0
311
+ min_lr .......................................... 1e-05
312
+ mlp_chunks_for_prefill .......................... 1
313
+ mmap_bin_files .................................. True
314
+ mock_data ....................................... False
315
+ moe_apply_probs_on_input ........................ False
316
+ moe_aux_loss_coeff .............................. 0.01
317
+ moe_capacity_monitor ............................ True
318
+ moe_deepep_num_sms .............................. 20
319
+ moe_echo_enable_random_offloading ............... False
320
+ moe_echo_expert_dispatch_overlap ................ False
321
+ moe_echo_expert_dispatcher_type ................. hybridep
322
+ moe_echo_recompute_expert_dispatch .............. False
323
+ moe_enable_deepep ............................... False
324
+ moe_enable_echo ................................. False
325
+ moe_enable_hybridep ............................. True
326
+ moe_enable_ultraep .............................. True
327
+ moe_expert_capacity_factor ...................... None
328
+ moe_extended_tp ................................. False
329
+ moe_ffn_hidden_size ............................. 1536
330
+ moe_grouped_gemm ................................ True
331
+ moe_hybridep_num_sms ............................ 16
332
+ moe_input_jitter_eps ............................ None
333
+ moe_layer_freq .................................. 1
334
+ moe_layer_recompute ............................. False
335
+ moe_num_echo_experts ............................ None
336
+ moe_num_redundant_experts_per_rank .............. 1
337
+ moe_pad_expert_input_to_capacity ................ False
338
+ moe_per_layer_logging ........................... False
339
+ moe_permute_fusion .............................. False
340
+ moe_received_token_capacity ..................... 2.40625
341
+ moe_router_bias_update_rate ..................... 0.001
342
+ moe_router_dtype ................................ fp32
343
+ moe_router_enable_expert_bias ................... False
344
+ moe_router_force_load_balancing ................. False
345
+ moe_router_group_topk ........................... None
346
+ moe_router_load_balancing_type .................. aux_loss
347
+ moe_router_num_groups ........................... None
348
+ moe_router_padding_for_fp8 ...................... False
349
+ moe_router_pre_softmax .......................... False
350
+ moe_router_score_function ....................... softmax
351
+ moe_router_topk ................................. 2
352
+ moe_router_topk_scaling_factor .................. None
353
+ moe_shared_expert_intermediate_size ............. None
354
+ moe_shared_expert_overlap ....................... False
355
+ moe_token_dispatcher_type ....................... flex
356
+ moe_token_drop_policy ........................... probs
357
+ moe_ultraep_max_microbatches .................... None
358
+ moe_ultraep_overlap_weight_sync ................. True
359
+ moe_ultraep_weight_sync_join .................... before_dispatch
360
+ moe_upcycling_granularity ....................... 1
361
+ moe_use_device_initiated_grouped_gemm ........... True
362
+ moe_use_legacy_grouped_gemm ..................... False
363
+ moe_use_upcycling ............................... False
364
+ moe_z_loss_coeff ................................ None
365
+ mrope_section ................................... None
366
+ mscale .......................................... 1.0
367
+ mscale_all_dim .................................. 0.0
368
+ mtp_loss_scaling_factor ......................... 0.1
369
+ mtp_num_layers .................................. None
370
+ multi_latent_attention .......................... False
371
+ nccl_all_reduce_for_prefill ..................... False
372
+ nccl_communicator_config_path ................... None
373
+ nccl_ub ......................................... False
374
+ no_load_optim ................................... None
375
+ no_load_rng ..................................... None
376
+ no_persist_layer_norm ........................... False
377
+ no_rope_freq .................................... None
378
+ no_save_optim ................................... None
379
+ no_save_rng ..................................... None
380
+ non_persistent_ckpt_type ........................ None
381
+ non_persistent_global_ckpt_dir .................. None
382
+ non_persistent_local_ckpt_algo .................. fully_parallel
383
+ non_persistent_local_ckpt_dir ................... None
384
+ non_persistent_save_interval .................... None
385
+ norm_epsilon .................................... 1e-05
386
+ normalization ................................... RMSNorm
387
+ num_attention_heads ............................. 32
388
+ num_channels .................................... 3
389
+ num_classes ..................................... 1000
390
+ num_dataset_builder_threads ..................... 1
391
+ num_distributed_optimizer_instances ............. 1
392
+ num_experts ..................................... 8
393
+ num_layers ...................................... 16
394
+ num_layers_at_end_in_bf16 ....................... 1
395
+ num_layers_at_start_in_bf16 ..................... 1
396
+ num_layers_per_virtual_pipeline_stage ........... None
397
+ num_query_groups ................................ 8
398
+ num_virtual_stages_per_pipeline_rank ............ None
399
+ num_workers ..................................... 2
400
+ object_storage_cache_path ....................... None
401
+ one_logger_async ................................ False
402
+ one_logger_project .............................. megatron-lm
403
+ one_logger_run_name ............................. None
404
+ onnx_safe ....................................... None
405
+ openai_gelu ..................................... False
406
+ optimizer ....................................... adam
407
+ optimizer_cpu_offload ........................... False
408
+ optimizer_offload_fraction ...................... 1.0
409
+ output_bert_embeddings .......................... False
410
+ overlap_cpu_optimizer_d2h_h2d ................... False
411
+ overlap_grad_reduce ............................. True
412
+ overlap_moe_expert_parallel_comm ................ False
413
+ overlap_p2p_comm ................................ False
414
+ overlap_p2p_comm_warmup_flush ................... False
415
+ overlap_param_gather ............................ True
416
+ overlap_param_gather_with_optimizer_step ........ False
417
+ override_opt_param_scheduler .................... False
418
+ params_dtype .................................... torch.bfloat16
419
+ patch_dim ....................................... 16
420
+ per_split_data_args_path ........................ None
421
+ perform_initialization .......................... True
422
+ pin_cpu_grads ................................... True
423
+ pin_cpu_params .................................. True
424
+ pipeline_model_parallel_comm_backend ............ None
425
+ pipeline_model_parallel_layout .................. None
426
+ pipeline_model_parallel_size .................... 1
427
+ position_embedding_type ......................... rope
428
+ pretrained_checkpoint ........................... None
429
+ profile ......................................... False
430
+ profile_ranks ................................... [0]
431
+ profile_step_end ................................ 12
432
+ profile_step_start .............................. 10
433
+ q_lora_rank ..................................... None
434
+ qk_head_dim ..................................... 128
435
+ qk_l2_norm ...................................... False
436
+ qk_layernorm .................................... False
437
+ qk_pos_emb_head_dim ............................. 64
438
+ query_in_block_prob ............................. 0.1
439
+ rampup_batch_size ............................... None
440
+ rank ............................................ 0
441
+ recompute_granularity ........................... None
442
+ recompute_method ................................ None
443
+ recompute_modules ............................... None
444
+ recompute_num_layers ............................ None
445
+ record_memory_history ........................... False
446
+ relative_attention_max_distance ................. 128
447
+ relative_attention_num_buckets .................. 32
448
+ replication ..................................... False
449
+ replication_factor .............................. 2
450
+ replication_jump ................................ None
451
+ rerun_mode ...................................... validate_results
452
+ reset_attention_mask ............................ False
453
+ reset_position_ids .............................. False
454
+ result_rejected_tracker_filename ................ None
455
+ retriever_report_topk_accuracies ................ []
456
+ retriever_score_scaling ......................... False
457
+ retriever_seq_length ............................ 256
458
+ retro_add_retriever ............................. False
459
+ retro_attention_gate ............................ 1
460
+ retro_cyclic_train_iters ........................ None
461
+ retro_encoder_attention_dropout ................. 0.1
462
+ retro_encoder_hidden_dropout .................... 0.1
463
+ retro_encoder_layers ............................ 2
464
+ retro_num_neighbors ............................. 2
465
+ retro_num_retrieved_chunks ...................... 2
466
+ retro_project_dir ............................... None
467
+ retro_verify_neighbor_count ..................... True
468
+ reuse_grad_buf_for_mxfp8_param_ag ............... False
469
+ rope_scaling_factor ............................. 8.0
470
+ rope_type ....................................... None
471
+ rotary_base ..................................... 1000000
472
+ rotary_interleaved .............................. False
473
+ rotary_percent .................................. 1.0
474
+ rotary_scaling_factor ........................... 1.0
475
+ rotary_seq_len_interpolation_factor ............. None
476
+ run_workload_inspector_server ................... False
477
+ sample_rate ..................................... 1.0
478
+ save ............................................ None
479
+ save_interval ................................... None
480
+ save_retain_interval ............................ None
481
+ scatter_gather_tensors_in_pipeline .............. True
482
+ seed ............................................ 1234
483
+ seq_length ...................................... 4096
484
+ sequence_parallel ............................... False
485
+ sft ............................................. False
486
+ sft_tokenizer_prompt_format ..................... nemotron-h-aligned
487
+ sgd_momentum .................................... 0.9
488
+ short_seq_prob .................................. 0.1
489
+ skip_train ...................................... False
490
+ skipped_train_samples ........................... 0
491
+ spec ............................................ None
492
+ split ........................................... 100,0,0
493
+ squared_relu .................................... False
494
+ start_weight_decay .............................. 0.1
495
+ straggler_ctrlr_port ............................ 65535
496
+ straggler_minmax_count .......................... 1
497
+ suggested_communication_unit_size ............... None
498
+ swiglu .......................................... True
499
+ swin_backbone_type .............................. tiny
500
+ symmetric_ar_type ............................... None
501
+ te_rng_tracker .................................. True
502
+ tensor_model_parallel_size ...................... 1
503
+ tensorboard_dir ................................. None
504
+ tensorboard_log_interval ........................ 1
505
+ tensorboard_queue_size .......................... 1000
506
+ test_data_path .................................. None
507
+ test_mode ....................................... False
508
+ tiktoken_num_special_tokens ..................... 1000
509
+ tiktoken_pattern ................................ None
510
+ tiktoken_special_tokens ......................... None
511
+ timing_log_level ................................ 0
512
+ timing_log_option ............................... minmax
513
+ titles_data_path ................................ None
514
+ tokenizer_model ................................. None
515
+ tokenizer_type .................................. GPT2BPETokenizer
516
+ torch_fsdp2_reshard_after_forward ............... True
517
+ tp_comm_bootstrap_backend ....................... nccl
518
+ tp_comm_bulk_dgrad .............................. True
519
+ tp_comm_bulk_wgrad .............................. True
520
+ tp_comm_overlap ................................. False
521
+ tp_comm_overlap_ag .............................. True
522
+ tp_comm_overlap_cfg ............................. None
523
+ tp_comm_overlap_rs .............................. True
524
+ tp_comm_overlap_rs_dgrad ........................ False
525
+ tp_comm_split_ag ................................ True
526
+ tp_comm_split_rs ................................ True
527
+ train_data_path ................................. None
528
+ train_iters ..................................... 12
529
+ train_samples ................................... None
530
+ train_sync_interval ............................. None
531
+ transformer_impl ................................ transformer_engine
532
+ transformer_pipeline_model_parallel_size ........ 1
533
+ untie_embeddings_and_output_weights ............. True
534
+ use_checkpoint_args ............................. False
535
+ use_checkpoint_opt_param_scheduler .............. False
536
+ use_cpu_initialization .......................... None
537
+ use_custom_fsdp ................................. False
538
+ use_dist_ckpt ................................... True
539
+ use_dist_ckpt_deprecated ........................ False
540
+ use_distributed_optimizer ....................... True
541
+ use_flash_attn .................................. False
542
+ use_legacy_models ............................... False
543
+ use_mp_args_from_checkpoint_args ................ False
544
+ use_one_sent_docs ............................... False
545
+ use_persistent_ckpt_worker ...................... False
546
+ use_precision_aware_optimizer ................... False
547
+ use_pytorch_profiler ............................ False
548
+ use_ring_exchange_p2p ........................... False
549
+ use_rope_scaling ................................ False
550
+ use_rotary_position_embeddings .................. False
551
+ use_sharp ....................................... False
552
+ use_tokenizer_model_from_checkpoint_args ........ True
553
+ use_torch_fsdp2 ................................. False
554
+ use_torch_optimizer_for_cpu_offload ............. False
555
+ use_tp_pp_dp_mapping ............................ False
556
+ v_head_dim ...................................... 128
557
+ valid_data_path ................................. None
558
+ variable_seq_lengths ............................ False
559
+ virtual_pipeline_model_parallel_size ............ None
560
+ vision_backbone_type ............................ vit
561
+ vision_pretraining .............................. False
562
+ vision_pretraining_type ......................... classify
563
+ vocab_extra_ids ................................. 0
564
+ vocab_file ...................................... /workspace/data/wikitext/gpt2-vocab.json
565
+ vocab_size ...................................... None
566
+ wandb_exp_name ..................................
567
+ wandb_project ...................................
568
+ wandb_save_dir ..................................
569
+ weight_decay .................................... 0.1
570
+ weight_decay_incr_style ......................... constant
571
+ wgrad_deferral_limit ............................ 0
572
+ world_size ...................................... 8
573
+ yaml_cfg ........................................ None
574
+ -------------------- end of arguments ---------------------
575
+ INFO:megatron.core.num_microbatches_calculator:setting number of microbatches to constant 2
576
+ > building GPT2BPETokenizer tokenizer ...
577
+ > padded vocab (size: 50257) with 47 dummy tokens (new size: 50304)
578
+ WARNING:megatron.core.rerun_state_machine:RerunStateMachine initialized in mode RerunMode.VALIDATE_RESULTS
579
+ > initializing torch distributed ...
580
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
581
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
582
+ [Gloo] Rank 7 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
583
+ [Gloo] Rank 0 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
584
+ [Gloo] Rank 1 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
585
+ [Gloo] Rank 4 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
586
+ [Gloo] Rank 2 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
587
+ [Gloo] Rank 3 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
588
+ [Gloo] Rank [Gloo] Rank 5 is connected to 67 is connected to peer ranks. 7Expected number of connected peer ranks is : peer ranks. 7Expected number of connected peer ranks is :
589
+ 7
590
+ [Gloo] Rank 1 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
591
+ [Gloo] Rank 0 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
592
+ [Gloo] Rank 3 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
593
+ [Gloo] Rank 2 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
594
+ [Gloo] Rank 4 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
595
+ [Gloo] Rank 6 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
596
+ [Gloo] Rank 5 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
597
+ [Gloo] Rank 7 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
598
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
599
+ > initialized tensor model parallel with size 1
600
+ > initialized pipeline model parallel with size 1
601
+ > setting random seeds to 1234 ...
602
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
603
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
604
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
605
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
606
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
607
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
608
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
609
+ [rank5]:[W830 07:18:02.550985330 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 5] using GPU 5 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
610
+ [rank7]:[W830 07:18:02.551640626 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 7] using GPU 7 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
611
+ [rank3]:[W830 07:18:02.557784250 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 3] using GPU 3 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
612
+ [rank2]:[W830 07:18:02.557819324 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 2] using GPU 2 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
613
+ [rank1]:[W830 07:18:02.557824167 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 1] using GPU 1 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
614
+ [rank4]:[W830 07:18:02.557828637 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 4] using GPU 4 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
615
+ [rank6]:[W830 07:18:02.557833838 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 6] using GPU 6 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
616
+ > compiling dataset index builder ...
617
+ make: Entering directory '/workspace/Megatron-UltraEP/megatron/core/datasets'
618
+ make: Nothing to be done for 'default'.
619
+ make: Leaving directory '/workspace/Megatron-UltraEP/megatron/core/datasets'
620
+ >>> done with dataset index builder. Compilation time: 0.035 seconds
621
+ WARNING: constraints for invoking optimized fused softmax kernel are not met. We default back to unfused kernel invocations.
622
+ > compiling and loading fused kernels ...
623
+ [rank0]:[W830 07:18:02.720898652 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 0] using GPU 0 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
624
+ >>> done with compiling and loading fused kernels. Compilation time: 3.589 seconds
625
+ time to initialize megatron (seconds): 6.869
626
+ [after megatron is initialized] datetime: 2026-08-30 07:18:08
627
+ building GPT model ...
628
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
629
+ warnings.warn(
630
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
631
+ warnings.warn(
632
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
633
+ warnings.warn(
634
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
635
+ warnings.warn(
636
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
637
+ warnings.warn(
638
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
639
+ warnings.warn(
640
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
641
+ warnings.warn(
642
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
643
+ warnings.warn(
644
+ [INFO] Use MNNVL fabric: False
645
+ [WARN] MAX_NUM_NVL_PEERS is not set. Using detected value 8.
646
+ NVSHMEM v3.7.2
647
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
648
+
649
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
650
+
651
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
652
+
653
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
654
+
655
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
656
+
657
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
658
+
659
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
660
+
661
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
662
+
663
+ UltraEP: NVL domain size <= 8; forcing weight_sync plan mode to direct (was adaptive).
664
+ > number of parameters on (tensor, pipeline) model parallel rank (0, 0): 1687818240
665
+ INFO:megatron.core.distributed.distributed_data_parallel:Setting up DistributedDataParallel with config DistributedDataParallelConfig(grad_reduce_in_fp32=True, overlap_grad_reduce=True, overlap_param_gather=True, align_param_gather=False, use_distributed_optimizer=True, num_distributed_optimizer_instances=1, check_for_nan_in_grad=False, check_for_large_grads=False, bucket_size=40000000, pad_buckets_for_high_nccl_busbw=False, average_in_collective=False, fp8_param_gather=False, reuse_grad_buf_for_mxfp8_param_ag=False, use_custom_fsdp=False, data_parallel_sharding_strategy='no_shard', gradient_reduce_div_fusion=True, suggested_communication_unit_size=None, preserve_fp32_weights=True, keep_fp8_transpose_cache_when_using_custom_fsdp=False, nccl_ub=False, fsdp_double_buffer=False)
666
+ INFO:megatron.core.distributed.param_and_grad_buffer:Number of buckets for gradient all-reduce / reduce-scatter: 18
667
+ Params for bucket 1 (206045184 elements, 206045184 padded size):
668
+ module.output_layer.weight
669
+ Params for bucket 2 (41988096 elements, 41988096 padded size):
670
+ module.decoder.final_layernorm.weight
671
+ module.decoder.layers.15.mlp.router.weight
672
+ module.decoder.layers.15.self_attention.linear_qkv.layer_norm_weight
673
+ module.decoder.layers.15.self_attention.linear_proj.weight
674
+ module.decoder.layers.15.self_attention.linear_qkv.weight
675
+ module.decoder.layers.15.pre_mlp_layernorm.weight
676
+ Params for bucket 3 (41984000 elements, 41984000 padded size):
677
+ module.decoder.layers.14.mlp.router.weight
678
+ module.decoder.layers.14.self_attention.linear_proj.weight
679
+ module.decoder.layers.14.self_attention.linear_qkv.layer_norm_weight
680
+ module.decoder.layers.14.pre_mlp_layernorm.weight
681
+ module.decoder.layers.14.self_attention.linear_qkv.weight
682
+ Params for bucket 4 (41984000 elements, 41984000 padded size):
683
+ module.decoder.layers.13.mlp.router.weight
684
+ module.decoder.layers.13.self_attention.linear_qkv.layer_norm_weight
685
+ module.decoder.layers.13.self_attention.linear_proj.weight
686
+ module.decoder.layers.13.self_attention.linear_qkv.weight
687
+ module.decoder.layers.13.pre_mlp_layernorm.weight
688
+ Params for bucket 5 (41984000 elements, 41984000 padded size):
689
+ module.decoder.layers.12.mlp.router.weight
690
+ module.decoder.layers.12.self_attention.linear_proj.weight
691
+ module.decoder.layers.12.self_attention.linear_qkv.layer_norm_weight
692
+ module.decoder.layers.12.self_attention.linear_qkv.weight
693
+ module.decoder.layers.12.pre_mlp_layernorm.weight
694
+ Params for bucket 6 (41984000 elements, 41984000 padded size):
695
+ module.decoder.layers.11.mlp.router.weight
696
+ module.decoder.layers.11.self_attention.linear_proj.weight
697
+ module.decoder.layers.11.self_attention.linear_qkv.layer_norm_weight
698
+ module.decoder.layers.11.pre_mlp_layernorm.weight
699
+ module.decoder.layers.11.self_attention.linear_qkv.weight
700
+ Params for bucket 7 (41984000 elements, 41984000 padded size):
701
+ module.decoder.layers.10.mlp.router.weight
702
+ module.decoder.layers.10.self_attention.linear_proj.weight
703
+ module.decoder.layers.10.self_attention.linear_qkv.layer_norm_weight
704
+ module.decoder.layers.10.self_attention.linear_qkv.weight
705
+ module.decoder.layers.10.pre_mlp_layernorm.weight
706
+ Params for bucket 8 (41984000 elements, 41984000 padded size):
707
+ module.decoder.layers.9.mlp.router.weight
708
+ module.decoder.layers.9.self_attention.linear_proj.weight
709
+ module.decoder.layers.9.self_attention.linear_qkv.layer_norm_weight
710
+ module.decoder.layers.9.self_attention.linear_qkv.weight
711
+ module.decoder.layers.9.pre_mlp_layernorm.weight
712
+ Params for bucket 9 (41984000 elements, 41984000 padded size):
713
+ module.decoder.layers.8.mlp.router.weight
714
+ module.decoder.layers.8.self_attention.linear_proj.weight
715
+ module.decoder.layers.8.self_attention.linear_qkv.layer_norm_weight
716
+ module.decoder.layers.8.pre_mlp_layernorm.weight
717
+ module.decoder.layers.8.self_attention.linear_qkv.weight
718
+ Params for bucket 10 (41984000 elements, 41984000 padded size):
719
+ module.decoder.layers.7.mlp.router.weight
720
+ module.decoder.layers.7.self_attention.linear_proj.weight
721
+ module.decoder.layers.7.self_attention.linear_qkv.layer_norm_weight
722
+ module.decoder.layers.7.self_attention.linear_qkv.weight
723
+ module.decoder.layers.7.pre_mlp_layernorm.weight
724
+ Params for bucket 11 (41984000 elements, 41984000 padded size):
725
+ module.decoder.layers.6.self_attention.linear_qkv.weight
726
+ module.decoder.layers.6.pre_mlp_layernorm.weight
727
+ module.decoder.layers.6.mlp.router.weight
728
+ module.decoder.layers.6.self_attention.linear_proj.weight
729
+ module.decoder.layers.6.self_attention.linear_qkv.layer_norm_weight
730
+ Params for bucket 12 (41984000 elements, 41984000 padded size):
731
+ module.decoder.layers.5.mlp.router.weight
732
+ module.decoder.layers.5.self_attention.linear_qkv.layer_norm_weight
733
+ module.decoder.layers.5.self_attention.linear_qkv.weight
734
+ module.decoder.layers.5.self_attention.linear_proj.weight
735
+ module.decoder.layers.5.pre_mlp_layernorm.weight
736
+ Params for bucket 13 (41984000 elements, 41984000 padded size):
737
+ module.decoder.layers.4.pre_mlp_layernorm.weight
738
+ module.decoder.layers.4.self_attention.linear_qkv.weight
739
+ module.decoder.layers.4.mlp.router.weight
740
+ module.decoder.layers.4.self_attention.linear_proj.weight
741
+ module.decoder.layers.4.self_attention.linear_qkv.layer_norm_weight
742
+ Params for bucket 14 (41984000 elements, 41984000 padded size):
743
+ module.decoder.layers.3.self_attention.linear_qkv.weight
744
+ module.decoder.layers.3.pre_mlp_layernorm.weight
745
+ module.decoder.layers.3.mlp.router.weight
746
+ module.decoder.layers.3.self_attention.linear_proj.weight
747
+ module.decoder.layers.3.self_attention.linear_qkv.layer_norm_weight
748
+ Params for bucket 15 (41984000 elements, 41984000 padded size):
749
+ module.decoder.layers.2.pre_mlp_layernorm.weight
750
+ module.decoder.layers.2.self_attention.linear_qkv.weight
751
+ module.decoder.layers.2.mlp.router.weight
752
+ module.decoder.layers.2.self_attention.linear_proj.weight
753
+ module.decoder.layers.2.self_attention.linear_qkv.layer_norm_weight
754
+ Params for bucket 16 (41984000 elements, 41984000 padded size):
755
+ module.decoder.layers.1.mlp.router.weight
756
+ module.decoder.layers.1.pre_mlp_layernorm.weight
757
+ module.decoder.layers.1.self_attention.linear_qkv.layer_norm_weight
758
+ module.decoder.layers.1.self_attention.linear_proj.weight
759
+ module.decoder.layers.1.self_attention.linear_qkv.weight
760
+ Params for bucket 17 (41984000 elements, 41984000 padded size):
761
+ module.decoder.layers.0.pre_mlp_layernorm.weight
762
+ module.decoder.layers.0.self_attention.linear_qkv.weight
763
+ module.decoder.layers.0.self_attention.linear_qkv.layer_norm_weight
764
+ module.decoder.layers.0.mlp.router.weight
765
+ module.decoder.layers.0.self_attention.linear_proj.weight
766
+ Params for bucket 18 (206045184 elements, 206045184 padded size):
767
+ module.embedding.word_embeddings.weight
768
+ INFO:megatron.core.distributed.param_and_grad_buffer:Number of buckets for gradient all-reduce / reduce-scatter: 7
769
+ Params for bucket 1 (44040192 elements, 44040192 padded size):
770
+ module.decoder.layers.14.mlp.experts.linear_fc1.weight0
771
+ module.decoder.layers.15.mlp.experts.linear_fc1.weight0
772
+ module.decoder.layers.15.mlp.experts.linear_fc2.weight0
773
+ module.decoder.layers.13.mlp.experts.linear_fc2.weight0
774
+ module.decoder.layers.14.mlp.experts.linear_fc2.weight0
775
+ Params for bucket 2 (50331648 elements, 50331648 padded size):
776
+ module.decoder.layers.13.mlp.experts.linear_fc1.weight0
777
+ module.decoder.layers.11.mlp.experts.linear_fc1.weight0
778
+ module.decoder.layers.12.mlp.experts.linear_fc1.weight0
779
+ module.decoder.layers.11.mlp.experts.linear_fc2.weight0
780
+ module.decoder.layers.12.mlp.experts.linear_fc2.weight0
781
+ Params for bucket 3 (44040192 elements, 44040192 padded size):
782
+ module.decoder.layers.10.mlp.experts.linear_fc1.weight0
783
+ module.decoder.layers.9.mlp.experts.linear_fc1.weight0
784
+ module.decoder.layers.10.mlp.experts.linear_fc2.weight0
785
+ module.decoder.layers.9.mlp.experts.linear_fc2.weight0
786
+ module.decoder.layers.8.mlp.experts.linear_fc2.weight0
787
+ Params for bucket 4 (50331648 elements, 50331648 padded size):
788
+ module.decoder.layers.8.mlp.experts.linear_fc1.weight0
789
+ module.decoder.layers.6.mlp.experts.linear_fc2.weight0
790
+ module.decoder.layers.7.mlp.experts.linear_fc1.weight0
791
+ module.decoder.layers.7.mlp.experts.linear_fc2.weight0
792
+ module.decoder.layers.6.mlp.experts.linear_fc1.weight0
793
+ Params for bucket 5 (44040192 elements, 44040192 padded size):
794
+ module.decoder.layers.5.mlp.experts.linear_fc2.weight0
795
+ module.decoder.layers.3.mlp.experts.linear_fc2.weight0
796
+ module.decoder.layers.4.mlp.experts.linear_fc2.weight0
797
+ module.decoder.layers.4.mlp.experts.linear_fc1.weight0
798
+ module.decoder.layers.5.mlp.experts.linear_fc1.weight0
799
+ Params for bucket 6 (50331648 elements, 50331648 padded size):
800
+ module.decoder.layers.1.mlp.experts.linear_fc1.weight0
801
+ module.decoder.layers.1.mlp.experts.linear_fc2.weight0
802
+ module.decoder.layers.2.mlp.experts.linear_fc2.weight0
803
+ module.decoder.layers.3.mlp.experts.linear_fc1.weight0
804
+ module.decoder.layers.2.mlp.experts.linear_fc1.weight0
805
+ Params for bucket 7 (18874368 elements, 18874368 padded size):
806
+ module.decoder.layers.0.mlp.experts.linear_fc2.weight0
807
+ module.decoder.layers.0.mlp.experts.linear_fc1.weight0
808
+ INFO:megatron.core.optimizer:Setting up optimizer with config OptimizerConfig(optimizer='adam', lr=0.0001, min_lr=1e-05, decoupled_lr=None, decoupled_min_lr=None, weight_decay=0.1, fp8_recipe='mxfp8', fp16=False, bf16=True, reuse_grad_buf_for_mxfp8_param_ag=False, params_dtype=torch.bfloat16, use_precision_aware_optimizer=False, store_param_remainders=True, main_grads_dtype=torch.float32, main_params_dtype=torch.float32, exp_avg_dtype=torch.float32, exp_avg_sq_dtype=torch.float32, loss_scale=None, initial_loss_scale=4294967296, min_loss_scale=1.0, loss_scale_window=1000, hysteresis=2, adam_beta1=0.9, adam_beta2=0.999, adam_eps=1e-08, sgd_momentum=0.9, use_distributed_optimizer=True, overlap_param_gather_with_optimizer_step=False, optimizer_cpu_offload=False, optimizer_offload_fraction=1.0, use_torch_optimizer_for_cpu_offload=False, overlap_cpu_optimizer_d2h_h2d=False, pin_cpu_grads=True, pin_cpu_params=True, clip_grad=1.0, log_num_zeros_in_grad=False, barrier_with_L1_time=True, timers=<megatron.core.timers.Timers object at 0x767f7ac592b0>, config_logger_dir='')
809
+ INFO:megatron.core.optimizer_param_scheduler:> learning rate decay style: cosine
810
+ [after model, optimizer, and learning rate scheduler are built] datetime: 2026-08-30 07:18:11
811
+ > building train, validation, and test datasets ...
812
+ > datasets target sizes (minimum size):
813
+ train: 192
814
+ validation: 0
815
+ test: 0
816
+ INFO:megatron.core.datasets.blended_megatron_dataset_config:Let split_matrix = [(0, 1.0), None, None]
817
+ > building train, validation, and test datasets for GPT ...
818
+ INFO:megatron.core.datasets.blended_megatron_dataset_builder:Building GPTDataset splits with sizes=(192, 0, 0) and config=GPTDatasetConfig(random_seed=1234, sequence_length=4096, blend=(['/workspace/data/wikitext/wikitext103_gpt2_text_document'], None), blend_per_split=None, split='100,0,0', split_matrix=[(0, 1.0), None, None], num_dataset_builder_threads=1, path_to_cache=None, mmap_bin_files=True, mock=False, tokenizer=<megatron.training.tokenizer.tokenizer._GPT2BPETokenizer object at 0x767f7aecab40>, mid_level_dataset_surplus=0.005, reset_position_ids=False, reset_attention_mask=False, eod_mask_loss=False, create_attention_mask=True, drop_last_partial_validation_sequence=True, add_extra_token_to_sequence=True, object_storage_cache_path=None)
819
+ INFO:megatron.core.datasets.indexed_dataset:Load the _IndexReader from /workspace/data/wikitext/wikitext103_gpt2_text_document.idx
820
+ INFO:megatron.core.datasets.indexed_dataset: Extract the sequence lengths
821
+ INFO:megatron.core.datasets.indexed_dataset: Extract the sequence pointers
822
+ INFO:megatron.core.datasets.indexed_dataset: Extract the document indices
823
+ INFO:megatron.core.datasets.indexed_dataset:> total number of sequences: 1165029
824
+ INFO:megatron.core.datasets.indexed_dataset:> total number of documents: 1801350
825
+ INFO:megatron.core.datasets.gpt_dataset:Load the GPTDataset train indices
826
+ INFO:megatron.core.datasets.gpt_dataset: Load the document index from 44ed1ee5bcb9e5b030387b8d27a6c406-GPTDataset-train-document_index.npy
827
+ INFO:megatron.core.datasets.gpt_dataset: Load the sample index from 44ed1ee5bcb9e5b030387b8d27a6c406-GPTDataset-train-sample_index.npy
828
+ INFO:megatron.core.datasets.gpt_dataset: Load the shuffle index from 44ed1ee5bcb9e5b030387b8d27a6c406-GPTDataset-train-shuffle_index.npy
829
+ INFO:megatron.core.datasets.gpt_dataset:> total number of samples: 29073
830
+ > finished creating GPT datasets ...
831
+ [after dataloaders are built] datetime: 2026-08-30 07:18:11
832
+ done with setup ...
833
+ (min, max) time across ranks (ms):
834
+ model-and-optimizer-setup ......................: (3118.92, 3119.00)
835
+ train/valid/test-data-iterators-setup ..........: (72.07, 92.15)
836
+ training ...
837
+ Overwriting rerun_state_machine.current_iteration from -1 to 0...
838
+ [before the start of training step] datetime: 2026-08-30 07:18:11
839
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
840
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
841
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
842
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
843
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
844
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
845
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
846
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
847
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
848
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
849
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
850
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
851
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
852
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
853
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
854
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
855
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
856
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
857
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
858
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
859
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
860
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
861
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
862
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
863
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
864
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
865
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
866
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
867
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
868
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
869
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
870
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
871
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
872
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
873
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
874
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
875
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
876
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
877
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
878
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
879
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
880
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
881
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
882
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
883
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
884
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
885
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
886
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
887
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
888
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
889
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
890
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
891
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
892
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
893
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
894
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
895
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
896
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
897
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
898
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
899
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
900
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
901
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
902
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
903
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
904
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
905
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
906
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
907
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
908
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
909
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
910
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
911
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
912
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
913
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
914
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
915
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
916
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
917
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
918
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
919
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
920
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
921
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
922
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
923
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
924
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
925
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
926
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
927
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
928
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
929
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
930
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
931
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
932
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
933
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
934
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
935
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
936
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
937
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
938
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
939
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
940
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
941
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
942
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
943
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
944
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
945
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
946
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
947
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
948
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
949
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
950
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
951
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
952
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
953
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
954
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
955
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
956
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
957
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
958
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
959
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
960
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
961
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
962
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
963
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
964
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
965
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
966
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
967
+ [2026-08-30 07:18:37] iteration 1/ 12 | consumed samples: 16 | elapsed time per iteration (ms): 25674.7 | throughput per GPU (TFLOP/s/GPU): 3.3 | learning rate: 2.000000E-05 | global batch size: 16 | lm loss: 1.103286E+01 | loss scale: 1.0 | grad norm: 44.353 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14592 | min safe capacity: 1.7812 |
968
+ Number of parameters in transformer block in billions: 3.09
969
+ Number of parameters in embedding layers in billions: 0.41
970
+ Total number of parameters in billions: 3.50
971
+ Number of parameters in most loaded shard in billions: 3.4994
972
+ Theoretical memory footprints: weight and optimizer=25029.43 MB
973
+ [Rank 0] (after 1 iterations) memory (MB) | allocated: 14864.61376953125 | max allocated: 22156.96533203125 | reserved: 22904.0 | max reserved: 22904.0
974
+ [2026-08-30 07:18:37] iteration 2/ 12 | consumed samples: 32 | elapsed time per iteration (ms): 272.5 | throughput per GPU (TFLOP/s/GPU): 315.5 | learning rate: 4.000000E-05 | global batch size: 16 | lm loss: 1.102278E+01 | loss scale: 1.0 | grad norm: 46.283 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14848 | min safe capacity: 1.8125 |
975
+ [2026-08-30 07:18:37] iteration 3/ 12 | consumed samples: 48 | elapsed time per iteration (ms): 255.3 | throughput per GPU (TFLOP/s/GPU): 336.8 | learning rate: 6.000000E-05 | global batch size: 16 | lm loss: 1.037333E+01 | loss scale: 1.0 | grad norm: 215.171 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14848 | min safe capacity: 1.8125 |
976
+ Capture CUDA graph for training!!!
977
+ CUDA graph capture done!!!
978
+ [2026-08-30 07:18:38] iteration 4/ 12 | consumed samples: 64 | elapsed time per iteration (ms): 993.9 | throughput per GPU (TFLOP/s/GPU): 86.5 | learning rate: 8.000000E-05 | global batch size: 16 | lm loss: 1.313653E+01 | loss scale: 1.0 | grad norm: 41.158 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14848 | min safe capacity: 1.8125 |
979
+ [2026-08-30 07:18:38] iteration 5/ 12 | consumed samples: 80 | elapsed time per iteration (ms): 177.1 | throughput per GPU (TFLOP/s/GPU): 485.6 | learning rate: 1.000000E-04 | global batch size: 16 | lm loss: 1.037862E+01 | loss scale: 1.0 | grad norm: 15.962 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14848 | min safe capacity: 1.8125 |
980
+ [2026-08-30 07:18:38] iteration 6/ 12 | consumed samples: 96 | elapsed time per iteration (ms): 176.3 | throughput per GPU (TFLOP/s/GPU): 487.9 | learning rate: 7.750000E-05 | global batch size: 16 | lm loss: 1.002902E+01 | loss scale: 1.0 | grad norm: 9.280 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14848 | min safe capacity: 1.8125 |
981
+ [2026-08-30 07:18:39] iteration 7/ 12 | consumed samples: 112 | elapsed time per iteration (ms): 177.0 | throughput per GPU (TFLOP/s/GPU): 485.9 | learning rate: 3.250000E-05 | global batch size: 16 | lm loss: 9.371662E+00 | loss scale: 1.0 | grad norm: 14.828 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14848 | min safe capacity: 1.8125 |
982
+ [2026-08-30 07:18:39] iteration 8/ 12 | consumed samples: 128 | elapsed time per iteration (ms): 176.5 | throughput per GPU (TFLOP/s/GPU): 487.1 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.945050E+00 | loss scale: 1.0 | grad norm: 6.170 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14848 | min safe capacity: 1.8125 |
983
+ [2026-08-30 07:18:39] iteration 9/ 12 | consumed samples: 144 | elapsed time per iteration (ms): 176.8 | throughput per GPU (TFLOP/s/GPU): 486.3 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.780246E+00 | loss scale: 1.0 | grad norm: 5.198 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14848 | min safe capacity: 1.8125 |
984
+ [2026-08-30 07:18:39] iteration 10/ 12 | consumed samples: 160 | elapsed time per iteration (ms): 175.9 | throughput per GPU (TFLOP/s/GPU): 488.9 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.695450E+00 | loss scale: 1.0 | grad norm: 5.511 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14848 | min safe capacity: 1.8125 |
985
+ [2026-08-30 07:18:39] iteration 11/ 12 | consumed samples: 176 | elapsed time per iteration (ms): 176.9 | throughput per GPU (TFLOP/s/GPU): 486.2 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.617610E+00 | loss scale: 1.0 | grad norm: 5.072 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14848 | min safe capacity: 1.8125 |
986
+ [2026-08-30 07:18:39] iteration 12/ 12 | consumed samples: 192 | elapsed time per iteration (ms): 177.3 | throughput per GPU (TFLOP/s/GPU): 485.0 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.518322E+00 | loss scale: 1.0 | grad norm: 4.735 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 14848 | min safe capacity: 1.8125 |
987
+ [after training is done] datetime: 2026-08-30 07:18:39
988
+ [rank0]:[W830 07:18:41.253369394 ProcessGroupNCCL.cpp:1505] Warning: WARNING: destroy_process_group() was not called before program exit, which can leak resources. For more info, please see https://pytorch.org/docs/stable/distributed.html#shutdown (function operator())
989
+ [rank3]:[W830 07:18:42.523363398 ProcessGroupNCCL.cpp:1505] Warning: WARNING: destroy_process_group() was not called before program exit, which can leak resources. For more info, please see https://pytorch.org/docs/stable/distributed.html#shutdown (function operator())
990
+ [rank5]:[W830 07:18:42.525268761 ProcessGroupNCCL.cpp:1505] Warning: WARNING: destroy_process_group() was not called before program exit, which can leak resources. For more info, please see https://pytorch.org/docs/stable/distributed.html#shutdown (function operator())
991
+ [rank6]:[W830 07:18:42.528742476 ProcessGroupNCCL.cpp:1505] Warning: WARNING: destroy_process_group() was not called before program exit, which can leak resources. For more info, please see https://pytorch.org/docs/stable/distributed.html#shutdown (function operator())
992
+ [rank2]:[W830 07:18:42.540765075 ProcessGroupNCCL.cpp:1505] Warning: WARNING: destroy_process_group() was not called before program exit, which can leak resources. For more info, please see https://pytorch.org/docs/stable/distributed.html#shutdown (function operator())
993
+ [rank4]:[W830 07:18:42.540769797 ProcessGroupNCCL.cpp:1505] Warning: WARNING: destroy_process_group() was not called before program exit, which can leak resources. For more info, please see https://pytorch.org/docs/stable/distributed.html#shutdown (function operator())
994
+ [rank1]:[W830 07:18:42.071117884 ProcessGroupNCCL.cpp:1505] Warning: WARNING: destroy_process_group() was not called before program exit, which can leak resources. For more info, please see https://pytorch.org/docs/stable/distributed.html#shutdown (function operator())
995
+ [rank7]:[W830 07:18:42.217894483 ProcessGroupNCCL.cpp:1505] Warning: WARNING: destroy_process_group() was not called before program exit, which can leak resources. For more info, please see https://pytorch.org/docs/stable/distributed.html#shutdown (function operator())
996
+ W0830 07:18:43.097000 736138 torch/distributed/elastic/agent/server/api.py:719] Received 15 death signal, shutting down workers
997
+ W0830 07:18:43.098000 736138 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 736229 closing signal SIGTERM
998
+ W0830 07:18:43.098000 736138 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 736230 closing signal SIGTERM
999
+ W0830 07:18:43.100000 736138 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 736231 closing signal SIGTERM
1000
+ W0830 07:18:43.102000 736138 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 736232 closing signal SIGTERM
1001
+ W0830 07:18:43.102000 736138 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 736233 closing signal SIGTERM
1002
+ W0830 07:18:43.103000 736138 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 736234 closing signal SIGTERM
1003
+ W0830 07:18:43.104000 736138 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 736235 closing signal SIGTERM
1004
+ W0830 07:18:43.105000 736138 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 736236 closing signal SIGTERM
1005
+ Traceback (most recent call last):
1006
+ File "/usr/local/bin/torchrun", line 8, in <module>
1007
+ sys.exit(main())
1008
+ ^^^^^^
1009
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/multiprocessing/errors/__init__.py", line 355, in wrapper
1010
+ return f(*args, **kwargs)
1011
+ ^^^^^^^^^^^^^^^^^^
1012
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/run.py", line 892, in main
1013
+ run(args)
1014
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/run.py", line 883, in run
1015
+ elastic_launch(
1016
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/launcher/api.py", line 139, in __call__
1017
+ return launch_agent(self._config, self._entrypoint, list(args))
1018
+ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1019
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/launcher/api.py", line 261, in launch_agent
1020
+ result = agent.run()
1021
+ ^^^^^^^^^^^
1022
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/metrics/api.py", line 138, in wrapper
1023
+ result = f(*args, **kwargs)
1024
+ ^^^^^^^^^^^^^^^^^^
1025
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/agent/server/api.py", line 711, in run
1026
+ result = self._invoke_run(role)
1027
+ ^^^^^^^^^^^^^^^^^^^^^^
1028
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/agent/server/api.py", line 870, in _invoke_run
1029
+ time.sleep(monitor_interval)
1030
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/multiprocessing/api.py", line 84, in _terminate_process_handler
1031
+ raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval)
1032
+ torch.distributed.elastic.multiprocessing.api.SignalException: Process 736138 got signal: 15
tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq8192__gbs16__cap2p40625/rep1/mem_samples.csv ADDED
@@ -0,0 +1,192 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 0, 0
2
+ 1, 0
3
+ 2, 0
4
+ 3, 0
5
+ 4, 0
6
+ 5, 0
7
+ 6, 0
8
+ 7, 0
9
+ 0, 0
10
+ 1, 0
11
+ 2, 0
12
+ 3, 0
13
+ 4, 0
14
+ 5, 0
15
+ 6, 0
16
+ 7, 0
17
+ 0, 1
18
+ 1, 1
19
+ 2, 1
20
+ 3, 1
21
+ 4, 1
22
+ 5, 1
23
+ 6, 1
24
+ 7, 1
25
+ 0, 4
26
+ 1, 4
27
+ 2, 4
28
+ 3, 4
29
+ 4, 1
30
+ 5, 1
31
+ 6, 1
32
+ 7, 1
33
+ 0, 4
34
+ 1, 4
35
+ 2, 4
36
+ 3, 4
37
+ 4, 4
38
+ 5, 4
39
+ 6, 4
40
+ 7, 4
41
+ 0, 4
42
+ 1, 4
43
+ 2, 4
44
+ 3, 629
45
+ 4, 4
46
+ 5, 4
47
+ 6, 4
48
+ 7, 4
49
+ 0, 821
50
+ 1, 821
51
+ 2, 821
52
+ 3, 821
53
+ 4, 821
54
+ 5, 821
55
+ 6, 821
56
+ 7, 821
57
+ 0, 2007
58
+ 1, 2007
59
+ 2, 2007
60
+ 3, 2007
61
+ 4, 2007
62
+ 5, 2007
63
+ 6, 2007
64
+ 7, 2007
65
+ 0, 2393
66
+ 1, 2379
67
+ 2, 2393
68
+ 3, 2373
69
+ 4, 2385
70
+ 5, 2379
71
+ 6, 2377
72
+ 7, 2391
73
+ 0, 14589
74
+ 1, 14587
75
+ 2, 14587
76
+ 3, 14587
77
+ 4, 14587
78
+ 5, 14587
79
+ 6, 14587
80
+ 7, 14587
81
+ 0, 15127
82
+ 1, 15125
83
+ 2, 15125
84
+ 3, 15125
85
+ 4, 15125
86
+ 5, 15125
87
+ 6, 15125
88
+ 7, 15125
89
+ 0, 16677
90
+ 1, 16675
91
+ 2, 16683
92
+ 3, 16683
93
+ 4, 16683
94
+ 5, 16683
95
+ 6, 16683
96
+ 7, 16683
97
+ 0, 16685
98
+ 1, 16683
99
+ 2, 16683
100
+ 3, 16683
101
+ 4, 16683
102
+ 5, 16683
103
+ 6, 16683
104
+ 7, 16683
105
+ 0, 16993
106
+ 1, 16991
107
+ 2, 16991
108
+ 3, 16991
109
+ 4, 16991
110
+ 5, 16991
111
+ 6, 16991
112
+ 7, 16991
113
+ 0, 17959
114
+ 1, 17957
115
+ 2, 17957
116
+ 3, 17957
117
+ 4, 17957
118
+ 5, 17957
119
+ 6, 17957
120
+ 7, 17957
121
+ 0, 35699
122
+ 1, 35697
123
+ 2, 35697
124
+ 3, 35697
125
+ 4, 35659
126
+ 5, 35697
127
+ 6, 35697
128
+ 7, 35697
129
+ 0, 35699
130
+ 1, 35697
131
+ 2, 35697
132
+ 3, 35697
133
+ 4, 35659
134
+ 5, 35697
135
+ 6, 35697
136
+ 7, 35697
137
+ 0, 35699
138
+ 1, 35697
139
+ 2, 35697
140
+ 3, 35697
141
+ 4, 35659
142
+ 5, 35697
143
+ 6, 35697
144
+ 7, 35697
145
+ 0, 37471
146
+ 1, 37449
147
+ 2, 37479
148
+ 3, 37493
149
+ 4, 37425
150
+ 5, 37493
151
+ 6, 37483
152
+ 7, 37479
153
+ 0, 38473
154
+ 1, 38451
155
+ 2, 38447
156
+ 3, 38449
157
+ 4, 38431
158
+ 5, 38477
159
+ 6, 38481
160
+ 7, 38479
161
+ 0, 39671
162
+ 1, 39697
163
+ 2, 39677
164
+ 3, 39737
165
+ 4, 39705
166
+ 5, 39679
167
+ 6, 39729
168
+ 7, 39745
169
+ 0, 51017
170
+ 1, 51011
171
+ 2, 51007
172
+ 3, 51011
173
+ 4, 50977
174
+ 5, 51009
175
+ 6, 51009
176
+ 7, 51009
177
+ 0, 51017
178
+ 1, 51011
179
+ 2, 51007
180
+ 3, 51011
181
+ 4, 50977
182
+ 5, 51009
183
+ 6, 51009
184
+ 7, 51009
185
+ 0, 3137
186
+ 1, 1540
187
+ 2, 1540
188
+ 3, 1540
189
+ 4, 1540
190
+ 5, 1540
191
+ 6, 1860
192
+ 7, 50485
tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq8192__gbs16__cap2p40625/rep1/run.env ADDED
@@ -0,0 +1,70 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ backend=ultraep
2
+ config=PONLY
3
+ family=tierA
4
+ repo=/workspace/Megatron-UltraEP
5
+ branch=ultraep-bracing-regions-through-kernels
6
+ commit=03d57ef7
7
+ num_layers=16
8
+ hidden=4096
9
+ ffn_hidden=14336
10
+ moe_ffn_hidden=1536
11
+ num_experts=8
12
+ topk=2
13
+ experts_per_rank=1
14
+ expert_params=18874368
15
+ heads=32
16
+ kv_groups=8
17
+ kv_channels=
18
+ qk_layernorm=0
19
+ shared_ffn=0
20
+ shared_expert_overlap=0
21
+ moe_token_dispatcher=flex
22
+ mla=0
23
+ dense_first_layers=0
24
+ moe_layers=16
25
+ rotary_base=1000000
26
+ rotary_scaling=
27
+ qkv_bias=0
28
+ router_balancing=aux_loss
29
+ aux_loss_coeff=1e-2
30
+ router_pre_softmax=0
31
+ router_fusion=0
32
+ router_dtype=fp32
33
+ device_initiated_grouped_gemm=1
34
+ norm_epsilon=1e-5
35
+ seq_length=8192
36
+ micro_batch_size=1
37
+ global_batch_size=16
38
+ num_microbatches=2
39
+ assignments_per_rank=16384
40
+ mean_tokens_per_expert=16384
41
+ slot_policy=one
42
+ slots_per_rank=1
43
+ capacity_mode=static
44
+ capacity=2.40625
45
+ cuda_graph=1
46
+ probe_mode=0
47
+ async_grad_reduce=0
48
+ branch_role=timing
49
+ memory_dump_mode=none
50
+ memory_snapshot_path=
51
+ cuda_graph_warmup_steps=3
52
+ recompute=none
53
+ force_balance=0
54
+ train_iters=12
55
+ ep_size=8
56
+ echo_dispatch_overlap=1
57
+ echo_dispatch_recompute=0
58
+ ultraep_quota_min_tokens=2048
59
+ ultraep_weight_sync_join=
60
+ profile_enable=0
61
+ kernel_trace=0
62
+ fp8_mode=mxfp8
63
+ fp8_recipe=mxfp8
64
+ training_seed=1234
65
+ learning_rate=1e-4
66
+ min_learning_rate=1.0e-5
67
+ lr_warmup_iters=5
68
+ state_dump_enable=0
69
+ state_dump_dir=
70
+ moonep_assert_no_pending_reduce=0
tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq8192__gbs16__cap2p40625/rep1/train.log ADDED
@@ -0,0 +1,1024 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ==============================================================
2
+ PONLY / ultraep (slots/rank=1 policy=one)
3
+ E=8 K=2 H'=1536 m=1
4
+ L=16 seq=8192 gbs=16 (2 microbatches)
5
+ cap_mode=static cap=2.40625 cg=1 recompute=none
6
+ memory_dump=none snapshots=/workspace/tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq8192__gbs16__cap2p40625/rep1/memory_snapshots/snapshot.pickle
7
+ repo=/workspace/Megatron-UltraEP @ 03d57ef7
8
+ ==============================================================
9
+ W0830 07:18:48.605000 739083 torch/distributed/run.py:766]
10
+ W0830 07:18:48.605000 739083 torch/distributed/run.py:766] *****************************************
11
+ W0830 07:18:48.605000 739083 torch/distributed/run.py:766] Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed.
12
+ W0830 07:18:48.605000 739083 torch/distributed/run.py:766] *****************************************
13
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
14
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
15
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
16
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
17
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
18
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
19
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
20
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
21
+ using world size: 8, data-parallel size: 8, context-parallel size: 1, hierarchical context-parallel sizes: None, tensor-model-parallel size: 1, pipeline-model-parallel size: 1
22
+ WARNING: overriding default arguments for tokenizer_type:GPT2BPETokenizer with tokenizer_type:GPT2BPETokenizer
23
+ Number of virtual stages per pipeline stage: None
24
+ accumulate and all-reduce gradients in fp32 for bfloat16 data type.
25
+ using torch.bfloat16 for parameters ...
26
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
27
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
28
+ ------------------------ arguments ------------------------
29
+ account_for_embedding_in_pipeline_split ......... False
30
+ account_for_loss_in_pipeline_split .............. False
31
+ accumulate_allreduce_grads_in_fp32 .............. True
32
+ adam_beta1 ...................................... 0.9
33
+ adam_beta2 ...................................... 0.999
34
+ adam_eps ........................................ 1e-08
35
+ add_bias_linear ................................. False
36
+ add_position_embedding .......................... False
37
+ add_qkv_bias .................................... False
38
+ adlr_autoresume ................................. False
39
+ adlr_autoresume_interval ........................ 1000
40
+ align_grad_reduce ............................... True
41
+ align_param_gather .............................. False
42
+ app_tag_run_name ................................ None
43
+ app_tag_run_version ............................. 0.0.0
44
+ apply_layernorm_1p .............................. False
45
+ apply_query_key_layer_scaling ................... False
46
+ apply_residual_connection_post_layernorm ........ False
47
+ apply_rope_fusion ............................... True
48
+ async_save ...................................... None
49
+ async_tensor_model_parallel_allreduce ........... True
50
+ attention_backend ............................... AttnBackend.auto
51
+ attention_dropout ............................... 0.0
52
+ attention_softmax_in_fp32 ....................... False
53
+ auto_detect_ckpt_format ......................... False
54
+ barrier_with_L1_time ............................ True
55
+ bert_binary_head ................................ True
56
+ bert_embedder_type .............................. megatron
57
+ bert_load ....................................... None
58
+ bf16 ............................................ True
59
+ bias_dropout_fusion ............................. True
60
+ bias_gelu_fusion ................................ False
61
+ bias_swiglu_fusion .............................. True
62
+ biencoder_projection_dim ........................ 0
63
+ biencoder_shared_query_context_model ............ False
64
+ block_data_path ................................. None
65
+ calc_ft_timeouts ................................ False
66
+ calculate_per_token_loss ........................ False
67
+ check_for_large_grads ........................... False
68
+ check_for_nan_in_loss_and_grad .................. False
69
+ check_for_spiky_loss ............................ False
70
+ check_weight_hash_across_dp_replicas_interval ... None
71
+ ckpt_assume_constant_structure .................. False
72
+ ckpt_convert_format ............................. None
73
+ ckpt_convert_save ............................... None
74
+ ckpt_convert_update_legacy_dist_opt_format ...... False
75
+ ckpt_format ..................................... torch_dist
76
+ ckpt_fully_parallel_load ........................ False
77
+ ckpt_fully_parallel_save ........................ True
78
+ ckpt_fully_parallel_save_deprecated ............. False
79
+ ckpt_step ....................................... None
80
+ classes_fraction ................................ 1.0
81
+ clip_grad ....................................... 1.0
82
+ clone_scatter_output_in_embedding ............... True
83
+ config_logger_dir ...............................
84
+ consumed_train_samples .......................... 0
85
+ consumed_valid_samples .......................... 0
86
+ context_parallel_size ........................... 1
87
+ cp_comm_type .................................... ['p2p']
88
+ create_attention_mask_in_dataloader ............. True
89
+ cross_entropy_fusion_impl ....................... native
90
+ cross_entropy_loss_fusion ....................... False
91
+ cuda_graph_scope ................................ full_iteration
92
+ cuda_graph_warmup_steps ......................... 3
93
+ data_args_path .................................. None
94
+ data_cache_path ................................. None
95
+ data_parallel_random_init ....................... False
96
+ data_parallel_sharding_strategy ................. no_shard
97
+ data_parallel_size .............................. 8
98
+ data_path ....................................... ['/workspace/data/wikitext/wikitext103_gpt2_text_document']
99
+ data_per_class_fraction ......................... 1.0
100
+ data_sharding ................................... True
101
+ dataloader_type ................................. single
102
+ ddp_average_in_collective ....................... False
103
+ ddp_bucket_size ................................. None
104
+ ddp_num_buckets ................................. None
105
+ ddp_pad_buckets_for_high_nccl_busbw ............. False
106
+ decoder_first_pipeline_num_layers ............... None
107
+ decoder_last_pipeline_num_layers ................ None
108
+ decoder_num_layers .............................. None
109
+ decoder_seq_length .............................. None
110
+ decoupled_lr .................................... None
111
+ decoupled_min_lr ................................ None
112
+ decrease_batch_size_if_needed ................... False
113
+ defer_embedding_wgrad_compute ................... False
114
+ delay_wgrad_compute ............................. False
115
+ deprecated_use_mcore_models ..................... True
116
+ deterministic_mode .............................. False
117
+ dino_bottleneck_size ............................ 256
118
+ dino_freeze_last_layer .......................... 1
119
+ dino_head_hidden_size ........................... 2048
120
+ dino_local_crops_number ......................... 10
121
+ dino_local_img_size ............................. 96
122
+ dino_norm_last_layer ............................ False
123
+ dino_teacher_temp ............................... 0.07
124
+ dino_warmup_teacher_temp ........................ 0.04
125
+ dino_warmup_teacher_temp_epochs ................. 30
126
+ disable_bf16_reduced_precision_matmul ........... False
127
+ disable_mamba_mem_eff_path ...................... False
128
+ disable_straggler_on_startup .................... False
129
+ dist_ckpt_format_deprecated ..................... None
130
+ dist_ckpt_strictness ............................ assume_ok_unexpected
131
+ distribute_saved_activations .................... False
132
+ distributed_backend ............................. nccl
133
+ distributed_timeout_minutes ..................... 10
134
+ embedding_init_method_std ....................... None
135
+ embedding_path .................................. None
136
+ empty_unused_memory_level ....................... 0
137
+ enable_cuda_graph ............................... True
138
+ enable_experimental ............................. False
139
+ enable_ft_package ............................... False
140
+ enable_gloo_process_groups ...................... True
141
+ enable_msc ...................................... True
142
+ enable_one_logger ............................... True
143
+ encoder_num_layers .............................. 16
144
+ encoder_seq_length .............................. 8192
145
+ end_weight_decay ................................ 0.1
146
+ eod_mask_loss ................................... False
147
+ error_injection_rate ............................ 0
148
+ error_injection_type ............................ transient_error
149
+ eval_interval ................................... 1000000
150
+ eval_iters ...................................... 0
151
+ evidence_data_path .............................. None
152
+ exit_duration_in_mins ........................... None
153
+ exit_interval ................................... None
154
+ exit_on_missing_checkpoint ...................... False
155
+ exit_signal_handler ............................. False
156
+ exp_avg_dtype ................................... torch.float32
157
+ exp_avg_sq_dtype ................................ torch.float32
158
+ expert_model_parallel_size ...................... 8
159
+ expert_tensor_parallel_size ..................... 1
160
+ export_force_local_attention .................... False
161
+ export_kd_cfg ................................... None
162
+ export_kd_teacher_ckpt_format ................... None
163
+ export_kd_teacher_load .......................... None
164
+ export_kv_cache_quant ........................... False
165
+ export_legacy_megatron .......................... False
166
+ export_model_type ............................... GPTModel
167
+ export_moe_apply_probs_on_input ................. False
168
+ export_qk_l2_norm ............................... False
169
+ export_quant_cfg ................................ None
170
+ export_real_quant_cfg ........................... None
171
+ export_te_mcore_model ........................... False
172
+ external_cuda_graph ............................. False
173
+ ffn_hidden_size ................................. 14336
174
+ finetune ........................................ False
175
+ finetune_data_split ............................. train
176
+ finetune_hf_dataset ............................. None
177
+ first_last_layers_bf16 .......................... False
178
+ flash_decode .................................... False
179
+ fp16 ............................................ False
180
+ fp16_lm_cross_entropy ........................... False
181
+ fp32_residual_connection ........................ False
182
+ fp8 ............................................. e4m3
183
+ fp8_amax_compute_algo ........................... most_recent
184
+ fp8_amax_history_len ............................ 1
185
+ fp8_interval .................................... 1
186
+ fp8_margin ...................................... 0
187
+ fp8_param_gather ................................ False
188
+ fp8_recipe ...................................... mxfp8
189
+ fp8_wgrad ....................................... True
190
+ fsdp_double_buffer .............................. False
191
+ global_batch_size ............................... 16
192
+ grad_reduce_in_bf16 ............................. False
193
+ gradient_accumulation_fusion .................... True
194
+ gradient_reduce_div_fusion ...................... True
195
+ group_query_attention ........................... True
196
+ head_lr_mult .................................... 1.0
197
+ heterogeneous_layers_config_encoded_json ........ None
198
+ heterogeneous_layers_config_path ................ None
199
+ hidden_dropout .................................. 0.0
200
+ hidden_size ..................................... 4096
201
+ hierarchical_context_parallel_sizes ............. None
202
+ high_priority_stream_groups ..................... []
203
+ hybrid_attention_ratio .......................... 0.0
204
+ hybrid_mlp_ratio ................................ 0.0
205
+ hybrid_override_pattern ......................... None
206
+ hysteresis ...................................... 2
207
+ ict_head_size ................................... None
208
+ ict_load ........................................ None
209
+ img_h ........................................... 224
210
+ img_w ........................................... 224
211
+ indexer_batch_size .............................. 128
212
+ indexer_log_interval ............................ 1000
213
+ inference_batch_times_seqlen_threshold .......... -1
214
+ inference_dynamic_batching ...................... False
215
+ inference_dynamic_batching_buffer_guaranteed_fraction 0.2
216
+ inference_dynamic_batching_buffer_overflow_factor None
217
+ inference_dynamic_batching_buffer_size_gb ....... 40.0
218
+ inference_dynamic_batching_chunk_size ........... 256
219
+ inference_dynamic_batching_max_requests_override None
220
+ inference_dynamic_batching_max_tokens_override .. None
221
+ inference_dynamic_batching_num_cuda_graphs ...... 16
222
+ inference_max_batch_size ........................ 8
223
+ inference_max_seq_length ........................ 2560
224
+ inference_rng_tracker ........................... False
225
+ init_method_std ................................. 0.01
226
+ init_method_xavier_uniform ...................... False
227
+ init_model_with_meta_device ..................... False
228
+ initial_loss_scale .............................. 4294967296
229
+ inprocess_active_world_size ..................... 8
230
+ inprocess_barrier_timeout ....................... 120
231
+ inprocess_completion_timeout .................... 120
232
+ inprocess_empty_cuda_cache ...................... False
233
+ inprocess_granularity ........................... node
234
+ inprocess_hard_timeout .......................... 90
235
+ inprocess_heartbeat_interval .................... 30
236
+ inprocess_heartbeat_timeout ..................... 60
237
+ inprocess_last_call_wait ........................ 1
238
+ inprocess_max_iterations ........................ None
239
+ inprocess_monitor_process_interval .............. 1.0
240
+ inprocess_monitor_thread_interval ............... 1.0
241
+ inprocess_progress_watchdog_interval ............ 1.0
242
+ inprocess_restart ............................... False
243
+ inprocess_soft_timeout .......................... 60
244
+ inprocess_termination_grace_time ................ 1
245
+ is_hybrid_model ................................. False
246
+ iter_per_epoch .................................. 1250
247
+ iterations_to_skip .............................. []
248
+ keep_fp8_transpose_cache_when_using_custom_fsdp . False
249
+ kitchen_config_file ............................. None
250
+ kitchen_recipe_number ........................... None
251
+ kv_channels ..................................... 128
252
+ kv_lora_rank .................................... 32
253
+ lazy_mpu_init ................................... None
254
+ load ............................................ None
255
+ load_main_params_from_ckpt ...................... None
256
+ load_model_opt_format ........................... False
257
+ local_rank ...................................... 0
258
+ log_energy ...................................... False
259
+ log_interval .................................... 1
260
+ log_loss_scale_to_tensorboard ................... True
261
+ log_memory_to_tensorboard ....................... False
262
+ log_num_zeros_in_grad ........................... False
263
+ log_params_norm ................................. False
264
+ log_progress .................................... False
265
+ log_straggler ................................... False
266
+ log_throughput .................................. True
267
+ log_timers_to_tensorboard ....................... False
268
+ log_validation_ppl_to_tensorboard ............... False
269
+ log_world_size_to_tensorboard ................... False
270
+ logging_level ................................... None
271
+ loss_scale ...................................... None
272
+ loss_scale_window ............................... 1000
273
+ lr .............................................. 0.0001
274
+ lr_decay_iters .................................. 8
275
+ lr_decay_samples ................................ None
276
+ lr_decay_style .................................. cosine
277
+ lr_warmup_fraction .............................. None
278
+ lr_warmup_init .................................. 0.0
279
+ lr_warmup_iters ................................. 5
280
+ lr_warmup_samples ............................... 0
281
+ lr_wsd_decay_iters .............................. None
282
+ lr_wsd_decay_samples ............................ None
283
+ lr_wsd_decay_style .............................. exponential
284
+ main_grads_dtype ................................ torch.float32
285
+ main_params_dtype ............................... torch.float32
286
+ make_vocab_size_divisible_by .................... 128
287
+ mamba_head_dim .................................. 64
288
+ mamba_num_groups ................................ 8
289
+ mamba_num_heads ................................. None
290
+ mamba_state_dim ................................. 128
291
+ manual_gc ....................................... False
292
+ manual_gc_eval .................................. True
293
+ manual_gc_interval .............................. 0
294
+ mask_factor ..................................... 1.0
295
+ mask_prob ....................................... 0.15
296
+ mask_type ....................................... random
297
+ masked_softmax_fusion ........................... False
298
+ max_position_embeddings ......................... 32768
299
+ max_tokens_to_oom ............................... 12000
300
+ memory_snapshot_path ............................ snapshot.pickle
301
+ merge_file ...................................... /workspace/data/wikitext/gpt2-merges.txt
302
+ micro_batch_size ................................ 1
303
+ microbatch_group_size_per_vp_stage .............. None
304
+ mid_level_dataset_surplus ....................... 0.005
305
+ min_loss_scale .................................. 1.0
306
+ min_lr .......................................... 1e-05
307
+ mlp_chunks_for_prefill .......................... 1
308
+ mmap_bin_files .................................. True
309
+ mock_data ....................................... False
310
+ moe_apply_probs_on_input ........................ False
311
+ moe_aux_loss_coeff .............................. 0.01
312
+ moe_capacity_monitor ............................ True
313
+ moe_deepep_num_sms .............................. 20
314
+ moe_echo_enable_random_offloading ............... False
315
+ moe_echo_expert_dispatch_overlap ................ False
316
+ moe_echo_expert_dispatcher_type ................. hybridep
317
+ moe_echo_recompute_expert_dispatch .............. False
318
+ moe_enable_deepep ............................... False
319
+ moe_enable_echo ................................. False
320
+ moe_enable_hybridep ............................. True
321
+ moe_enable_ultraep .............................. True
322
+ moe_expert_capacity_factor ...................... None
323
+ moe_extended_tp ................................. False
324
+ moe_ffn_hidden_size ............................. 1536
325
+ moe_grouped_gemm ................................ True
326
+ moe_hybridep_num_sms ............................ 16
327
+ moe_input_jitter_eps ............................ None
328
+ moe_layer_freq .................................. 1
329
+ moe_layer_recompute ............................. False
330
+ moe_num_echo_experts ............................ None
331
+ moe_num_redundant_experts_per_rank .............. 1
332
+ moe_pad_expert_input_to_capacity ................ False
333
+ moe_per_layer_logging ........................... False
334
+ moe_permute_fusion .............................. False
335
+ moe_received_token_capacity ..................... 2.40625
336
+ moe_router_bias_update_rate ..................... 0.001
337
+ moe_router_dtype ................................ fp32
338
+ moe_router_enable_expert_bias ................... False
339
+ moe_router_force_load_balancing ................. False
340
+ moe_router_group_topk ........................... None
341
+ moe_router_load_balancing_type .................. aux_loss
342
+ moe_router_num_groups ........................... None
343
+ moe_router_padding_for_fp8 ...................... False
344
+ moe_router_pre_softmax .......................... False
345
+ moe_router_score_function ....................... softmax
346
+ moe_router_topk ................................. 2
347
+ moe_router_topk_scaling_factor .................. None
348
+ moe_shared_expert_intermediate_size ............. None
349
+ moe_shared_expert_overlap ....................... False
350
+ moe_token_dispatcher_type ....................... flex
351
+ moe_token_drop_policy ........................... probs
352
+ moe_ultraep_max_microbatches .................... None
353
+ moe_ultraep_overlap_weight_sync ................. True
354
+ moe_ultraep_weight_sync_join .................... before_dispatch
355
+ moe_upcycling_granularity ....................... 1
356
+ moe_use_device_initiated_grouped_gemm ........... True
357
+ moe_use_legacy_grouped_gemm ..................... False
358
+ moe_use_upcycling ............................... False
359
+ moe_z_loss_coeff ................................ None
360
+ mrope_section ................................... None
361
+ mscale .......................................... 1.0
362
+ mscale_all_dim .................................. 0.0
363
+ mtp_loss_scaling_factor ......................... 0.1
364
+ mtp_num_layers .................................. None
365
+ multi_latent_attention .......................... False
366
+ nccl_all_reduce_for_prefill ..................... False
367
+ nccl_communicator_config_path ................... None
368
+ nccl_ub ......................................... False
369
+ no_load_optim ................................... None
370
+ no_load_rng ..................................... None
371
+ no_persist_layer_norm ........................... False
372
+ no_rope_freq .................................... None
373
+ no_save_optim ................................... None
374
+ no_save_rng ..................................... None
375
+ non_persistent_ckpt_type ........................ None
376
+ non_persistent_global_ckpt_dir .................. None
377
+ non_persistent_local_ckpt_algo .................. fully_parallel
378
+ non_persistent_local_ckpt_dir ................... None
379
+ non_persistent_save_interval .................... None
380
+ norm_epsilon .................................... 1e-05
381
+ normalization ................................... RMSNorm
382
+ num_attention_heads ............................. 32
383
+ num_channels .................................... 3
384
+ num_classes ..................................... 1000
385
+ num_dataset_builder_threads ..................... 1
386
+ num_distributed_optimizer_instances ............. 1
387
+ num_experts ..................................... 8
388
+ num_layers ...................................... 16
389
+ num_layers_at_end_in_bf16 ....................... 1
390
+ num_layers_at_start_in_bf16 ..................... 1
391
+ num_layers_per_virtual_pipeline_stage ........... None
392
+ num_query_groups ................................ 8
393
+ num_virtual_stages_per_pipeline_rank ............ None
394
+ num_workers ..................................... 2
395
+ object_storage_cache_path ....................... None
396
+ one_logger_async ................................ False
397
+ one_logger_project .............................. megatron-lm
398
+ one_logger_run_name ............................. None
399
+ onnx_safe ....................................... None
400
+ openai_gelu ..................................... False
401
+ optimizer ....................................... adam
402
+ optimizer_cpu_offload ........................... False
403
+ optimizer_offload_fraction ...................... 1.0
404
+ output_bert_embeddings .......................... False
405
+ overlap_cpu_optimizer_d2h_h2d ................... False
406
+ overlap_grad_reduce ............................. True
407
+ overlap_moe_expert_parallel_comm ................ False
408
+ overlap_p2p_comm ................................ False
409
+ overlap_p2p_comm_warmup_flush ................... False
410
+ overlap_param_gather ............................ True
411
+ overlap_param_gather_with_optimizer_step ........ False
412
+ override_opt_param_scheduler .................... False
413
+ params_dtype .................................... torch.bfloat16
414
+ patch_dim ....................................... 16
415
+ per_split_data_args_path ........................ None
416
+ perform_initialization .......................... True
417
+ pin_cpu_grads ................................... True
418
+ pin_cpu_params .................................. True
419
+ pipeline_model_parallel_comm_backend ............ None
420
+ pipeline_model_parallel_layout .................. None
421
+ pipeline_model_parallel_size .................... 1
422
+ position_embedding_type ......................... rope
423
+ pretrained_checkpoint ........................... None
424
+ profile ......................................... False
425
+ profile_ranks ................................... [0]
426
+ profile_step_end ................................ 12
427
+ profile_step_start .............................. 10
428
+ q_lora_rank ..................................... None
429
+ qk_head_dim ..................................... 128
430
+ qk_l2_norm ...................................... False
431
+ qk_layernorm .................................... False
432
+ qk_pos_emb_head_dim ............................. 64
433
+ query_in_block_prob ............................. 0.1
434
+ rampup_batch_size ............................... None
435
+ rank ............................................ 0
436
+ recompute_granularity ........................... None
437
+ recompute_method ................................ None
438
+ recompute_modules ............................... None
439
+ recompute_num_layers ............................ None
440
+ record_memory_history ........................... False
441
+ relative_attention_max_distance ................. 128
442
+ relative_attention_num_buckets .................. 32
443
+ replication ..................................... False
444
+ replication_factor .............................. 2
445
+ replication_jump ................................ None
446
+ rerun_mode ...................................... validate_results
447
+ reset_attention_mask ............................ False
448
+ reset_position_ids .............................. False
449
+ result_rejected_tracker_filename ................ None
450
+ retriever_report_topk_accuracies ................ []
451
+ retriever_score_scaling ......................... False
452
+ retriever_seq_length ............................ 256
453
+ retro_add_retriever ............................. False
454
+ retro_attention_gate ............................ 1
455
+ retro_cyclic_train_iters ........................ None
456
+ retro_encoder_attention_dropout ................. 0.1
457
+ retro_encoder_hidden_dropout .................... 0.1
458
+ retro_encoder_layers ............................ 2
459
+ retro_num_neighbors ............................. 2
460
+ retro_num_retrieved_chunks ...................... 2
461
+ retro_project_dir ............................... None
462
+ retro_verify_neighbor_count ..................... True
463
+ reuse_grad_buf_for_mxfp8_param_ag ............... False
464
+ rope_scaling_factor ............................. 8.0
465
+ rope_type ....................................... None
466
+ rotary_base ..................................... 1000000
467
+ rotary_interleaved .............................. False
468
+ rotary_percent .................................. 1.0
469
+ rotary_scaling_factor ........................... 1.0
470
+ rotary_seq_len_interpolation_factor ............. None
471
+ run_workload_inspector_server ................... False
472
+ sample_rate ..................................... 1.0
473
+ save ............................................ None
474
+ save_interval ................................... None
475
+ save_retain_interval ............................ None
476
+ scatter_gather_tensors_in_pipeline .............. True
477
+ seed ............................................ 1234
478
+ seq_length ...................................... 8192
479
+ sequence_parallel ............................... False
480
+ sft ............................................. False
481
+ sft_tokenizer_prompt_format ..................... nemotron-h-aligned
482
+ sgd_momentum .................................... 0.9
483
+ short_seq_prob .................................. 0.1
484
+ skip_train ...................................... False
485
+ skipped_train_samples ........................... 0
486
+ spec ............................................ None
487
+ split ........................................... 100,0,0
488
+ squared_relu .................................... False
489
+ start_weight_decay .............................. 0.1
490
+ straggler_ctrlr_port ............................ 65535
491
+ straggler_minmax_count .......................... 1
492
+ suggested_communication_unit_size ............... None
493
+ swiglu .......................................... True
494
+ swin_backbone_type .............................. tiny
495
+ symmetric_ar_type ............................... None
496
+ te_rng_tracker .................................. True
497
+ tensor_model_parallel_size ...................... 1
498
+ tensorboard_dir ................................. None
499
+ tensorboard_log_interval ........................ 1
500
+ tensorboard_queue_size .......................... 1000
501
+ test_data_path .................................. None
502
+ test_mode ....................................... False
503
+ tiktoken_num_special_tokens ..................... 1000
504
+ tiktoken_pattern ................................ None
505
+ tiktoken_special_tokens ......................... None
506
+ timing_log_level ................................ 0
507
+ timing_log_option ............................... minmax
508
+ titles_data_path ................................ None
509
+ tokenizer_model ................................. None
510
+ tokenizer_type .................................. GPT2BPETokenizer
511
+ torch_fsdp2_reshard_after_forward ............... True
512
+ tp_comm_bootstrap_backend ....................... nccl
513
+ tp_comm_bulk_dgrad .............................. True
514
+ tp_comm_bulk_wgrad .............................. True
515
+ tp_comm_overlap ................................. False
516
+ tp_comm_overlap_ag .............................. True
517
+ tp_comm_overlap_cfg ............................. None
518
+ tp_comm_overlap_rs .............................. True
519
+ tp_comm_overlap_rs_dgrad ........................ False
520
+ tp_comm_split_ag ................................ True
521
+ tp_comm_split_rs ................................ True
522
+ train_data_path ................................. None
523
+ train_iters ..................................... 12
524
+ train_samples ................................... None
525
+ train_sync_interval ............................. None
526
+ transformer_impl ................................ transformer_engine
527
+ transformer_pipeline_model_parallel_size ........ 1
528
+ untie_embeddings_and_output_weights ............. True
529
+ use_checkpoint_args ............................. False
530
+ use_checkpoint_opt_param_scheduler .............. False
531
+ use_cpu_initialization .......................... None
532
+ use_custom_fsdp ................................. False
533
+ use_dist_ckpt ................................... True
534
+ use_dist_ckpt_deprecated ........................ False
535
+ use_distributed_optimizer ....................... True
536
+ use_flash_attn .................................. False
537
+ use_legacy_models ............................... False
538
+ use_mp_args_from_checkpoint_args ................ False
539
+ use_one_sent_docs ............................... False
540
+ use_persistent_ckpt_worker ...................... False
541
+ use_precision_aware_optimizer ................... False
542
+ use_pytorch_profiler ............................ False
543
+ use_ring_exchange_p2p ........................... False
544
+ use_rope_scaling ................................ False
545
+ use_rotary_position_embeddings .................. False
546
+ use_sharp ....................................... False
547
+ use_tokenizer_model_from_checkpoint_args ........ True
548
+ use_torch_fsdp2 ................................. False
549
+ use_torch_optimizer_for_cpu_offload ............. False
550
+ use_tp_pp_dp_mapping ............................ False
551
+ v_head_dim ...................................... 128
552
+ valid_data_path ................................. None
553
+ variable_seq_lengths ............................ False
554
+ virtual_pipeline_model_parallel_size ............ None
555
+ vision_backbone_type ............................ vit
556
+ vision_pretraining .............................. False
557
+ vision_pretraining_type ......................... classify
558
+ vocab_extra_ids ................................. 0
559
+ vocab_file ...................................... /workspace/data/wikitext/gpt2-vocab.json
560
+ vocab_size ...................................... None
561
+ wandb_exp_name ..................................
562
+ wandb_project ...................................
563
+ wandb_save_dir ..................................
564
+ weight_decay .................................... 0.1
565
+ weight_decay_incr_style ......................... constant
566
+ wgrad_deferral_limit ............................ 0
567
+ world_size ...................................... 8
568
+ yaml_cfg ........................................ None
569
+ -------------------- end of arguments ---------------------
570
+ INFO:megatron.core.num_microbatches_calculator:setting number of microbatches to constant 2
571
+ > building GPT2BPETokenizer tokenizer ...
572
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
573
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
574
+ > padded vocab (size: 50257) with 47 dummy tokens (new size: 50304)
575
+ WARNING:megatron.core.rerun_state_machine:RerunStateMachine initialized in mode RerunMode.VALIDATE_RESULTS
576
+ > initializing torch distributed ...
577
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
578
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
579
+ WARNING: one_logger package is required to enable e2e metrics tracking. please go to https://confluence.nvidia.com/display/MLWFO/Package+Repositories for details to install it
580
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
581
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
582
+ [Gloo] Rank 0 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
583
+ [Gloo] Rank 4 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
584
+ [Gloo] Rank 1 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
585
+ [Gloo] Rank 2 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
586
+ [Gloo] Rank 3 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
587
+ [Gloo] Rank [Gloo] Rank 5 is connected to 67 is connected to peer ranks. 7Expected number of connected peer ranks is : peer ranks. 7Expected number of connected peer ranks is :
588
+ 7
589
+ [Gloo] Rank 7 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
590
+ [Gloo] Rank 0 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
591
+ [Gloo] Rank 1 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
592
+ [Gloo] Rank 2 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
593
+ [Gloo] Rank [Gloo] Rank 36 is connected to is connected to 77 peer ranks. peer ranks. Expected number of connected peer ranks is : Expected number of connected peer ranks is : 77
594
+
595
+ [Gloo] Rank 4 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
596
+ [Gloo] Rank 5 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
597
+ [Gloo] Rank 7 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
598
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
599
+ > initialized tensor model parallel with size 1
600
+ > initialized pipeline model parallel with size 1
601
+ > setting random seeds to 1234 ...
602
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
603
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
604
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
605
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
606
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
607
+ [Gloo] Rank [Gloo] Rank 00 is connected to is connected to 00 peer ranks. peer ranks. Expected number of connected peer ranks is : Expected number of connected peer ranks is : 00
608
+
609
+ [rank2]:[W830 07:19:06.693799211 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 2] using GPU 2 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
610
+ [rank5]:[W830 07:19:06.693809764 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 5] using GPU 5 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
611
+ [rank7]:[W830 07:19:06.693898300 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 7] using GPU 7 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
612
+ [rank1]:[W830 07:19:06.693934259 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 1] using GPU 1 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
613
+ [rank6]:[W830 07:19:06.693988089 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 6] using GPU 6 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
614
+ [rank4]:[W830 07:19:06.694048781 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 4] using GPU 4 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
615
+ [rank3]:[W830 07:19:06.695635923 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 3] using GPU 3 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
616
+ > compiling dataset index builder ...
617
+ make: Entering directory '/workspace/Megatron-UltraEP/megatron/core/datasets'
618
+ make: Nothing to be done for 'default'.
619
+ make: Leaving directory '/workspace/Megatron-UltraEP/megatron/core/datasets'
620
+ >>> done with dataset index builder. Compilation time: 0.034 seconds
621
+ WARNING: constraints for invoking optimized fused softmax kernel are not met. We default back to unfused kernel invocations.
622
+ > compiling and loading fused kernels ...
623
+ [rank0]:[W830 07:19:06.851927949 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 0] using GPU 0 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
624
+ >>> done with compiling and loading fused kernels. Compilation time: 3.972 seconds
625
+ time to initialize megatron (seconds): 7.370
626
+ [after megatron is initialized] datetime: 2026-08-30 07:19:12
627
+ building GPT model ...
628
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
629
+ warnings.warn(
630
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
631
+ warnings.warn(
632
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
633
+ warnings.warn(
634
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
635
+ warnings.warn(
636
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
637
+ warnings.warn(
638
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
639
+ warnings.warn(
640
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
641
+ warnings.warn(
642
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
643
+ warnings.warn(
644
+ [INFO] Use MNNVL fabric: False
645
+ [WARN] MAX_NUM_NVL_PEERS is not set. Using detected value 8.
646
+ NVSHMEM v3.7.2
647
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
648
+
649
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
650
+
651
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
652
+
653
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
654
+
655
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
656
+
657
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
658
+
659
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
660
+
661
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
662
+
663
+ UltraEP: NVL domain size <= 8; forcing weight_sync plan mode to direct (was adaptive).
664
+ > number of parameters on (tensor, pipeline) model parallel rank (0, 0): 1687818240
665
+ INFO:megatron.core.distributed.distributed_data_parallel:Setting up DistributedDataParallel with config DistributedDataParallelConfig(grad_reduce_in_fp32=True, overlap_grad_reduce=True, overlap_param_gather=True, align_param_gather=False, use_distributed_optimizer=True, num_distributed_optimizer_instances=1, check_for_nan_in_grad=False, check_for_large_grads=False, bucket_size=40000000, pad_buckets_for_high_nccl_busbw=False, average_in_collective=False, fp8_param_gather=False, reuse_grad_buf_for_mxfp8_param_ag=False, use_custom_fsdp=False, data_parallel_sharding_strategy='no_shard', gradient_reduce_div_fusion=True, suggested_communication_unit_size=None, preserve_fp32_weights=True, keep_fp8_transpose_cache_when_using_custom_fsdp=False, nccl_ub=False, fsdp_double_buffer=False)
666
+ INFO:megatron.core.distributed.param_and_grad_buffer:Number of buckets for gradient all-reduce / reduce-scatter: 18
667
+ Params for bucket 1 (206045184 elements, 206045184 padded size):
668
+ module.output_layer.weight
669
+ Params for bucket 2 (41988096 elements, 41988096 padded size):
670
+ module.decoder.final_layernorm.weight
671
+ module.decoder.layers.15.mlp.router.weight
672
+ module.decoder.layers.15.self_attention.linear_qkv.layer_norm_weight
673
+ module.decoder.layers.15.self_attention.linear_proj.weight
674
+ module.decoder.layers.15.self_attention.linear_qkv.weight
675
+ module.decoder.layers.15.pre_mlp_layernorm.weight
676
+ Params for bucket 3 (41984000 elements, 41984000 padded size):
677
+ module.decoder.layers.14.pre_mlp_layernorm.weight
678
+ module.decoder.layers.14.self_attention.linear_qkv.weight
679
+ module.decoder.layers.14.mlp.router.weight
680
+ module.decoder.layers.14.self_attention.linear_proj.weight
681
+ module.decoder.layers.14.self_attention.linear_qkv.layer_norm_weight
682
+ Params for bucket 4 (41984000 elements, 41984000 padded size):
683
+ module.decoder.layers.13.self_attention.linear_qkv.weight
684
+ module.decoder.layers.13.pre_mlp_layernorm.weight
685
+ module.decoder.layers.13.mlp.router.weight
686
+ module.decoder.layers.13.self_attention.linear_qkv.layer_norm_weight
687
+ module.decoder.layers.13.self_attention.linear_proj.weight
688
+ Params for bucket 5 (41984000 elements, 41984000 padded size):
689
+ module.decoder.layers.12.pre_mlp_layernorm.weight
690
+ module.decoder.layers.12.self_attention.linear_qkv.weight
691
+ module.decoder.layers.12.mlp.router.weight
692
+ module.decoder.layers.12.self_attention.linear_proj.weight
693
+ module.decoder.layers.12.self_attention.linear_qkv.layer_norm_weight
694
+ Params for bucket 6 (41984000 elements, 41984000 padded size):
695
+ module.decoder.layers.11.mlp.router.weight
696
+ module.decoder.layers.11.self_attention.linear_proj.weight
697
+ module.decoder.layers.11.self_attention.linear_qkv.layer_norm_weight
698
+ module.decoder.layers.11.pre_mlp_layernorm.weight
699
+ module.decoder.layers.11.self_attention.linear_qkv.weight
700
+ Params for bucket 7 (41984000 elements, 41984000 padded size):
701
+ module.decoder.layers.10.mlp.router.weight
702
+ module.decoder.layers.10.self_attention.linear_qkv.layer_norm_weight
703
+ module.decoder.layers.10.self_attention.linear_proj.weight
704
+ module.decoder.layers.10.self_attention.linear_qkv.weight
705
+ module.decoder.layers.10.pre_mlp_layernorm.weight
706
+ Params for bucket 8 (41984000 elements, 41984000 padded size):
707
+ module.decoder.layers.9.mlp.router.weight
708
+ module.decoder.layers.9.self_attention.linear_proj.weight
709
+ module.decoder.layers.9.self_attention.linear_qkv.layer_norm_weight
710
+ module.decoder.layers.9.pre_mlp_layernorm.weight
711
+ module.decoder.layers.9.self_attention.linear_qkv.weight
712
+ Params for bucket 9 (41984000 elements, 41984000 padded size):
713
+ module.decoder.layers.8.mlp.router.weight
714
+ module.decoder.layers.8.self_attention.linear_proj.weight
715
+ module.decoder.layers.8.self_attention.linear_qkv.layer_norm_weight
716
+ module.decoder.layers.8.self_attention.linear_qkv.weight
717
+ module.decoder.layers.8.pre_mlp_layernorm.weight
718
+ Params for bucket 10 (41984000 elements, 41984000 padded size):
719
+ module.decoder.layers.7.mlp.router.weight
720
+ module.decoder.layers.7.self_attention.linear_qkv.layer_norm_weight
721
+ module.decoder.layers.7.self_attention.linear_proj.weight
722
+ module.decoder.layers.7.pre_mlp_layernorm.weight
723
+ module.decoder.layers.7.self_attention.linear_qkv.weight
724
+ Params for bucket 11 (41984000 elements, 41984000 padded size):
725
+ module.decoder.layers.6.mlp.router.weight
726
+ module.decoder.layers.6.self_attention.linear_proj.weight
727
+ module.decoder.layers.6.self_attention.linear_qkv.layer_norm_weight
728
+ module.decoder.layers.6.self_attention.linear_qkv.weight
729
+ module.decoder.layers.6.pre_mlp_layernorm.weight
730
+ Params for bucket 12 (41984000 elements, 41984000 padded size):
731
+ module.decoder.layers.5.self_attention.linear_proj.weight
732
+ module.decoder.layers.5.pre_mlp_layernorm.weight
733
+ module.decoder.layers.5.self_attention.linear_qkv.weight
734
+ module.decoder.layers.5.mlp.router.weight
735
+ module.decoder.layers.5.self_attention.linear_qkv.layer_norm_weight
736
+ Params for bucket 13 (41984000 elements, 41984000 padded size):
737
+ module.decoder.layers.4.mlp.router.weight
738
+ module.decoder.layers.4.self_attention.linear_qkv.layer_norm_weight
739
+ module.decoder.layers.4.self_attention.linear_proj.weight
740
+ module.decoder.layers.4.self_attention.linear_qkv.weight
741
+ module.decoder.layers.4.pre_mlp_layernorm.weight
742
+ Params for bucket 14 (41984000 elements, 41984000 padded size):
743
+ module.decoder.layers.3.mlp.router.weight
744
+ module.decoder.layers.3.self_attention.linear_proj.weight
745
+ module.decoder.layers.3.self_attention.linear_qkv.layer_norm_weight
746
+ module.decoder.layers.3.self_attention.linear_qkv.weight
747
+ module.decoder.layers.3.pre_mlp_layernorm.weight
748
+ Params for bucket 15 (41984000 elements, 41984000 padded size):
749
+ module.decoder.layers.2.mlp.router.weight
750
+ module.decoder.layers.2.self_attention.linear_qkv.layer_norm_weight
751
+ module.decoder.layers.2.self_attention.linear_proj.weight
752
+ module.decoder.layers.2.self_attention.linear_qkv.weight
753
+ module.decoder.layers.2.pre_mlp_layernorm.weight
754
+ Params for bucket 16 (41984000 elements, 41984000 padded size):
755
+ module.decoder.layers.1.mlp.router.weight
756
+ module.decoder.layers.1.self_attention.linear_qkv.weight
757
+ module.decoder.layers.1.pre_mlp_layernorm.weight
758
+ module.decoder.layers.1.self_attention.linear_proj.weight
759
+ module.decoder.layers.1.self_attention.linear_qkv.layer_norm_weight
760
+ Params for bucket 17 (41984000 elements, 41984000 padded size):
761
+ module.decoder.layers.0.mlp.router.weight
762
+ module.decoder.layers.0.self_attention.linear_qkv.weight
763
+ module.decoder.layers.0.self_attention.linear_qkv.layer_norm_weight
764
+ module.decoder.layers.0.pre_mlp_layernorm.weight
765
+ module.decoder.layers.0.self_attention.linear_proj.weight
766
+ Params for bucket 18 (206045184 elements, 206045184 padded size):
767
+ module.embedding.word_embeddings.weight
768
+ INFO:megatron.core.distributed.param_and_grad_buffer:Number of buckets for gradient all-reduce / reduce-scatter: 7
769
+ Params for bucket 1 (44040192 elements, 44040192 padded size):
770
+ module.decoder.layers.13.mlp.experts.linear_fc2.weight0
771
+ module.decoder.layers.15.mlp.experts.linear_fc1.weight0
772
+ module.decoder.layers.15.mlp.experts.linear_fc2.weight0
773
+ module.decoder.layers.14.mlp.experts.linear_fc2.weight0
774
+ module.decoder.layers.14.mlp.experts.linear_fc1.weight0
775
+ Params for bucket 2 (50331648 elements, 50331648 padded size):
776
+ module.decoder.layers.12.mlp.experts.linear_fc2.weight0
777
+ module.decoder.layers.11.mlp.experts.linear_fc1.weight0
778
+ module.decoder.layers.13.mlp.experts.linear_fc1.weight0
779
+ module.decoder.layers.11.mlp.experts.linear_fc2.weight0
780
+ module.decoder.layers.12.mlp.experts.linear_fc1.weight0
781
+ Params for bucket 3 (44040192 elements, 44040192 padded size):
782
+ module.decoder.layers.10.mlp.experts.linear_fc1.weight0
783
+ module.decoder.layers.9.mlp.experts.linear_fc1.weight0
784
+ module.decoder.layers.10.mlp.experts.linear_fc2.weight0
785
+ module.decoder.layers.8.mlp.experts.linear_fc2.weight0
786
+ module.decoder.layers.9.mlp.experts.linear_fc2.weight0
787
+ Params for bucket 4 (50331648 elements, 50331648 padded size):
788
+ module.decoder.layers.8.mlp.experts.linear_fc1.weight0
789
+ module.decoder.layers.6.mlp.experts.linear_fc1.weight0
790
+ module.decoder.layers.7.mlp.experts.linear_fc1.weight0
791
+ module.decoder.layers.6.mlp.experts.linear_fc2.weight0
792
+ module.decoder.layers.7.mlp.experts.linear_fc2.weight0
793
+ Params for bucket 5 (44040192 elements, 44040192 padded size):
794
+ module.decoder.layers.5.mlp.experts.linear_fc1.weight0
795
+ module.decoder.layers.4.mlp.experts.linear_fc1.weight0
796
+ module.decoder.layers.5.mlp.experts.linear_fc2.weight0
797
+ module.decoder.layers.4.mlp.experts.linear_fc2.weight0
798
+ module.decoder.layers.3.mlp.experts.linear_fc2.weight0
799
+ Params for bucket 6 (50331648 elements, 50331648 padded size):
800
+ module.decoder.layers.3.mlp.experts.linear_fc1.weight0
801
+ module.decoder.layers.1.mlp.experts.linear_fc1.weight0
802
+ module.decoder.layers.2.mlp.experts.linear_fc1.weight0
803
+ module.decoder.layers.2.mlp.experts.linear_fc2.weight0
804
+ module.decoder.layers.1.mlp.experts.linear_fc2.weight0
805
+ Params for bucket 7 (18874368 elements, 18874368 padded size):
806
+ module.decoder.layers.0.mlp.experts.linear_fc2.weight0
807
+ module.decoder.layers.0.mlp.experts.linear_fc1.weight0
808
+ INFO:megatron.core.optimizer:Setting up optimizer with config OptimizerConfig(optimizer='adam', lr=0.0001, min_lr=1e-05, decoupled_lr=None, decoupled_min_lr=None, weight_decay=0.1, fp8_recipe='mxfp8', fp16=False, bf16=True, reuse_grad_buf_for_mxfp8_param_ag=False, params_dtype=torch.bfloat16, use_precision_aware_optimizer=False, store_param_remainders=True, main_grads_dtype=torch.float32, main_params_dtype=torch.float32, exp_avg_dtype=torch.float32, exp_avg_sq_dtype=torch.float32, loss_scale=None, initial_loss_scale=4294967296, min_loss_scale=1.0, loss_scale_window=1000, hysteresis=2, adam_beta1=0.9, adam_beta2=0.999, adam_eps=1e-08, sgd_momentum=0.9, use_distributed_optimizer=True, overlap_param_gather_with_optimizer_step=False, optimizer_cpu_offload=False, optimizer_offload_fraction=1.0, use_torch_optimizer_for_cpu_offload=False, overlap_cpu_optimizer_d2h_h2d=False, pin_cpu_grads=True, pin_cpu_params=True, clip_grad=1.0, log_num_zeros_in_grad=False, barrier_with_L1_time=True, timers=<megatron.core.timers.Timers object at 0x717d7a13cfb0>, config_logger_dir='')
809
+ INFO:megatron.core.optimizer_param_scheduler:> learning rate decay style: cosine
810
+ [after model, optimizer, and learning rate scheduler are built] datetime: 2026-08-30 07:19:15
811
+ > building train, validation, and test datasets ...
812
+ > datasets target sizes (minimum size):
813
+ train: 192
814
+ validation: 0
815
+ test: 0
816
+ INFO:megatron.core.datasets.blended_megatron_dataset_config:Let split_matrix = [(0, 1.0), None, None]
817
+ > building train, validation, and test datasets for GPT ...
818
+ INFO:megatron.core.datasets.blended_megatron_dataset_builder:Building GPTDataset splits with sizes=(192, 0, 0) and config=GPTDatasetConfig(random_seed=1234, sequence_length=8192, blend=(['/workspace/data/wikitext/wikitext103_gpt2_text_document'], None), blend_per_split=None, split='100,0,0', split_matrix=[(0, 1.0), None, None], num_dataset_builder_threads=1, path_to_cache=None, mmap_bin_files=True, mock=False, tokenizer=<megatron.training.tokenizer.tokenizer._GPT2BPETokenizer object at 0x717d7b1f9070>, mid_level_dataset_surplus=0.005, reset_position_ids=False, reset_attention_mask=False, eod_mask_loss=False, create_attention_mask=True, drop_last_partial_validation_sequence=True, add_extra_token_to_sequence=True, object_storage_cache_path=None)
819
+ INFO:megatron.core.datasets.indexed_dataset:Load the _IndexReader from /workspace/data/wikitext/wikitext103_gpt2_text_document.idx
820
+ INFO:megatron.core.datasets.indexed_dataset: Extract the sequence lengths
821
+ INFO:megatron.core.datasets.indexed_dataset: Extract the sequence pointers
822
+ INFO:megatron.core.datasets.indexed_dataset: Extract the document indices
823
+ INFO:megatron.core.datasets.indexed_dataset:> total number of sequences: 1165029
824
+ INFO:megatron.core.datasets.indexed_dataset:> total number of documents: 1801350
825
+ INFO:megatron.core.datasets.gpt_dataset:Load the GPTDataset train indices
826
+ INFO:megatron.core.datasets.gpt_dataset: Load the document index from b16a908ae7f33ed67d5ec87f6f633afe-GPTDataset-train-document_index.npy
827
+ INFO:megatron.core.datasets.gpt_dataset: Load the sample index from b16a908ae7f33ed67d5ec87f6f633afe-GPTDataset-train-sample_index.npy
828
+ INFO:megatron.core.datasets.gpt_dataset: Load the shuffle index from b16a908ae7f33ed67d5ec87f6f633afe-GPTDataset-train-shuffle_index.npy
829
+ INFO:megatron.core.datasets.gpt_dataset:> total number of samples: 14536
830
+ > finished creating GPT datasets ...
831
+ [after dataloaders are built] datetime: 2026-08-30 07:19:15
832
+ done with setup ...
833
+ training ...
834
+ Overwriting rerun_state_machine.current_iteration from -1 to 0...
835
+ (min, max) time across ranks (ms):
836
+ model-and-optimizer-setup ......................: (3175.25, 3175.39)
837
+ train/valid/test-data-iterators-setup ..........: (74.03, 93.30)
838
+ [before the start of training step] datetime: 2026-08-30 07:19:15
839
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
840
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
841
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
842
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
843
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
844
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
845
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
846
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
847
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
848
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
849
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
850
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
851
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
852
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
853
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
854
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
855
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
856
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
857
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
858
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
859
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
860
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
861
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
862
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
863
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
864
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
865
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
866
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
867
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
868
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
869
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
870
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
871
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
872
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
873
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
874
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
875
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
876
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
877
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
878
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
879
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
880
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
881
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
882
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
883
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
884
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
885
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
886
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
887
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
888
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
889
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
890
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
891
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
892
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
893
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
894
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
895
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
896
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
897
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
898
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
899
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
900
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
901
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
902
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
903
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
904
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
905
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
906
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
907
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
908
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
909
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
910
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
911
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
912
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
913
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
914
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
915
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
916
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
917
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
918
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
919
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
920
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
921
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
922
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
923
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
924
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
925
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
926
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
927
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
928
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
929
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
930
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
931
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
932
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
933
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
934
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
935
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
936
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
937
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
938
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
939
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
940
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
941
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
942
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
943
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
944
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
945
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
946
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
947
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
948
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
949
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
950
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
951
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
952
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
953
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
954
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
955
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
956
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
957
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
958
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
959
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
960
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
961
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
962
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
963
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
964
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
965
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
966
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
967
+ [2026-08-30 07:19:42] iteration 1/ 12 | consumed samples: 16 | elapsed time per iteration (ms): 26292.0 | throughput per GPU (TFLOP/s/GPU): 7.5 | learning rate: 2.000000E-05 | global batch size: 16 | lm loss: 1.103051E+01 | loss scale: 1.0 | grad norm: 50.241 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 28416 | min safe capacity: 1.7344 |
968
+ Number of parameters in transformer block in billions: 3.09
969
+ Number of parameters in embedding layers in billions: 0.41
970
+ Total number of parameters in billions: 3.50
971
+ Number of parameters in most loaded shard in billions: 3.4994
972
+ Theoretical memory footprints: weight and optimizer=25029.43 MB
973
+ [Rank 0] (after 1 iterations) memory (MB) | allocated: 14963.09814453125 | max allocated: 30454.23095703125 | reserved: 31264.0 | max reserved: 31264.0
974
+ [2026-08-30 07:19:42] iteration 2/ 12 | consumed samples: 32 | elapsed time per iteration (ms): 338.7 | throughput per GPU (TFLOP/s/GPU): 585.7 | learning rate: 4.000000E-05 | global batch size: 16 | lm loss: 1.101006E+01 | loss scale: 1.0 | grad norm: 51.122 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 28416 | min safe capacity: 1.7344 |
975
+ [2026-08-30 07:19:42] iteration 3/ 12 | consumed samples: 48 | elapsed time per iteration (ms): 330.1 | throughput per GPU (TFLOP/s/GPU): 601.0 | learning rate: 6.000000E-05 | global batch size: 16 | lm loss: 1.053779E+01 | loss scale: 1.0 | grad norm: 249.887 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 28416 | min safe capacity: 1.7344 |
976
+ Capture CUDA graph for training!!!
977
+ CUDA graph capture done!!!
978
+ [2026-08-30 07:19:43] iteration 4/ 12 | consumed samples: 64 | elapsed time per iteration (ms): 1123.3 | throughput per GPU (TFLOP/s/GPU): 176.6 | learning rate: 8.000000E-05 | global batch size: 16 | lm loss: 1.320004E+01 | loss scale: 1.0 | grad norm: 35.691 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 28416 | min safe capacity: 1.7344 |
979
+ [2026-08-30 07:19:44] iteration 5/ 12 | consumed samples: 80 | elapsed time per iteration (ms): 292.9 | throughput per GPU (TFLOP/s/GPU): 677.4 | learning rate: 1.000000E-04 | global batch size: 16 | lm loss: 1.052084E+01 | loss scale: 1.0 | grad norm: 17.496 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 28416 | min safe capacity: 1.7344 |
980
+ [2026-08-30 07:19:44] iteration 6/ 12 | consumed samples: 96 | elapsed time per iteration (ms): 295.1 | throughput per GPU (TFLOP/s/GPU): 672.2 | learning rate: 7.750000E-05 | global batch size: 16 | lm loss: 1.027154E+01 | loss scale: 1.0 | grad norm: 9.627 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 28416 | min safe capacity: 1.7344 |
981
+ [2026-08-30 07:19:44] iteration 7/ 12 | consumed samples: 112 | elapsed time per iteration (ms): 290.7 | throughput per GPU (TFLOP/s/GPU): 682.3 | learning rate: 3.250000E-05 | global batch size: 16 | lm loss: 9.463491E+00 | loss scale: 1.0 | grad norm: 7.590 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 28416 | min safe capacity: 1.7344 |
982
+ [2026-08-30 07:19:45] iteration 8/ 12 | consumed samples: 128 | elapsed time per iteration (ms): 291.6 | throughput per GPU (TFLOP/s/GPU): 680.3 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.751801E+00 | loss scale: 1.0 | grad norm: 6.024 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 28416 | min safe capacity: 1.7344 |
983
+ [2026-08-30 07:19:45] iteration 9/ 12 | consumed samples: 144 | elapsed time per iteration (ms): 291.6 | throughput per GPU (TFLOP/s/GPU): 680.4 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.540117E+00 | loss scale: 1.0 | grad norm: 7.212 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 28416 | min safe capacity: 1.7344 |
984
+ [2026-08-30 07:19:45] iteration 10/ 12 | consumed samples: 160 | elapsed time per iteration (ms): 292.1 | throughput per GPU (TFLOP/s/GPU): 679.2 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.444377E+00 | loss scale: 1.0 | grad norm: 6.087 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 28416 | min safe capacity: 1.7344 |
985
+ [2026-08-30 07:19:46] iteration 11/ 12 | consumed samples: 176 | elapsed time per iteration (ms): 295.4 | throughput per GPU (TFLOP/s/GPU): 671.6 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.379338E+00 | loss scale: 1.0 | grad norm: 4.765 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 28416 | min safe capacity: 1.7344 |
986
+ [2026-08-30 07:19:46] iteration 12/ 12 | consumed samples: 192 | elapsed time per iteration (ms): 292.8 | throughput per GPU (TFLOP/s/GPU): 677.4 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.274484E+00 | loss scale: 1.0 | grad norm: 4.067 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 28416 | min safe capacity: 1.7344 |
987
+ [after training is done] datetime: 2026-08-30 07:19:46
988
+ W0830 07:19:47.299000 739083 torch/distributed/elastic/agent/server/api.py:719] Received 15 death signal, shutting down workers
989
+ W0830 07:19:47.300000 739083 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 739171 closing signal SIGTERM
990
+ W0830 07:19:47.301000 739083 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 739172 closing signal SIGTERM
991
+ W0830 07:19:47.302000 739083 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 739173 closing signal SIGTERM
992
+ W0830 07:19:47.305000 739083 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 739174 closing signal SIGTERM
993
+ W0830 07:19:47.306000 739083 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 739175 closing signal SIGTERM
994
+ W0830 07:19:47.307000 739083 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 739176 closing signal SIGTERM
995
+ W0830 07:19:47.308000 739083 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 739177 closing signal SIGTERM
996
+ W0830 07:19:47.310000 739083 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 739178 closing signal SIGTERM
997
+ Traceback (most recent call last):
998
+ File "/usr/local/bin/torchrun", line 8, in <module>
999
+ sys.exit(main())
1000
+ ^^^^^^
1001
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/multiprocessing/errors/__init__.py", line 355, in wrapper
1002
+ return f(*args, **kwargs)
1003
+ ^^^^^^^^^^^^^^^^^^
1004
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/run.py", line 892, in main
1005
+ run(args)
1006
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/run.py", line 883, in run
1007
+ elastic_launch(
1008
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/launcher/api.py", line 139, in __call__
1009
+ return launch_agent(self._config, self._entrypoint, list(args))
1010
+ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1011
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/launcher/api.py", line 261, in launch_agent
1012
+ result = agent.run()
1013
+ ^^^^^^^^^^^
1014
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/metrics/api.py", line 138, in wrapper
1015
+ result = f(*args, **kwargs)
1016
+ ^^^^^^^^^^^^^^^^^^
1017
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/agent/server/api.py", line 711, in run
1018
+ result = self._invoke_run(role)
1019
+ ^^^^^^^^^^^^^^^^^^^^^^
1020
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/agent/server/api.py", line 870, in _invoke_run
1021
+ time.sleep(monitor_interval)
1022
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/multiprocessing/api.py", line 84, in _terminate_process_handler
1023
+ raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval)
1024
+ torch.distributed.elastic.multiprocessing.api.SignalException: Process 739083 got signal: 15
tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq8192__gbs16__cap2p40625/rep2/mem_samples.csv ADDED
@@ -0,0 +1,168 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 0, 0
2
+ 1, 0
3
+ 2, 0
4
+ 3, 0
5
+ 4, 0
6
+ 5, 0
7
+ 6, 0
8
+ 7, 0
9
+ 0, 0
10
+ 1, 0
11
+ 2, 0
12
+ 3, 0
13
+ 4, 0
14
+ 5, 0
15
+ 6, 0
16
+ 7, 0
17
+ 0, 1
18
+ 1, 1
19
+ 2, 1
20
+ 3, 1
21
+ 4, 1
22
+ 5, 1
23
+ 6, 1
24
+ 7, 1
25
+ 0, 4
26
+ 1, 4
27
+ 2, 4
28
+ 3, 4
29
+ 4, 1
30
+ 5, 1
31
+ 6, 1
32
+ 7, 1
33
+ 0, 4
34
+ 1, 4
35
+ 2, 4
36
+ 3, 4
37
+ 4, 4
38
+ 5, 4
39
+ 6, 4
40
+ 7, 4
41
+ 0, 821
42
+ 1, 819
43
+ 2, 819
44
+ 3, 819
45
+ 4, 819
46
+ 5, 821
47
+ 6, 819
48
+ 7, 819
49
+ 0, 1537
50
+ 1, 1537
51
+ 2, 1537
52
+ 3, 1537
53
+ 4, 1537
54
+ 5, 1537
55
+ 6, 1537
56
+ 7, 1537
57
+ 0, 2679
58
+ 1, 2679
59
+ 2, 2679
60
+ 3, 2679
61
+ 4, 2679
62
+ 5, 2679
63
+ 6, 2679
64
+ 7, 2679
65
+ 0, 2903
66
+ 1, 2903
67
+ 2, 2903
68
+ 3, 2903
69
+ 4, 2903
70
+ 5, 2903
71
+ 6, 2903
72
+ 7, 2903
73
+ 0, 14589
74
+ 1, 14587
75
+ 2, 14587
76
+ 3, 14587
77
+ 4, 14587
78
+ 5, 14587
79
+ 6, 14587
80
+ 7, 14587
81
+ 0, 15127
82
+ 1, 15125
83
+ 2, 15125
84
+ 3, 15125
85
+ 4, 15125
86
+ 5, 15125
87
+ 6, 15125
88
+ 7, 15125
89
+ 0, 16685
90
+ 1, 16683
91
+ 2, 16683
92
+ 3, 16683
93
+ 4, 16683
94
+ 5, 16683
95
+ 6, 16683
96
+ 7, 16683
97
+ 0, 16685
98
+ 1, 16683
99
+ 2, 16683
100
+ 3, 16683
101
+ 4, 16683
102
+ 5, 16683
103
+ 6, 16683
104
+ 7, 16683
105
+ 0, 16993
106
+ 1, 16991
107
+ 2, 16991
108
+ 3, 16991
109
+ 4, 16991
110
+ 5, 16991
111
+ 6, 16991
112
+ 7, 16991
113
+ 0, 17959
114
+ 1, 17957
115
+ 2, 17957
116
+ 3, 17957
117
+ 4, 17957
118
+ 5, 17957
119
+ 6, 17957
120
+ 7, 17957
121
+ 0, 35699
122
+ 1, 35697
123
+ 2, 35697
124
+ 3, 35697
125
+ 4, 35659
126
+ 5, 35697
127
+ 6, 35697
128
+ 7, 35697
129
+ 0, 35699
130
+ 1, 35697
131
+ 2, 35697
132
+ 3, 35697
133
+ 4, 35659
134
+ 5, 35697
135
+ 6, 35697
136
+ 7, 35697
137
+ 0, 35699
138
+ 1, 35697
139
+ 2, 35697
140
+ 3, 35697
141
+ 4, 35659
142
+ 5, 35697
143
+ 6, 35697
144
+ 7, 35697
145
+ 0, 37523
146
+ 1, 37541
147
+ 2, 37501
148
+ 3, 37523
149
+ 4, 37483
150
+ 5, 37481
151
+ 6, 37515
152
+ 7, 37517
153
+ 0, 38501
154
+ 1, 38511
155
+ 2, 38493
156
+ 3, 38529
157
+ 4, 38497
158
+ 5, 38517
159
+ 6, 38529
160
+ 7, 38497
161
+ 0, 39647
162
+ 1, 39639
163
+ 2, 39653
164
+ 3, 39641
165
+ 4, 39651
166
+ 5, 39689
167
+ 6, 39641
168
+ 7, 39707
tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq8192__gbs16__cap2p40625/rep2/run.env ADDED
@@ -0,0 +1,70 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ backend=ultraep
2
+ config=PONLY
3
+ family=tierA
4
+ repo=/workspace/Megatron-UltraEP
5
+ branch=ultraep-bracing-regions-through-kernels
6
+ commit=03d57ef7
7
+ num_layers=16
8
+ hidden=4096
9
+ ffn_hidden=14336
10
+ moe_ffn_hidden=1536
11
+ num_experts=8
12
+ topk=2
13
+ experts_per_rank=1
14
+ expert_params=18874368
15
+ heads=32
16
+ kv_groups=8
17
+ kv_channels=
18
+ qk_layernorm=0
19
+ shared_ffn=0
20
+ shared_expert_overlap=0
21
+ moe_token_dispatcher=flex
22
+ mla=0
23
+ dense_first_layers=0
24
+ moe_layers=16
25
+ rotary_base=1000000
26
+ rotary_scaling=
27
+ qkv_bias=0
28
+ router_balancing=aux_loss
29
+ aux_loss_coeff=1e-2
30
+ router_pre_softmax=0
31
+ router_fusion=0
32
+ router_dtype=fp32
33
+ device_initiated_grouped_gemm=1
34
+ norm_epsilon=1e-5
35
+ seq_length=8192
36
+ micro_batch_size=1
37
+ global_batch_size=16
38
+ num_microbatches=2
39
+ assignments_per_rank=16384
40
+ mean_tokens_per_expert=16384
41
+ slot_policy=one
42
+ slots_per_rank=1
43
+ capacity_mode=static
44
+ capacity=2.40625
45
+ cuda_graph=1
46
+ probe_mode=0
47
+ async_grad_reduce=0
48
+ branch_role=timing
49
+ memory_dump_mode=none
50
+ memory_snapshot_path=
51
+ cuda_graph_warmup_steps=3
52
+ recompute=none
53
+ force_balance=0
54
+ train_iters=12
55
+ ep_size=8
56
+ echo_dispatch_overlap=1
57
+ echo_dispatch_recompute=0
58
+ ultraep_quota_min_tokens=2048
59
+ ultraep_weight_sync_join=
60
+ profile_enable=0
61
+ kernel_trace=0
62
+ fp8_mode=mxfp8
63
+ fp8_recipe=mxfp8
64
+ training_seed=1234
65
+ learning_rate=1e-4
66
+ min_learning_rate=1.0e-5
67
+ lr_warmup_iters=5
68
+ state_dump_enable=0
69
+ state_dump_dir=
70
+ moonep_assert_no_pending_reduce=0
tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq8192__gbs16__cap2p40625/rep2/train.log ADDED
@@ -0,0 +1,977 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ==============================================================
2
+ PONLY / ultraep (slots/rank=1 policy=one)
3
+ E=8 K=2 H'=1536 m=1
4
+ L=16 seq=8192 gbs=16 (2 microbatches)
5
+ cap_mode=static cap=2.40625 cg=1 recompute=none
6
+ memory_dump=none snapshots=/workspace/tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq8192__gbs16__cap2p40625/rep2/memory_snapshots/snapshot.pickle
7
+ repo=/workspace/Megatron-UltraEP @ 03d57ef7
8
+ ==============================================================
9
+ W0830 07:19:53.798000 741931 torch/distributed/run.py:766]
10
+ W0830 07:19:53.798000 741931 torch/distributed/run.py:766] *****************************************
11
+ W0830 07:19:53.798000 741931 torch/distributed/run.py:766] Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed.
12
+ W0830 07:19:53.798000 741931 torch/distributed/run.py:766] *****************************************
13
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
14
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
15
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
16
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
17
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
18
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
19
+ WARNING: one_logger package is required to enable e2e metrics tracking. please go to https://confluence.nvidia.com/display/MLWFO/Package+Repositories for details to install it
20
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
21
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
22
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
23
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
24
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
25
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
26
+ using world size: 8, data-parallel size: 8, context-parallel size: 1, hierarchical context-parallel sizes: None, tensor-model-parallel size: 1, pipeline-model-parallel size: 1
27
+ WARNING: overriding default arguments for tokenizer_type:GPT2BPETokenizer with tokenizer_type:GPT2BPETokenizer
28
+ Number of virtual stages per pipeline stage: None
29
+ accumulate and all-reduce gradients in fp32 for bfloat16 data type.
30
+ using torch.bfloat16 for parameters ...
31
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
32
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
33
+ ------------------------ arguments ------------------------
34
+ account_for_embedding_in_pipeline_split ......... False
35
+ account_for_loss_in_pipeline_split .............. False
36
+ accumulate_allreduce_grads_in_fp32 .............. True
37
+ adam_beta1 ...................................... 0.9
38
+ adam_beta2 ...................................... 0.999
39
+ adam_eps ........................................ 1e-08
40
+ add_bias_linear ................................. False
41
+ add_position_embedding .......................... False
42
+ add_qkv_bias .................................... False
43
+ adlr_autoresume ................................. False
44
+ adlr_autoresume_interval ........................ 1000
45
+ align_grad_reduce ............................... True
46
+ align_param_gather .............................. False
47
+ app_tag_run_name ................................ None
48
+ app_tag_run_version ............................. 0.0.0
49
+ apply_layernorm_1p .............................. False
50
+ apply_query_key_layer_scaling ................... False
51
+ apply_residual_connection_post_layernorm ........ False
52
+ apply_rope_fusion ............................... True
53
+ async_save ...................................... None
54
+ async_tensor_model_parallel_allreduce ........... True
55
+ attention_backend ............................... AttnBackend.auto
56
+ attention_dropout ............................... 0.0
57
+ attention_softmax_in_fp32 ....................... False
58
+ auto_detect_ckpt_format ......................... False
59
+ barrier_with_L1_time ............................ True
60
+ bert_binary_head ................................ True
61
+ bert_embedder_type .............................. megatron
62
+ bert_load ....................................... None
63
+ bf16 ............................................ True
64
+ bias_dropout_fusion ............................. True
65
+ bias_gelu_fusion ................................ False
66
+ bias_swiglu_fusion .............................. True
67
+ biencoder_projection_dim ........................ 0
68
+ biencoder_shared_query_context_model ............ False
69
+ block_data_path ................................. None
70
+ calc_ft_timeouts ................................ False
71
+ calculate_per_token_loss ........................ False
72
+ check_for_large_grads ........................... False
73
+ check_for_nan_in_loss_and_grad .................. False
74
+ check_for_spiky_loss ............................ False
75
+ check_weight_hash_across_dp_replicas_interval ... None
76
+ ckpt_assume_constant_structure .................. False
77
+ ckpt_convert_format ............................. None
78
+ ckpt_convert_save ............................... None
79
+ ckpt_convert_update_legacy_dist_opt_format ...... False
80
+ ckpt_format ..................................... torch_dist
81
+ ckpt_fully_parallel_load ........................ False
82
+ ckpt_fully_parallel_save ........................ True
83
+ ckpt_fully_parallel_save_deprecated ............. False
84
+ ckpt_step ....................................... None
85
+ classes_fraction ................................ 1.0
86
+ clip_grad ....................................... 1.0
87
+ clone_scatter_output_in_embedding ............... True
88
+ config_logger_dir ...............................
89
+ consumed_train_samples .......................... 0
90
+ consumed_valid_samples .......................... 0
91
+ context_parallel_size ........................... 1
92
+ cp_comm_type .................................... ['p2p']
93
+ create_attention_mask_in_dataloader ............. True
94
+ cross_entropy_fusion_impl ....................... native
95
+ cross_entropy_loss_fusion ....................... False
96
+ cuda_graph_scope ................................ full_iteration
97
+ cuda_graph_warmup_steps ......................... 3
98
+ data_args_path .................................. None
99
+ data_cache_path ................................. None
100
+ data_parallel_random_init ....................... False
101
+ data_parallel_sharding_strategy ................. no_shard
102
+ data_parallel_size .............................. 8
103
+ data_path ....................................... ['/workspace/data/wikitext/wikitext103_gpt2_text_document']
104
+ data_per_class_fraction ......................... 1.0
105
+ data_sharding ................................... True
106
+ dataloader_type ................................. single
107
+ ddp_average_in_collective ....................... False
108
+ ddp_bucket_size ................................. None
109
+ ddp_num_buckets ................................. None
110
+ ddp_pad_buckets_for_high_nccl_busbw ............. False
111
+ decoder_first_pipeline_num_layers ............... None
112
+ decoder_last_pipeline_num_layers ................ None
113
+ decoder_num_layers .............................. None
114
+ decoder_seq_length .............................. None
115
+ decoupled_lr .................................... None
116
+ decoupled_min_lr ................................ None
117
+ decrease_batch_size_if_needed ................... False
118
+ defer_embedding_wgrad_compute ................... False
119
+ delay_wgrad_compute ............................. False
120
+ deprecated_use_mcore_models ..................... True
121
+ deterministic_mode .............................. False
122
+ dino_bottleneck_size ............................ 256
123
+ dino_freeze_last_layer .......................... 1
124
+ dino_head_hidden_size ........................... 2048
125
+ dino_local_crops_number ......................... 10
126
+ dino_local_img_size ............................. 96
127
+ dino_norm_last_layer ............................ False
128
+ dino_teacher_temp ............................... 0.07
129
+ dino_warmup_teacher_temp ........................ 0.04
130
+ dino_warmup_teacher_temp_epochs ................. 30
131
+ disable_bf16_reduced_precision_matmul ........... False
132
+ disable_mamba_mem_eff_path ...................... False
133
+ disable_straggler_on_startup .................... False
134
+ dist_ckpt_format_deprecated ..................... None
135
+ dist_ckpt_strictness ............................ assume_ok_unexpected
136
+ distribute_saved_activations .................... False
137
+ distributed_backend ............................. nccl
138
+ distributed_timeout_minutes ..................... 10
139
+ embedding_init_method_std ....................... None
140
+ embedding_path .................................. None
141
+ empty_unused_memory_level ....................... 0
142
+ enable_cuda_graph ............................... True
143
+ enable_experimental ............................. False
144
+ enable_ft_package ............................... False
145
+ enable_gloo_process_groups ...................... True
146
+ enable_msc ...................................... True
147
+ enable_one_logger ............................... True
148
+ encoder_num_layers .............................. 16
149
+ encoder_seq_length .............................. 8192
150
+ end_weight_decay ................................ 0.1
151
+ eod_mask_loss ................................... False
152
+ error_injection_rate ............................ 0
153
+ error_injection_type ............................ transient_error
154
+ eval_interval ................................... 1000000
155
+ eval_iters ...................................... 0
156
+ evidence_data_path .............................. None
157
+ exit_duration_in_mins ........................... None
158
+ exit_interval ................................... None
159
+ exit_on_missing_checkpoint ...................... False
160
+ exit_signal_handler ............................. False
161
+ exp_avg_dtype ................................... torch.float32
162
+ exp_avg_sq_dtype ................................ torch.float32
163
+ expert_model_parallel_size ...................... 8
164
+ expert_tensor_parallel_size ..................... 1
165
+ export_force_local_attention .................... False
166
+ export_kd_cfg ................................... None
167
+ export_kd_teacher_ckpt_format ................... None
168
+ export_kd_teacher_load .......................... None
169
+ export_kv_cache_quant ........................... False
170
+ export_legacy_megatron .......................... False
171
+ export_model_type ............................... GPTModel
172
+ export_moe_apply_probs_on_input ................. False
173
+ export_qk_l2_norm ............................... False
174
+ export_quant_cfg ................................ None
175
+ export_real_quant_cfg ........................... None
176
+ export_te_mcore_model ........................... False
177
+ external_cuda_graph ............................. False
178
+ ffn_hidden_size ................................. 14336
179
+ finetune ........................................ False
180
+ finetune_data_split ............................. train
181
+ finetune_hf_dataset ............................. None
182
+ first_last_layers_bf16 .......................... False
183
+ flash_decode .................................... False
184
+ fp16 ............................................ False
185
+ fp16_lm_cross_entropy ........................... False
186
+ fp32_residual_connection ........................ False
187
+ fp8 ............................................. e4m3
188
+ fp8_amax_compute_algo ........................... most_recent
189
+ fp8_amax_history_len ............................ 1
190
+ fp8_interval .................................... 1
191
+ fp8_margin ...................................... 0
192
+ fp8_param_gather ................................ False
193
+ fp8_recipe ...................................... mxfp8
194
+ fp8_wgrad ....................................... True
195
+ fsdp_double_buffer .............................. False
196
+ global_batch_size ............................... 16
197
+ grad_reduce_in_bf16 ............................. False
198
+ gradient_accumulation_fusion .................... True
199
+ gradient_reduce_div_fusion ...................... True
200
+ group_query_attention ........................... True
201
+ head_lr_mult .................................... 1.0
202
+ heterogeneous_layers_config_encoded_json ........ None
203
+ heterogeneous_layers_config_path ................ None
204
+ hidden_dropout .................................. 0.0
205
+ hidden_size ..................................... 4096
206
+ hierarchical_context_parallel_sizes ............. None
207
+ high_priority_stream_groups ..................... []
208
+ hybrid_attention_ratio .......................... 0.0
209
+ hybrid_mlp_ratio ................................ 0.0
210
+ hybrid_override_pattern ......................... None
211
+ hysteresis ...................................... 2
212
+ ict_head_size ................................... None
213
+ ict_load ........................................ None
214
+ img_h ........................................... 224
215
+ img_w ........................................... 224
216
+ indexer_batch_size .............................. 128
217
+ indexer_log_interval ............................ 1000
218
+ inference_batch_times_seqlen_threshold .......... -1
219
+ inference_dynamic_batching ...................... False
220
+ inference_dynamic_batching_buffer_guaranteed_fraction 0.2
221
+ inference_dynamic_batching_buffer_overflow_factor None
222
+ inference_dynamic_batching_buffer_size_gb ....... 40.0
223
+ inference_dynamic_batching_chunk_size ........... 256
224
+ inference_dynamic_batching_max_requests_override None
225
+ inference_dynamic_batching_max_tokens_override .. None
226
+ inference_dynamic_batching_num_cuda_graphs ...... 16
227
+ inference_max_batch_size ........................ 8
228
+ inference_max_seq_length ........................ 2560
229
+ inference_rng_tracker ........................... False
230
+ init_method_std ................................. 0.01
231
+ init_method_xavier_uniform ...................... False
232
+ init_model_with_meta_device ..................... False
233
+ initial_loss_scale .............................. 4294967296
234
+ inprocess_active_world_size ..................... 8
235
+ inprocess_barrier_timeout ....................... 120
236
+ inprocess_completion_timeout .................... 120
237
+ inprocess_empty_cuda_cache ...................... False
238
+ inprocess_granularity ........................... node
239
+ inprocess_hard_timeout .......................... 90
240
+ inprocess_heartbeat_interval .................... 30
241
+ inprocess_heartbeat_timeout ..................... 60
242
+ inprocess_last_call_wait ........................ 1
243
+ inprocess_max_iterations ........................ None
244
+ inprocess_monitor_process_interval .............. 1.0
245
+ inprocess_monitor_thread_interval ............... 1.0
246
+ inprocess_progress_watchdog_interval ............ 1.0
247
+ inprocess_restart ............................... False
248
+ inprocess_soft_timeout .......................... 60
249
+ inprocess_termination_grace_time ................ 1
250
+ is_hybrid_model ................................. False
251
+ iter_per_epoch .................................. 1250
252
+ iterations_to_skip .............................. []
253
+ keep_fp8_transpose_cache_when_using_custom_fsdp . False
254
+ kitchen_config_file ............................. None
255
+ kitchen_recipe_number ........................... None
256
+ kv_channels ..................................... 128
257
+ kv_lora_rank .................................... 32
258
+ lazy_mpu_init ................................... None
259
+ load ............................................ None
260
+ load_main_params_from_ckpt ...................... None
261
+ load_model_opt_format ........................... False
262
+ local_rank ...................................... 0
263
+ log_energy ...................................... False
264
+ log_interval .................................... 1
265
+ log_loss_scale_to_tensorboard ................... True
266
+ log_memory_to_tensorboard ....................... False
267
+ log_num_zeros_in_grad ........................... False
268
+ log_params_norm ................................. False
269
+ log_progress .................................... False
270
+ log_straggler ................................... False
271
+ log_throughput .................................. True
272
+ log_timers_to_tensorboard ....................... False
273
+ log_validation_ppl_to_tensorboard ............... False
274
+ log_world_size_to_tensorboard ................... False
275
+ logging_level ................................... None
276
+ loss_scale ...................................... None
277
+ loss_scale_window ............................... 1000
278
+ lr .............................................. 0.0001
279
+ lr_decay_iters .................................. 8
280
+ lr_decay_samples ................................ None
281
+ lr_decay_style .................................. cosine
282
+ lr_warmup_fraction .............................. None
283
+ lr_warmup_init .................................. 0.0
284
+ lr_warmup_iters ................................. 5
285
+ lr_warmup_samples ............................... 0
286
+ lr_wsd_decay_iters .............................. None
287
+ lr_wsd_decay_samples ............................ None
288
+ lr_wsd_decay_style .............................. exponential
289
+ main_grads_dtype ................................ torch.float32
290
+ main_params_dtype ............................... torch.float32
291
+ make_vocab_size_divisible_by .................... 128
292
+ mamba_head_dim .................................. 64
293
+ mamba_num_groups ................................ 8
294
+ mamba_num_heads ................................. None
295
+ mamba_state_dim ................................. 128
296
+ manual_gc ....................................... False
297
+ manual_gc_eval .................................. True
298
+ manual_gc_interval .............................. 0
299
+ mask_factor ..................................... 1.0
300
+ mask_prob ....................................... 0.15
301
+ mask_type ....................................... random
302
+ masked_softmax_fusion ........................... False
303
+ max_position_embeddings ......................... 32768
304
+ max_tokens_to_oom ............................... 12000
305
+ memory_snapshot_path ............................ snapshot.pickle
306
+ merge_file ...................................... /workspace/data/wikitext/gpt2-merges.txt
307
+ micro_batch_size ................................ 1
308
+ microbatch_group_size_per_vp_stage .............. None
309
+ mid_level_dataset_surplus ....................... 0.005
310
+ min_loss_scale .................................. 1.0
311
+ min_lr .......................................... 1e-05
312
+ mlp_chunks_for_prefill .......................... 1
313
+ mmap_bin_files .................................. True
314
+ mock_data ....................................... False
315
+ moe_apply_probs_on_input ........................ False
316
+ moe_aux_loss_coeff .............................. 0.01
317
+ moe_capacity_monitor ............................ True
318
+ moe_deepep_num_sms .............................. 20
319
+ moe_echo_enable_random_offloading ............... False
320
+ moe_echo_expert_dispatch_overlap ................ False
321
+ moe_echo_expert_dispatcher_type ................. hybridep
322
+ moe_echo_recompute_expert_dispatch .............. False
323
+ moe_enable_deepep ............................... False
324
+ moe_enable_echo ................................. False
325
+ moe_enable_hybridep ............................. True
326
+ moe_enable_ultraep .............................. True
327
+ moe_expert_capacity_factor ...................... None
328
+ moe_extended_tp ................................. False
329
+ moe_ffn_hidden_size ............................. 1536
330
+ moe_grouped_gemm ................................ True
331
+ moe_hybridep_num_sms ............................ 16
332
+ moe_input_jitter_eps ............................ None
333
+ moe_layer_freq .................................. 1
334
+ moe_layer_recompute ............................. False
335
+ moe_num_echo_experts ............................ None
336
+ moe_num_redundant_experts_per_rank .............. 1
337
+ moe_pad_expert_input_to_capacity ................ False
338
+ moe_per_layer_logging ........................... False
339
+ moe_permute_fusion .............................. False
340
+ moe_received_token_capacity ..................... 2.40625
341
+ moe_router_bias_update_rate ..................... 0.001
342
+ moe_router_dtype ................................ fp32
343
+ moe_router_enable_expert_bias ................... False
344
+ moe_router_force_load_balancing ................. False
345
+ moe_router_group_topk ........................... None
346
+ moe_router_load_balancing_type .................. aux_loss
347
+ moe_router_num_groups ........................... None
348
+ moe_router_padding_for_fp8 ...................... False
349
+ moe_router_pre_softmax .......................... False
350
+ moe_router_score_function ....................... softmax
351
+ moe_router_topk ................................. 2
352
+ moe_router_topk_scaling_factor .................. None
353
+ moe_shared_expert_intermediate_size ............. None
354
+ moe_shared_expert_overlap ....................... False
355
+ moe_token_dispatcher_type ....................... flex
356
+ moe_token_drop_policy ........................... probs
357
+ moe_ultraep_max_microbatches .................... None
358
+ moe_ultraep_overlap_weight_sync ................. True
359
+ moe_ultraep_weight_sync_join .................... before_dispatch
360
+ moe_upcycling_granularity ....................... 1
361
+ moe_use_device_initiated_grouped_gemm ........... True
362
+ moe_use_legacy_grouped_gemm ..................... False
363
+ moe_use_upcycling ............................... False
364
+ moe_z_loss_coeff ................................ None
365
+ mrope_section ................................... None
366
+ mscale .......................................... 1.0
367
+ mscale_all_dim .................................. 0.0
368
+ mtp_loss_scaling_factor ......................... 0.1
369
+ mtp_num_layers .................................. None
370
+ multi_latent_attention .......................... False
371
+ nccl_all_reduce_for_prefill ..................... False
372
+ nccl_communicator_config_path ................... None
373
+ nccl_ub ......................................... False
374
+ no_load_optim ................................... None
375
+ no_load_rng ..................................... None
376
+ no_persist_layer_norm ........................... False
377
+ no_rope_freq .................................... None
378
+ no_save_optim ................................... None
379
+ no_save_rng ..................................... None
380
+ non_persistent_ckpt_type ........................ None
381
+ non_persistent_global_ckpt_dir .................. None
382
+ non_persistent_local_ckpt_algo .................. fully_parallel
383
+ non_persistent_local_ckpt_dir ................... None
384
+ non_persistent_save_interval .................... None
385
+ norm_epsilon .................................... 1e-05
386
+ normalization ................................... RMSNorm
387
+ num_attention_heads ............................. 32
388
+ num_channels .................................... 3
389
+ num_classes ..................................... 1000
390
+ num_dataset_builder_threads ..................... 1
391
+ num_distributed_optimizer_instances ............. 1
392
+ num_experts ..................................... 8
393
+ num_layers ...................................... 16
394
+ num_layers_at_end_in_bf16 ....................... 1
395
+ num_layers_at_start_in_bf16 ..................... 1
396
+ num_layers_per_virtual_pipeline_stage ........... None
397
+ num_query_groups ................................ 8
398
+ num_virtual_stages_per_pipeline_rank ............ None
399
+ num_workers ..................................... 2
400
+ object_storage_cache_path ....................... None
401
+ one_logger_async ................................ False
402
+ one_logger_project .............................. megatron-lm
403
+ one_logger_run_name ............................. None
404
+ onnx_safe ....................................... None
405
+ openai_gelu ..................................... False
406
+ optimizer ....................................... adam
407
+ optimizer_cpu_offload ........................... False
408
+ optimizer_offload_fraction ...................... 1.0
409
+ output_bert_embeddings .......................... False
410
+ overlap_cpu_optimizer_d2h_h2d ................... False
411
+ overlap_grad_reduce ............................. True
412
+ overlap_moe_expert_parallel_comm ................ False
413
+ overlap_p2p_comm ................................ False
414
+ overlap_p2p_comm_warmup_flush ................... False
415
+ overlap_param_gather ............................ True
416
+ overlap_param_gather_with_optimizer_step ........ False
417
+ override_opt_param_scheduler .................... False
418
+ params_dtype .................................... torch.bfloat16
419
+ patch_dim ....................................... 16
420
+ per_split_data_args_path ........................ None
421
+ perform_initialization .......................... True
422
+ pin_cpu_grads ................................... True
423
+ pin_cpu_params .................................. True
424
+ pipeline_model_parallel_comm_backend ............ None
425
+ pipeline_model_parallel_layout .................. None
426
+ pipeline_model_parallel_size .................... 1
427
+ position_embedding_type ......................... rope
428
+ pretrained_checkpoint ........................... None
429
+ profile ......................................... False
430
+ profile_ranks ................................... [0]
431
+ profile_step_end ................................ 12
432
+ profile_step_start .............................. 10
433
+ q_lora_rank ..................................... None
434
+ qk_head_dim ..................................... 128
435
+ qk_l2_norm ...................................... False
436
+ qk_layernorm .................................... False
437
+ qk_pos_emb_head_dim ............................. 64
438
+ query_in_block_prob ............................. 0.1
439
+ rampup_batch_size ............................... None
440
+ rank ............................................ 0
441
+ recompute_granularity ........................... None
442
+ recompute_method ................................ None
443
+ recompute_modules ............................... None
444
+ recompute_num_layers ............................ None
445
+ record_memory_history ........................... False
446
+ relative_attention_max_distance ................. 128
447
+ relative_attention_num_buckets .................. 32
448
+ replication ..................................... False
449
+ replication_factor .............................. 2
450
+ replication_jump ................................ None
451
+ rerun_mode ...................................... validate_results
452
+ reset_attention_mask ............................ False
453
+ reset_position_ids .............................. False
454
+ result_rejected_tracker_filename ................ None
455
+ retriever_report_topk_accuracies ................ []
456
+ retriever_score_scaling ......................... False
457
+ retriever_seq_length ............................ 256
458
+ retro_add_retriever ............................. False
459
+ retro_attention_gate ............................ 1
460
+ retro_cyclic_train_iters ........................ None
461
+ retro_encoder_attention_dropout ................. 0.1
462
+ retro_encoder_hidden_dropout .................... 0.1
463
+ retro_encoder_layers ............................ 2
464
+ retro_num_neighbors ............................. 2
465
+ retro_num_retrieved_chunks ...................... 2
466
+ retro_project_dir ............................... None
467
+ retro_verify_neighbor_count ..................... True
468
+ reuse_grad_buf_for_mxfp8_param_ag ............... False
469
+ rope_scaling_factor ............................. 8.0
470
+ rope_type ....................................... None
471
+ rotary_base ..................................... 1000000
472
+ rotary_interleaved .............................. False
473
+ rotary_percent .................................. 1.0
474
+ rotary_scaling_factor ........................... 1.0
475
+ rotary_seq_len_interpolation_factor ............. None
476
+ run_workload_inspector_server ................... False
477
+ sample_rate ..................................... 1.0
478
+ save ............................................ None
479
+ save_interval ................................... None
480
+ save_retain_interval ............................ None
481
+ scatter_gather_tensors_in_pipeline .............. True
482
+ seed ............................................ 1234
483
+ seq_length ...................................... 8192
484
+ sequence_parallel ............................... False
485
+ sft ............................................. False
486
+ sft_tokenizer_prompt_format ..................... nemotron-h-aligned
487
+ sgd_momentum .................................... 0.9
488
+ short_seq_prob .................................. 0.1
489
+ skip_train ...................................... False
490
+ skipped_train_samples ........................... 0
491
+ spec ............................................ None
492
+ split ........................................... 100,0,0
493
+ squared_relu .................................... False
494
+ start_weight_decay .............................. 0.1
495
+ straggler_ctrlr_port ............................ 65535
496
+ straggler_minmax_count .......................... 1
497
+ suggested_communication_unit_size ............... None
498
+ swiglu .......................................... True
499
+ swin_backbone_type .............................. tiny
500
+ symmetric_ar_type ............................... None
501
+ te_rng_tracker .................................. True
502
+ tensor_model_parallel_size ...................... 1
503
+ tensorboard_dir ................................. None
504
+ tensorboard_log_interval ........................ 1
505
+ tensorboard_queue_size .......................... 1000
506
+ test_data_path .................................. None
507
+ test_mode ....................................... False
508
+ tiktoken_num_special_tokens ..................... 1000
509
+ tiktoken_pattern ................................ None
510
+ tiktoken_special_tokens ......................... None
511
+ timing_log_level ................................ 0
512
+ timing_log_option ............................... minmax
513
+ titles_data_path ................................ None
514
+ tokenizer_model ................................. None
515
+ tokenizer_type .................................. GPT2BPETokenizer
516
+ torch_fsdp2_reshard_after_forward ............... True
517
+ tp_comm_bootstrap_backend ....................... nccl
518
+ tp_comm_bulk_dgrad .............................. True
519
+ tp_comm_bulk_wgrad .............................. True
520
+ tp_comm_overlap ................................. False
521
+ tp_comm_overlap_ag .............................. True
522
+ tp_comm_overlap_cfg ............................. None
523
+ tp_comm_overlap_rs .............................. True
524
+ tp_comm_overlap_rs_dgrad ........................ False
525
+ tp_comm_split_ag ................................ True
526
+ tp_comm_split_rs ................................ True
527
+ train_data_path ................................. None
528
+ train_iters ..................................... 12
529
+ train_samples ................................... None
530
+ train_sync_interval ............................. None
531
+ transformer_impl ................................ transformer_engine
532
+ transformer_pipeline_model_parallel_size ........ 1
533
+ untie_embeddings_and_output_weights ............. True
534
+ use_checkpoint_args ............................. False
535
+ use_checkpoint_opt_param_scheduler .............. False
536
+ use_cpu_initialization .......................... None
537
+ use_custom_fsdp ................................. False
538
+ use_dist_ckpt ................................... True
539
+ use_dist_ckpt_deprecated ........................ False
540
+ use_distributed_optimizer ....................... True
541
+ use_flash_attn .................................. False
542
+ use_legacy_models ............................... False
543
+ use_mp_args_from_checkpoint_args ................ False
544
+ use_one_sent_docs ............................... False
545
+ use_persistent_ckpt_worker ...................... False
546
+ use_precision_aware_optimizer ................... False
547
+ use_pytorch_profiler ............................ False
548
+ use_ring_exchange_p2p ........................... False
549
+ use_rope_scaling ................................ False
550
+ use_rotary_position_embeddings .................. False
551
+ use_sharp ....................................... False
552
+ use_tokenizer_model_from_checkpoint_args ........ True
553
+ use_torch_fsdp2 ................................. False
554
+ use_torch_optimizer_for_cpu_offload ............. False
555
+ use_tp_pp_dp_mapping ............................ False
556
+ v_head_dim ...................................... 128
557
+ valid_data_path ................................. None
558
+ variable_seq_lengths ............................ False
559
+ virtual_pipeline_model_parallel_size ............ None
560
+ vision_backbone_type ............................ vit
561
+ vision_pretraining .............................. False
562
+ vision_pretraining_type ......................... classify
563
+ vocab_extra_ids ................................. 0
564
+ vocab_file ...................................... /workspace/data/wikitext/gpt2-vocab.json
565
+ vocab_size ...................................... None
566
+ wandb_exp_name ..................................
567
+ wandb_project ...................................
568
+ wandb_save_dir ..................................
569
+ weight_decay .................................... 0.1
570
+ weight_decay_incr_style ......................... constant
571
+ wgrad_deferral_limit ............................ 0
572
+ world_size ...................................... 8
573
+ yaml_cfg ........................................ None
574
+ -------------------- end of arguments ---------------------
575
+ INFO:megatron.core.num_microbatches_calculator:setting number of microbatches to constant 2
576
+ > building GPT2BPETokenizer tokenizer ...
577
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
578
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
579
+ > padded vocab (size: 50257) with 47 dummy tokens (new size: 50304)
580
+ WARNING:megatron.core.rerun_state_machine:RerunStateMachine initialized in mode RerunMode.VALIDATE_RESULTS
581
+ > initializing torch distributed ...
582
+ [Gloo] Rank 4 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
583
+ [Gloo] Rank 0 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
584
+ [Gloo] Rank 1 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
585
+ [Gloo] Rank 2 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
586
+ [Gloo] Rank 3 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
587
+ [Gloo] Rank 6 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
588
+ [Gloo] Rank 5 is connected to 7 peer ranks. Expected number of connected peer ranks is : [Gloo] Rank 77
589
+ is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
590
+ [Gloo] Rank 2 is connected to 7 peer ranks. Expected number of connected peer ranks is : [Gloo] Rank 7
591
+ 0 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
592
+ [Gloo] Rank 6 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
593
+ [Gloo] Rank 1 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
594
+ [Gloo] Rank [Gloo] Rank 4 is connected to 37 is connected to peer ranks. 7Expected number of connected peer ranks is : peer ranks. 7Expected number of connected peer ranks is :
595
+ 7
596
+ [Gloo] Rank 5 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
597
+ [Gloo] Rank 7 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
598
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
599
+ > initialized tensor model parallel with size 1
600
+ > initialized pipeline model parallel with size 1
601
+ > setting random seeds to 1234 ...
602
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
603
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
604
+ [Gloo] Rank [Gloo] Rank 00 is connected to is connected to 00 peer ranks. peer ranks. Expected number of connected peer ranks is : Expected number of connected peer ranks is : 00
605
+
606
+ [Gloo] Rank [Gloo] Rank 0[Gloo] Rank 0 is connected to 0 is connected to 0 is connected to 0 peer ranks. 0 peer ranks. Expected number of connected peer ranks is : peer ranks. Expected number of connected peer ranks is : 0Expected number of connected peer ranks is : 0
607
+ 0
608
+
609
+ [rank5]:[W830 07:20:11.563056647 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 5] using GPU 5 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
610
+ [rank6]:[W830 07:20:11.563138526 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 6] using GPU 6 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
611
+ [rank2]:[W830 07:20:11.563141602 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 2] using GPU 2 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
612
+ [rank4]:[W830 07:20:11.563175617 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 4] using GPU 4 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
613
+ [rank7]:[W830 07:20:11.567434592 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 7] using GPU 7 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
614
+ [rank1]:[W830 07:20:11.567440190 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 1] using GPU 1 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
615
+ [rank3]:[W830 07:20:11.569357811 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 3] using GPU 3 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
616
+ > compiling dataset index builder ...
617
+ make: Entering directory '/workspace/Megatron-UltraEP/megatron/core/datasets'
618
+ make: Nothing to be done for 'default'.
619
+ make: Leaving directory '/workspace/Megatron-UltraEP/megatron/core/datasets'
620
+ >>> done with dataset index builder. Compilation time: 0.034 seconds
621
+ WARNING: constraints for invoking optimized fused softmax kernel are not met. We default back to unfused kernel invocations.
622
+ > compiling and loading fused kernels ...
623
+ [rank0]:[W830 07:20:11.715944073 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 0] using GPU 0 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
624
+ >>> done with compiling and loading fused kernels. Compilation time: 3.736 seconds
625
+ time to initialize megatron (seconds): 7.282
626
+ [after megatron is initialized] datetime: 2026-08-30 07:20:17
627
+ building GPT model ...
628
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
629
+ warnings.warn(
630
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
631
+ warnings.warn(
632
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
633
+ warnings.warn(
634
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
635
+ warnings.warn(
636
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
637
+ warnings.warn(
638
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
639
+ warnings.warn(
640
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
641
+ warnings.warn(
642
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
643
+ warnings.warn(
644
+ [INFO] Use MNNVL fabric: False
645
+ [WARN] MAX_NUM_NVL_PEERS is not set. Using detected value 8.
646
+ NVSHMEM v3.7.2
647
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
648
+
649
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
650
+
651
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
652
+
653
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
654
+
655
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
656
+
657
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
658
+
659
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
660
+
661
+ /dvs/p4/build/sw/rel/gpgpu/toolkit/r12.9/main_nvshmem/src/modules/transport/ibrc/ibrc.cpp:1754: error status: 7 (NVSHMEMX_ERROR_INTERNAL) Device enumeration failed.
662
+
663
+ UltraEP: NVL domain size <= 8; forcing weight_sync plan mode to direct (was adaptive).
664
+ > number of parameters on (tensor, pipeline) model parallel rank (0, 0): 1687818240
665
+ INFO:megatron.core.distributed.distributed_data_parallel:Setting up DistributedDataParallel with config DistributedDataParallelConfig(grad_reduce_in_fp32=True, overlap_grad_reduce=True, overlap_param_gather=True, align_param_gather=False, use_distributed_optimizer=True, num_distributed_optimizer_instances=1, check_for_nan_in_grad=False, check_for_large_grads=False, bucket_size=40000000, pad_buckets_for_high_nccl_busbw=False, average_in_collective=False, fp8_param_gather=False, reuse_grad_buf_for_mxfp8_param_ag=False, use_custom_fsdp=False, data_parallel_sharding_strategy='no_shard', gradient_reduce_div_fusion=True, suggested_communication_unit_size=None, preserve_fp32_weights=True, keep_fp8_transpose_cache_when_using_custom_fsdp=False, nccl_ub=False, fsdp_double_buffer=False)
666
+ INFO:megatron.core.distributed.param_and_grad_buffer:Number of buckets for gradient all-reduce / reduce-scatter: 18
667
+ Params for bucket 1 (206045184 elements, 206045184 padded size):
668
+ module.output_layer.weight
669
+ Params for bucket 2 (41988096 elements, 41988096 padded size):
670
+ module.decoder.final_layernorm.weight
671
+ module.decoder.layers.15.self_attention.linear_qkv.layer_norm_weight
672
+ module.decoder.layers.15.mlp.router.weight
673
+ module.decoder.layers.15.self_attention.linear_proj.weight
674
+ module.decoder.layers.15.self_attention.linear_qkv.weight
675
+ module.decoder.layers.15.pre_mlp_layernorm.weight
676
+ Params for bucket 3 (41984000 elements, 41984000 padded size):
677
+ module.decoder.layers.14.mlp.router.weight
678
+ module.decoder.layers.14.self_attention.linear_proj.weight
679
+ module.decoder.layers.14.self_attention.linear_qkv.layer_norm_weight
680
+ module.decoder.layers.14.pre_mlp_layernorm.weight
681
+ module.decoder.layers.14.self_attention.linear_qkv.weight
682
+ Params for bucket 4 (41984000 elements, 41984000 padded size):
683
+ module.decoder.layers.13.mlp.router.weight
684
+ module.decoder.layers.13.self_attention.linear_qkv.layer_norm_weight
685
+ module.decoder.layers.13.self_attention.linear_proj.weight
686
+ module.decoder.layers.13.self_attention.linear_qkv.weight
687
+ module.decoder.layers.13.pre_mlp_layernorm.weight
688
+ Params for bucket 5 (41984000 elements, 41984000 padded size):
689
+ module.decoder.layers.12.mlp.router.weight
690
+ module.decoder.layers.12.self_attention.linear_qkv.layer_norm_weight
691
+ module.decoder.layers.12.self_attention.linear_proj.weight
692
+ module.decoder.layers.12.pre_mlp_layernorm.weight
693
+ module.decoder.layers.12.self_attention.linear_qkv.weight
694
+ Params for bucket 6 (41984000 elements, 41984000 padded size):
695
+ module.decoder.layers.11.mlp.router.weight
696
+ module.decoder.layers.11.self_attention.linear_qkv.layer_norm_weight
697
+ module.decoder.layers.11.self_attention.linear_proj.weight
698
+ module.decoder.layers.11.self_attention.linear_qkv.weight
699
+ module.decoder.layers.11.pre_mlp_layernorm.weight
700
+ Params for bucket 7 (41984000 elements, 41984000 padded size):
701
+ module.decoder.layers.10.mlp.router.weight
702
+ module.decoder.layers.10.self_attention.linear_proj.weight
703
+ module.decoder.layers.10.self_attention.linear_qkv.layer_norm_weight
704
+ module.decoder.layers.10.self_attention.linear_qkv.weight
705
+ module.decoder.layers.10.pre_mlp_layernorm.weight
706
+ Params for bucket 8 (41984000 elements, 41984000 padded size):
707
+ module.decoder.layers.9.self_attention.linear_qkv.weight
708
+ module.decoder.layers.9.pre_mlp_layernorm.weight
709
+ module.decoder.layers.9.mlp.router.weight
710
+ module.decoder.layers.9.self_attention.linear_proj.weight
711
+ module.decoder.layers.9.self_attention.linear_qkv.layer_norm_weight
712
+ Params for bucket 9 (41984000 elements, 41984000 padded size):
713
+ module.decoder.layers.8.mlp.router.weight
714
+ module.decoder.layers.8.self_attention.linear_proj.weight
715
+ module.decoder.layers.8.self_attention.linear_qkv.layer_norm_weight
716
+ module.decoder.layers.8.pre_mlp_layernorm.weight
717
+ module.decoder.layers.8.self_attention.linear_qkv.weight
718
+ Params for bucket 10 (41984000 elements, 41984000 padded size):
719
+ module.decoder.layers.7.mlp.router.weight
720
+ module.decoder.layers.7.self_attention.linear_qkv.layer_norm_weight
721
+ module.decoder.layers.7.self_attention.linear_proj.weight
722
+ module.decoder.layers.7.self_attention.linear_qkv.weight
723
+ module.decoder.layers.7.pre_mlp_layernorm.weight
724
+ Params for bucket 11 (41984000 elements, 41984000 padded size):
725
+ module.decoder.layers.6.mlp.router.weight
726
+ module.decoder.layers.6.self_attention.linear_proj.weight
727
+ module.decoder.layers.6.self_attention.linear_qkv.layer_norm_weight
728
+ module.decoder.layers.6.pre_mlp_layernorm.weight
729
+ module.decoder.layers.6.self_attention.linear_qkv.weight
730
+ Params for bucket 12 (41984000 elements, 41984000 padded size):
731
+ module.decoder.layers.5.mlp.router.weight
732
+ module.decoder.layers.5.self_attention.linear_qkv.layer_norm_weight
733
+ module.decoder.layers.5.self_attention.linear_qkv.weight
734
+ module.decoder.layers.5.self_attention.linear_proj.weight
735
+ module.decoder.layers.5.pre_mlp_layernorm.weight
736
+ Params for bucket 13 (41984000 elements, 41984000 padded size):
737
+ module.decoder.layers.4.pre_mlp_layernorm.weight
738
+ module.decoder.layers.4.self_attention.linear_qkv.weight
739
+ module.decoder.layers.4.mlp.router.weight
740
+ module.decoder.layers.4.self_attention.linear_qkv.layer_norm_weight
741
+ module.decoder.layers.4.self_attention.linear_proj.weight
742
+ Params for bucket 14 (41984000 elements, 41984000 padded size):
743
+ module.decoder.layers.3.mlp.router.weight
744
+ module.decoder.layers.3.self_attention.linear_proj.weight
745
+ module.decoder.layers.3.self_attention.linear_qkv.layer_norm_weight
746
+ module.decoder.layers.3.self_attention.linear_qkv.weight
747
+ module.decoder.layers.3.pre_mlp_layernorm.weight
748
+ Params for bucket 15 (41984000 elements, 41984000 padded size):
749
+ module.decoder.layers.2.mlp.router.weight
750
+ module.decoder.layers.2.self_attention.linear_proj.weight
751
+ module.decoder.layers.2.self_attention.linear_qkv.layer_norm_weight
752
+ module.decoder.layers.2.self_attention.linear_qkv.weight
753
+ module.decoder.layers.2.pre_mlp_layernorm.weight
754
+ Params for bucket 16 (41984000 elements, 41984000 padded size):
755
+ module.decoder.layers.1.mlp.router.weight
756
+ module.decoder.layers.1.self_attention.linear_qkv.weight
757
+ module.decoder.layers.1.pre_mlp_layernorm.weight
758
+ module.decoder.layers.1.self_attention.linear_qkv.layer_norm_weight
759
+ module.decoder.layers.1.self_attention.linear_proj.weight
760
+ Params for bucket 17 (41984000 elements, 41984000 padded size):
761
+ module.decoder.layers.0.mlp.router.weight
762
+ module.decoder.layers.0.self_attention.linear_qkv.layer_norm_weight
763
+ module.decoder.layers.0.pre_mlp_layernorm.weight
764
+ module.decoder.layers.0.self_attention.linear_qkv.weight
765
+ module.decoder.layers.0.self_attention.linear_proj.weight
766
+ Params for bucket 18 (206045184 elements, 206045184 padded size):
767
+ module.embedding.word_embeddings.weight
768
+ INFO:megatron.core.distributed.param_and_grad_buffer:Number of buckets for gradient all-reduce / reduce-scatter: 7
769
+ Params for bucket 1 (44040192 elements, 44040192 padded size):
770
+ module.decoder.layers.15.mlp.experts.linear_fc1.weight0
771
+ module.decoder.layers.14.mlp.experts.linear_fc1.weight0
772
+ module.decoder.layers.15.mlp.experts.linear_fc2.weight0
773
+ module.decoder.layers.13.mlp.experts.linear_fc2.weight0
774
+ module.decoder.layers.14.mlp.experts.linear_fc2.weight0
775
+ Params for bucket 2 (50331648 elements, 50331648 padded size):
776
+ module.decoder.layers.13.mlp.experts.linear_fc1.weight0
777
+ module.decoder.layers.11.mlp.experts.linear_fc1.weight0
778
+ module.decoder.layers.12.mlp.experts.linear_fc1.weight0
779
+ module.decoder.layers.12.mlp.experts.linear_fc2.weight0
780
+ module.decoder.layers.11.mlp.experts.linear_fc2.weight0
781
+ Params for bucket 3 (44040192 elements, 44040192 padded size):
782
+ module.decoder.layers.9.mlp.experts.linear_fc2.weight0
783
+ module.decoder.layers.10.mlp.experts.linear_fc1.weight0
784
+ module.decoder.layers.10.mlp.experts.linear_fc2.weight0
785
+ module.decoder.layers.8.mlp.experts.linear_fc2.weight0
786
+ module.decoder.layers.9.mlp.experts.linear_fc1.weight0
787
+ Params for bucket 4 (50331648 elements, 50331648 padded size):
788
+ module.decoder.layers.8.mlp.experts.linear_fc1.weight0
789
+ module.decoder.layers.7.mlp.experts.linear_fc1.weight0
790
+ module.decoder.layers.6.mlp.experts.linear_fc1.weight0
791
+ module.decoder.layers.7.mlp.experts.linear_fc2.weight0
792
+ module.decoder.layers.6.mlp.experts.linear_fc2.weight0
793
+ Params for bucket 5 (44040192 elements, 44040192 padded size):
794
+ module.decoder.layers.5.mlp.experts.linear_fc2.weight0
795
+ module.decoder.layers.3.mlp.experts.linear_fc2.weight0
796
+ module.decoder.layers.4.mlp.experts.linear_fc2.weight0
797
+ module.decoder.layers.5.mlp.experts.linear_fc1.weight0
798
+ module.decoder.layers.4.mlp.experts.linear_fc1.weight0
799
+ Params for bucket 6 (50331648 elements, 50331648 padded size):
800
+ module.decoder.layers.3.mlp.experts.linear_fc1.weight0
801
+ module.decoder.layers.2.mlp.experts.linear_fc1.weight0
802
+ module.decoder.layers.2.mlp.experts.linear_fc2.weight0
803
+ module.decoder.layers.1.mlp.experts.linear_fc1.weight0
804
+ module.decoder.layers.1.mlp.experts.linear_fc2.weight0
805
+ Params for bucket 7 (18874368 elements, 18874368 padded size):
806
+ module.decoder.layers.0.mlp.experts.linear_fc2.weight0
807
+ module.decoder.layers.0.mlp.experts.linear_fc1.weight0
808
+ INFO:megatron.core.optimizer:Setting up optimizer with config OptimizerConfig(optimizer='adam', lr=0.0001, min_lr=1e-05, decoupled_lr=None, decoupled_min_lr=None, weight_decay=0.1, fp8_recipe='mxfp8', fp16=False, bf16=True, reuse_grad_buf_for_mxfp8_param_ag=False, params_dtype=torch.bfloat16, use_precision_aware_optimizer=False, store_param_remainders=True, main_grads_dtype=torch.float32, main_params_dtype=torch.float32, exp_avg_dtype=torch.float32, exp_avg_sq_dtype=torch.float32, loss_scale=None, initial_loss_scale=4294967296, min_loss_scale=1.0, loss_scale_window=1000, hysteresis=2, adam_beta1=0.9, adam_beta2=0.999, adam_eps=1e-08, sgd_momentum=0.9, use_distributed_optimizer=True, overlap_param_gather_with_optimizer_step=False, optimizer_cpu_offload=False, optimizer_offload_fraction=1.0, use_torch_optimizer_for_cpu_offload=False, overlap_cpu_optimizer_d2h_h2d=False, pin_cpu_grads=True, pin_cpu_params=True, clip_grad=1.0, log_num_zeros_in_grad=False, barrier_with_L1_time=True, timers=<megatron.core.timers.Timers object at 0x756aba3ad010>, config_logger_dir='')
809
+ INFO:megatron.core.optimizer_param_scheduler:> learning rate decay style: cosine
810
+ [after model, optimizer, and learning rate scheduler are built] datetime: 2026-08-30 07:20:20
811
+ > building train, validation, and test datasets ...
812
+ > datasets target sizes (minimum size):
813
+ train: 192
814
+ validation: 0
815
+ test: 0
816
+ INFO:megatron.core.datasets.blended_megatron_dataset_config:Let split_matrix = [(0, 1.0), None, None]
817
+ > building train, validation, and test datasets for GPT ...
818
+ INFO:megatron.core.datasets.blended_megatron_dataset_builder:Building GPTDataset splits with sizes=(192, 0, 0) and config=GPTDatasetConfig(random_seed=1234, sequence_length=8192, blend=(['/workspace/data/wikitext/wikitext103_gpt2_text_document'], None), blend_per_split=None, split='100,0,0', split_matrix=[(0, 1.0), None, None], num_dataset_builder_threads=1, path_to_cache=None, mmap_bin_files=True, mock=False, tokenizer=<megatron.training.tokenizer.tokenizer._GPT2BPETokenizer object at 0x756abb6d4c80>, mid_level_dataset_surplus=0.005, reset_position_ids=False, reset_attention_mask=False, eod_mask_loss=False, create_attention_mask=True, drop_last_partial_validation_sequence=True, add_extra_token_to_sequence=True, object_storage_cache_path=None)
819
+ INFO:megatron.core.datasets.indexed_dataset:Load the _IndexReader from /workspace/data/wikitext/wikitext103_gpt2_text_document.idx
820
+ INFO:megatron.core.datasets.indexed_dataset: Extract the sequence lengths
821
+ INFO:megatron.core.datasets.indexed_dataset: Extract the sequence pointers
822
+ INFO:megatron.core.datasets.indexed_dataset: Extract the document indices
823
+ INFO:megatron.core.datasets.indexed_dataset:> total number of sequences: 1165029
824
+ INFO:megatron.core.datasets.indexed_dataset:> total number of documents: 1801350
825
+ INFO:megatron.core.datasets.gpt_dataset:Load the GPTDataset train indices
826
+ INFO:megatron.core.datasets.gpt_dataset: Load the document index from b16a908ae7f33ed67d5ec87f6f633afe-GPTDataset-train-document_index.npy
827
+ INFO:megatron.core.datasets.gpt_dataset: Load the sample index from b16a908ae7f33ed67d5ec87f6f633afe-GPTDataset-train-sample_index.npy
828
+ INFO:megatron.core.datasets.gpt_dataset: Load the shuffle index from b16a908ae7f33ed67d5ec87f6f633afe-GPTDataset-train-shuffle_index.npy
829
+ INFO:megatron.core.datasets.gpt_dataset:> total number of samples: 14536
830
+ > finished creating GPT datasets ...
831
+ [after dataloaders are built] datetime: 2026-08-30 07:20:20
832
+ done with setup ...
833
+ training ...
834
+ Overwriting rerun_state_machine.current_iteration from -1 to 0...
835
+ (min, max) time across ranks (ms):
836
+ model-and-optimizer-setup ......................: (3468.02, 3468.25)
837
+ train/valid/test-data-iterators-setup ..........: (71.00, 93.80)
838
+ [before the start of training step] datetime: 2026-08-30 07:20:20
839
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
840
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
841
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
842
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
843
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
844
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
845
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
846
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
847
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
848
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
849
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
850
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
851
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
852
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
853
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
854
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
855
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
856
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
857
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
858
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
859
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
860
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
861
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
862
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
863
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
864
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
865
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
866
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
867
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
868
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
869
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
870
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
871
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
872
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
873
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
874
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
875
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
876
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
877
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
878
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
879
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
880
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
881
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
882
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
883
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
884
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
885
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
886
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
887
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
888
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
889
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
890
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
891
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
892
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
893
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
894
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
895
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
896
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
897
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
898
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
899
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
900
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
901
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
902
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
903
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
904
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
905
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
906
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
907
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
908
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
909
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
910
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
911
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
912
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
913
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
914
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
915
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
916
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
917
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
918
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
919
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
920
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
921
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
922
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
923
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
924
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
925
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
926
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
927
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
928
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
929
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
930
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
931
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
932
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
933
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
934
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
935
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
936
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
937
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
938
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
939
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
940
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
941
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
942
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
943
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
944
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
945
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
946
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
947
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
948
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
949
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
950
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
951
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
952
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
953
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
954
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
955
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
956
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
957
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
958
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
959
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
960
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
961
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
962
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
963
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
964
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
965
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
966
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
967
+ [2026-08-30 07:20:46] iteration 1/ 12 | consumed samples: 16 | elapsed time per iteration (ms): 25702.4 | throughput per GPU (TFLOP/s/GPU): 7.7 | learning rate: 2.000000E-05 | global batch size: 16 | lm loss: 1.103051E+01 | loss scale: 1.0 | grad norm: 50.241 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 28416 | min safe capacity: 1.7344 |
968
+ Number of parameters in transformer block in billions: 3.09
969
+ Number of parameters in embedding layers in billions: 0.41
970
+ Total number of parameters in billions: 3.50
971
+ Number of parameters in most loaded shard in billions: 3.4994
972
+ Theoretical memory footprints: weight and optimizer=25029.43 MB
973
+ [Rank 0] (after 1 iterations) memory (MB) | allocated: 14963.09814453125 | max allocated: 30454.23095703125 | reserved: 31264.0 | max reserved: 31264.0
974
+ [2026-08-30 07:20:46] iteration 2/ 12 | consumed samples: 32 | elapsed time per iteration (ms): 337.6 | throughput per GPU (TFLOP/s/GPU): 587.7 | learning rate: 4.000000E-05 | global batch size: 16 | lm loss: 1.101006E+01 | loss scale: 1.0 | grad norm: 51.123 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 28416 | min safe capacity: 1.7344 |
975
+ [2026-08-30 07:20:47] iteration 3/ 12 | consumed samples: 48 | elapsed time per iteration (ms): 329.7 | throughput per GPU (TFLOP/s/GPU): 601.7 | learning rate: 6.000000E-05 | global batch size: 16 | lm loss: 1.053758E+01 | loss scale: 1.0 | grad norm: 249.877 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 28416 | min safe capacity: 1.7344 |
976
+ Capture CUDA graph for training!!!
977
+ CUDA graph capture done!!!
tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/results.tsv ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ config backend policy slots seq gbs cap recompute rep status iter_ms_mean iter_ms_min peak_mib run_dir
2
+ PONLY ultraep one 1 4096 16 2.40625 none 1 OK 176.80 175.80 38329 /workspace/tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq4096__gbs16__cap2p40625/rep1
3
+ PONLY ultraep one 1 4096 16 2.40625 none 2 OK 176.80 175.80 38329 /workspace/tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq4096__gbs16__cap2p40625/rep2
4
+ PONLY ultraep one 1 4096 16 2.40625 none 3 OK 176.76 175.80 38331 /workspace/tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq4096__gbs16__cap2p40625/rep3
5
+ PONLY ultraep one 1 8192 16 2.40625 none 1 OK 292.76 290.70 51017 /workspace/tiera-rerun-fixed-moonep/ultraep/ponly/ponly_gbs16/PONLY__ultraep__one__seq8192__gbs16__cap2p40625/rep1
tiera-rerun-fixed-moonep/ultraep/ponly/provenance.txt ADDED
@@ -0,0 +1,90 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ==============================================================
2
+ PROVENANCE: ultraep__ponly__gbs16
3
+ ==============================================================
4
+ captured_at : 2026-08-30T07:15:19+00:00
5
+ hostname : sp-aiinfra-interactive-aghosh-7jtjl-8xb200-moe-cg-0
6
+ kernel : 6.8.0-1043-nvidia
7
+ cwd : /workspace/blackwell_run_scripts
8
+ command : run_ladder.sh --phase ponly --gbs 16 --out /workspace/tiera-rerun-fixed-moonep/ultraep/ponly
9
+
10
+ ---- GPU / driver --------------------------------------------
11
+ 0, NVIDIA B200, 10.0, 183359 MiB, 580.95.05
12
+ 1, NVIDIA B200, 10.0, 183359 MiB, 580.95.05
13
+ 2, NVIDIA B200, 10.0, 183359 MiB, 580.95.05
14
+ 3, NVIDIA B200, 10.0, 183359 MiB, 580.95.05
15
+ 4, NVIDIA B200, 10.0, 183359 MiB, 580.95.05
16
+ 5, NVIDIA B200, 10.0, 183359 MiB, 580.95.05
17
+ 6, NVIDIA B200, 10.0, 183359 MiB, 580.95.05
18
+ 7, NVIDIA B200, 10.0, 183359 MiB, 580.95.05
19
+ persistence/clocks:
20
+ 0, 120 MHz, 1965 MHz, 1000.00 W, 29
21
+ 1, 195 MHz, 1965 MHz, 1000.00 W, 34
22
+ 2, 330 MHz, 1965 MHz, 1000.00 W, 35
23
+ 3, 120 MHz, 1965 MHz, 1000.00 W, 29
24
+ 4, 135 MHz, 1965 MHz, 1000.00 W, 31
25
+ 5, 180 MHz, 1965 MHz, 1000.00 W, 34
26
+ 6, 135 MHz, 1965 MHz, 1000.00 W, 34
27
+ 7, 120 MHz, 1965 MHz, 1000.00 W, 30
28
+
29
+ NVLink topology:
30
+ GPU0 GPU1 GPU2 GPU3 GPU4 GPU5 GPU6 GPU7 NIC0 NIC1 NIC2 NIC3 NIC4 NIC5 NIC6 NIC7 NIC8 CPU Affinity NUMA Affinity GPU NUMA ID
31
+ GPU0 X NV18 NV18 NV18 NV18 NV18 NV18 NV18 PIX NODE SYS SYS SYS SYS SYS SYS NODE 0-27 0 N/A
32
+ GPU1 NV18 X NV18 NV18 NV18 NV18 NV18 NV18 NODE PIX SYS SYS SYS SYS SYS SYS NODE 0-27 0 N/A
33
+ GPU2 NV18 NV18 X NV18 NV18 NV18 NV18 NV18 SYS SYS PIX NODE SYS SYS SYS SYS SYS 28-55 1 N/A
34
+ GPU3 NV18 NV18 NV18 X NV18 NV18 NV18 NV18 SYS SYS NODE PIX SYS SYS SYS SYS SYS 28-55 1 N/A
35
+ GPU4 NV18 NV18 NV18 NV18 X NV18 NV18 NV18 SYS SYS SYS SYS PIX NODE SYS SYS SYS 56-83 2 N/A
36
+ GPU5 NV18 NV18 NV18 NV18 NV18 X NV18 NV18 SYS SYS SYS SYS NODE PIX SYS SYS SYS 56-83 2 N/A
37
+ GPU6 NV18 NV18 NV18 NV18 NV18 NV18 X NV18 SYS SYS SYS SYS SYS SYS PIX NODE SYS 84-111 3 N/A
38
+ GPU7 NV18 NV18 NV18 NV18 NV18 NV18 NV18 X SYS SYS SYS SYS SYS SYS NODE PIX SYS 84-111 3 N/A
39
+ NIC0 PIX NODE SYS SYS SYS SYS SYS SYS X NODE SYS SYS SYS SYS SYS SYS NODE
40
+ NIC1 NODE PIX SYS SYS SYS SYS SYS SYS NODE X SYS SYS SYS SYS SYS SYS NODE
41
+ NIC2 SYS SYS PIX NODE SYS SYS SYS SYS SYS SYS X NODE SYS SYS SYS SYS SYS
42
+ NIC3 SYS SYS NODE PIX SYS SYS SYS SYS SYS SYS NODE X SYS SYS SYS SYS SYS
43
+ NIC4 SYS SYS SYS SYS PIX NODE SYS SYS SYS SYS SYS SYS X NODE SYS SYS SYS
44
+ unavailable
45
+
46
+ ---- toolchain -----------------------------------------------
47
+ nvcc : release 12.9, V12.9.86
48
+ gcc : gcc (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0
49
+ python : 3.12.3
50
+ torch : 2.8.0a0+5228986c39.nv25.06 cuda=12.9 nccl=2.27.3
51
+ torch arch : ['sm_75', 'sm_80', 'sm_86', 'sm_90', 'sm_100', 'sm_120', 'compute_120']
52
+ transformer_engine: 2.10.0.dev0+10f6128d [/usr/local/lib/python3.12/dist-packages/transformer_engine/__init__.py]
53
+ deep_ep : 1.2.1+34152ae [/workspace/DeepEP/deep_ep/__init__.py]
54
+ ultra_ep : 1.0.0 [/workspace/UltraEP/ultra_ep/__init__.py]
55
+ moonep : 0.0.1 [/workspace/MoonEP/moonep/__init__.py]
56
+ megatron : 0.14.0rc5 [None]
57
+ flash_attn : 2.7.4.post1 [/usr/local/lib/python3.12/dist-packages/flash_attn/__init__.py]
58
+ numpy : 1.26.4 [/usr/local/lib/python3.12/dist-packages/numpy/__init__.py]
59
+ huggingface_hub: 1.29.0 [/usr/local/lib/python3.12/dist-packages/huggingface_hub/__init__.py]
60
+
61
+ ---- repositories --------------------------------------------
62
+ Megatron-MoE-Echo bracing-regions-through-kernels 9a23679390d7
63
+ Megatron-UltraEP ultraep-bracing-regions-through-kernels 03d57ef7d476
64
+ UltraEP main 94cab099b44f
65
+ DeepEP HEAD 34152ae28f80 (+3 untracked: build output)
66
+ MoonEP megatron-ddp-symmetric-params d8bc6667fad7
67
+ blackwell_run_scripts granularity-ultraep 5e3993b15bed
68
+
69
+ ---- environment ---------------------------------------------
70
+ CUDA_ARCH_LIST=7.5 8.0 8.6 9.0 10.0 12.0
71
+ CUDA_DRIVER_VERSION=575.57.08
72
+ CUDA_HOME=/usr/local/cuda
73
+ CUDA_MODULE_LOADING=LAZY
74
+ CUDA_VERSION=12.9.1.010
75
+ HF_REMOTE_ROOT=tiera-rerun-fixed-moonep
76
+ NCCL_VERSION=2.27.3
77
+ NVSHMEM_VERSION=3.2.5
78
+ PYTORCH_BUILD_NUMBER=0
79
+ PYTORCH_BUILD_VERSION=2.8.0a0+5228986
80
+ PYTORCH_HOME=/opt/pytorch/pytorch
81
+ PYTORCH_VERSION=2.8.0a0+5228986
82
+ TORCH_ALLOW_TF32_CUBLAS_OVERRIDE=1
83
+ TORCH_CUDA_ARCH_LIST=7.5 8.0 8.6 9.0 10.0 12.0+PTX
84
+ TORCH_NCCL_USE_COMM_NONBLOCKING=0
85
+
86
+ ---- shared library resolution -------------------------------
87
+ libnvshmem_host.so: not in ldconfig
88
+ libnccl.so: /lib/x86_64-linux-gnu/libnccl.so.2
89
+ lrwxrwxrwx 1 root root 20 Aug 29 16:41 /usr/local/lib/python3.12/dist-packages/nvidia/nvshmem/lib/libnvshmem_host.so -> libnvshmem_host.so.3
90
+ -rw-r--r-- 1 root root 179045744 Aug 29 16:37 /usr/local/lib/python3.12/dist-packages/nvidia/nvshmem/lib/libnvshmem_host.so.3