AbhishekGhosh1998 commited on
Commit
a9aa42c
·
verified ·
1 Parent(s): 467b50d

granularity results tiera-rerun-fixed-moonep @ 2026-08-30T09:03:26Z from sp-aiinfra-interactive-aghosh-7jtjl-8xb200-moe-cg-0 (part 4)

Browse files
Files changed (14) hide show
  1. .gitattributes +2 -0
  2. tiera-rerun-fixed-moonep/ultraep/tierbtrace/manifest.txt +19 -0
  3. tiera-rerun-fixed-moonep/ultraep/tierbtrace/pip_freeze.txt +317 -0
  4. tiera-rerun-fixed-moonep/ultraep/tierbtrace/provenance.txt +90 -0
  5. tiera-rerun-fixed-moonep/ultraep/tierbtrace/tierbtrace_gbs16/G1__ultraep__matched__seq4096__gbs16__cap2p40625/nsys/G1__ultraep__matched__seq4096__gbs16__cap2p40625.nsys-rep +3 -0
  6. tiera-rerun-fixed-moonep/ultraep/tierbtrace/tierbtrace_gbs16/G1__ultraep__matched__seq4096__gbs16__cap2p40625/nsys/run.env +70 -0
  7. tiera-rerun-fixed-moonep/ultraep/tierbtrace/tierbtrace_gbs16/G1__ultraep__matched__seq4096__gbs16__cap2p40625/nsys/train.log +1075 -0
  8. tiera-rerun-fixed-moonep/ultraep/tierbtrace/tierbtrace_gbs16/G1__ultraep__matched__seq4096__gbs16__cap2p40625/nsys/verify_trace_ranks.log +6 -0
  9. tiera-rerun-fixed-moonep/ultraep/tierbtrace/tierbtrace_gbs16/G2__ultraep__matched__seq4096__gbs16__cap1p08203125/nsys/G2__ultraep__matched__seq4096__gbs16__cap1p08203125.nsys-rep +3 -0
  10. tiera-rerun-fixed-moonep/ultraep/tierbtrace/tierbtrace_gbs16/G2__ultraep__matched__seq4096__gbs16__cap1p08203125/nsys/run.env +70 -0
  11. tiera-rerun-fixed-moonep/ultraep/tierbtrace/tierbtrace_gbs16/G2__ultraep__matched__seq4096__gbs16__cap1p08203125/nsys/train.log +1132 -0
  12. tiera-rerun-fixed-moonep/ultraep/tierbtrace/tierbtrace_gbs16/results.tsv +1 -0
  13. tiera-rerun-fixed-moonep/ultraep/tierbtrace/tierbtrace_gbs16/trace_extension_build.log +58 -0
  14. tiera-rerun-fixed-moonep/ultraep/tierbtrace/tierbtrace_gbs16/trace_results.tsv +2 -0
.gitattributes CHANGED
@@ -74,3 +74,5 @@ tiera-rerun-fixed-moonep/echo/tierbtrace/tierbtrace_gbs16/G1__echo__matched__seq
74
  tiera-rerun-fixed-moonep/echo/tierbtrace/tierbtrace_gbs16/G2__echo__matched__seq4096__gbs16__cap1p08203125/nsys/G2__echo__matched__seq4096__gbs16__cap1p08203125.nsys-rep filter=lfs diff=lfs merge=lfs -text
75
  tiera-rerun-fixed-moonep/echo/tierbtrace/tierbtrace_gbs16/G3__echo__matched__seq4096__gbs16__cap1p072265625/nsys/G3__echo__matched__seq4096__gbs16__cap1p072265625.nsys-rep filter=lfs diff=lfs merge=lfs -text
76
  tiera-rerun-fixed-moonep/echo/tierbtrace/tierbtrace_gbs16/G4__echo__matched__seq4096__gbs16__cap1p095703125/nsys/G4__echo__matched__seq4096__gbs16__cap1p095703125.nsys-rep filter=lfs diff=lfs merge=lfs -text
 
 
 
74
  tiera-rerun-fixed-moonep/echo/tierbtrace/tierbtrace_gbs16/G2__echo__matched__seq4096__gbs16__cap1p08203125/nsys/G2__echo__matched__seq4096__gbs16__cap1p08203125.nsys-rep filter=lfs diff=lfs merge=lfs -text
75
  tiera-rerun-fixed-moonep/echo/tierbtrace/tierbtrace_gbs16/G3__echo__matched__seq4096__gbs16__cap1p072265625/nsys/G3__echo__matched__seq4096__gbs16__cap1p072265625.nsys-rep filter=lfs diff=lfs merge=lfs -text
76
  tiera-rerun-fixed-moonep/echo/tierbtrace/tierbtrace_gbs16/G4__echo__matched__seq4096__gbs16__cap1p095703125/nsys/G4__echo__matched__seq4096__gbs16__cap1p095703125.nsys-rep filter=lfs diff=lfs merge=lfs -text
77
+ tiera-rerun-fixed-moonep/ultraep/tierbtrace/tierbtrace_gbs16/G1__ultraep__matched__seq4096__gbs16__cap2p40625/nsys/G1__ultraep__matched__seq4096__gbs16__cap2p40625.nsys-rep filter=lfs diff=lfs merge=lfs -text
78
+ tiera-rerun-fixed-moonep/ultraep/tierbtrace/tierbtrace_gbs16/G2__ultraep__matched__seq4096__gbs16__cap1p08203125/nsys/G2__ultraep__matched__seq4096__gbs16__cap1p08203125.nsys-rep filter=lfs diff=lfs merge=lfs -text
tiera-rerun-fixed-moonep/ultraep/tierbtrace/manifest.txt ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ timestamp: 2026-08-30T08:54:25+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: tierbtrace
10
+ gbs: 16
11
+ seq: 4096
12
+ configs: G1 G2 G3 G4
13
+ capacity: 1.25
14
+ repeats: 3
15
+ nsys: 1
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/tierbtrace/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/tierbtrace/provenance.txt ADDED
@@ -0,0 +1,90 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ==============================================================
2
+ PROVENANCE: ultraep__tierbtrace__gbs16
3
+ ==============================================================
4
+ captured_at : 2026-08-30T08:54:11+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 tierbtrace --gbs 16 --out /workspace/tiera-rerun-fixed-moonep/ultraep/tierbtrace --nsys
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, 28
21
+ 1, 120 MHz, 1965 MHz, 1000.00 W, 32
22
+ 2, 120 MHz, 1965 MHz, 1000.00 W, 32
23
+ 3, 120 MHz, 1965 MHz, 1000.00 W, 28
24
+ 4, 120 MHz, 1965 MHz, 1000.00 W, 29
25
+ 5, 120 MHz, 1965 MHz, 1000.00 W, 32
26
+ 6, 120 MHz, 1965 MHz, 1000.00 W, 33
27
+ 7, 120 MHz, 1965 MHz, 1000.00 W, 28
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
tiera-rerun-fixed-moonep/ultraep/tierbtrace/tierbtrace_gbs16/G1__ultraep__matched__seq4096__gbs16__cap2p40625/nsys/G1__ultraep__matched__seq4096__gbs16__cap2p40625.nsys-rep ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:42f80a9659c34fed681c85e6bc76463298042813c5fb46d19b802053f4d6c515
3
+ size 10098534
tiera-rerun-fixed-moonep/ultraep/tierbtrace/tierbtrace_gbs16/G1__ultraep__matched__seq4096__gbs16__cap2p40625/nsys/run.env ADDED
@@ -0,0 +1,70 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ backend=ultraep
2
+ config=G1
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=14336
11
+ num_experts=8
12
+ topk=2
13
+ experts_per_rank=1
14
+ expert_params=176160768
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=matched
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=10
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=1
61
+ kernel_trace=1
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/tierbtrace/tierbtrace_gbs16/G1__ultraep__matched__seq4096__gbs16__cap2p40625/nsys/train.log ADDED
@@ -0,0 +1,1075 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ==============================================================
2
+ G1 / ultraep (slots/rank=1 policy=matched)
3
+ E=8 K=2 H'=14336 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/tierbtrace/tierbtrace_gbs16/G1__ultraep__matched__seq4096__gbs16__cap2p40625/nsys/memory_snapshots/snapshot.pickle
7
+ repo=/workspace/Megatron-UltraEP @ 03d57ef7
8
+ ==============================================================
9
+ W0830 08:58:39.576000 903525 torch/distributed/run.py:766]
10
+ W0830 08:58:39.576000 903525 torch/distributed/run.py:766] *****************************************
11
+ W0830 08:58:39.576000 903525 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 08:58:39.576000 903525 torch/distributed/run.py:766] *****************************************
13
+ 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
14
+ WARNING: overriding default arguments for tokenizer_type:GPT2BPETokenizer with tokenizer_type:GPT2BPETokenizer
15
+ Number of virtual stages per pipeline stage: None
16
+ accumulate and all-reduce gradients in fp32 for bfloat16 data type.
17
+ using torch.bfloat16 for parameters ...
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
+ ------------------------ arguments ------------------------
21
+ account_for_embedding_in_pipeline_split ......... False
22
+ account_for_loss_in_pipeline_split .............. False
23
+ accumulate_allreduce_grads_in_fp32 .............. True
24
+ adam_beta1 ...................................... 0.9
25
+ adam_beta2 ...................................... 0.999
26
+ adam_eps ........................................ 1e-08
27
+ add_bias_linear ................................. False
28
+ add_position_embedding .......................... False
29
+ add_qkv_bias .................................... False
30
+ adlr_autoresume ................................. False
31
+ adlr_autoresume_interval ........................ 1000
32
+ align_grad_reduce ............................... True
33
+ align_param_gather .............................. False
34
+ app_tag_run_name ................................ None
35
+ app_tag_run_version ............................. 0.0.0
36
+ apply_layernorm_1p .............................. False
37
+ apply_query_key_layer_scaling ................... False
38
+ apply_residual_connection_post_layernorm ........ False
39
+ apply_rope_fusion ............................... True
40
+ async_save ...................................... None
41
+ async_tensor_model_parallel_allreduce ........... True
42
+ attention_backend ............................... AttnBackend.auto
43
+ attention_dropout ............................... 0.0
44
+ attention_softmax_in_fp32 ....................... False
45
+ auto_detect_ckpt_format ......................... False
46
+ barrier_with_L1_time ............................ True
47
+ bert_binary_head ................................ True
48
+ bert_embedder_type .............................. megatron
49
+ bert_load ....................................... None
50
+ bf16 ............................................ True
51
+ bias_dropout_fusion ............................. True
52
+ bias_gelu_fusion ................................ False
53
+ bias_swiglu_fusion .............................. True
54
+ biencoder_projection_dim ........................ 0
55
+ biencoder_shared_query_context_model ............ False
56
+ block_data_path ................................. None
57
+ calc_ft_timeouts ................................ False
58
+ calculate_per_token_loss ........................ False
59
+ check_for_large_grads ........................... False
60
+ check_for_nan_in_loss_and_grad .................. False
61
+ check_for_spiky_loss ............................ False
62
+ check_weight_hash_across_dp_replicas_interval ... None
63
+ ckpt_assume_constant_structure .................. False
64
+ ckpt_convert_format ............................. None
65
+ ckpt_convert_save ............................... None
66
+ ckpt_convert_update_legacy_dist_opt_format ...... False
67
+ ckpt_format ..................................... torch_dist
68
+ ckpt_fully_parallel_load ........................ False
69
+ ckpt_fully_parallel_save ........................ True
70
+ ckpt_fully_parallel_save_deprecated ............. False
71
+ ckpt_step ....................................... None
72
+ classes_fraction ................................ 1.0
73
+ clip_grad ....................................... 1.0
74
+ clone_scatter_output_in_embedding ............... True
75
+ config_logger_dir ...............................
76
+ consumed_train_samples .......................... 0
77
+ consumed_valid_samples .......................... 0
78
+ context_parallel_size ........................... 1
79
+ cp_comm_type .................................... ['p2p']
80
+ create_attention_mask_in_dataloader ............. True
81
+ cross_entropy_fusion_impl ....................... native
82
+ cross_entropy_loss_fusion ....................... False
83
+ cuda_graph_scope ................................ full_iteration
84
+ cuda_graph_warmup_steps ......................... 3
85
+ data_args_path .................................. None
86
+ data_cache_path ................................. None
87
+ data_parallel_random_init ....................... False
88
+ data_parallel_sharding_strategy ................. no_shard
89
+ data_parallel_size .............................. 8
90
+ data_path ....................................... ['/workspace/data/wikitext/wikitext103_gpt2_text_document']
91
+ data_per_class_fraction ......................... 1.0
92
+ data_sharding ................................... True
93
+ dataloader_type ................................. single
94
+ ddp_average_in_collective ....................... False
95
+ ddp_bucket_size ................................. None
96
+ ddp_num_buckets ................................. None
97
+ ddp_pad_buckets_for_high_nccl_busbw ............. False
98
+ decoder_first_pipeline_num_layers ............... None
99
+ decoder_last_pipeline_num_layers ................ None
100
+ decoder_num_layers .............................. None
101
+ decoder_seq_length .............................. None
102
+ decoupled_lr .................................... None
103
+ decoupled_min_lr ................................ None
104
+ decrease_batch_size_if_needed ................... False
105
+ defer_embedding_wgrad_compute ................... False
106
+ delay_wgrad_compute ............................. False
107
+ deprecated_use_mcore_models ..................... True
108
+ deterministic_mode .............................. False
109
+ dino_bottleneck_size ............................ 256
110
+ dino_freeze_last_layer .......................... 1
111
+ dino_head_hidden_size ........................... 2048
112
+ dino_local_crops_number ......................... 10
113
+ dino_local_img_size ............................. 96
114
+ dino_norm_last_layer ............................ False
115
+ dino_teacher_temp ............................... 0.07
116
+ dino_warmup_teacher_temp ........................ 0.04
117
+ dino_warmup_teacher_temp_epochs ................. 30
118
+ disable_bf16_reduced_precision_matmul ........... False
119
+ disable_mamba_mem_eff_path ...................... False
120
+ disable_straggler_on_startup .................... False
121
+ dist_ckpt_format_deprecated ..................... None
122
+ dist_ckpt_strictness ............................ assume_ok_unexpected
123
+ distribute_saved_activations .................... False
124
+ distributed_backend ............................. nccl
125
+ distributed_timeout_minutes ..................... 10
126
+ embedding_init_method_std ....................... None
127
+ embedding_path .................................. None
128
+ empty_unused_memory_level ....................... 0
129
+ enable_cuda_graph ............................... True
130
+ enable_experimental ............................. False
131
+ enable_ft_package ............................... False
132
+ enable_gloo_process_groups ...................... True
133
+ enable_msc ...................................... True
134
+ enable_one_logger ............................... True
135
+ encoder_num_layers .............................. 16
136
+ encoder_seq_length .............................. 4096
137
+ end_weight_decay ................................ 0.1
138
+ eod_mask_loss ................................... False
139
+ error_injection_rate ............................ 0
140
+ error_injection_type ............................ transient_error
141
+ eval_interval ................................... 1000000
142
+ eval_iters ...................................... 0
143
+ evidence_data_path .............................. None
144
+ exit_duration_in_mins ........................... None
145
+ exit_interval ................................... None
146
+ exit_on_missing_checkpoint ...................... False
147
+ exit_signal_handler ............................. False
148
+ exp_avg_dtype ................................... torch.float32
149
+ exp_avg_sq_dtype ................................ torch.float32
150
+ expert_model_parallel_size ...................... 8
151
+ expert_tensor_parallel_size ..................... 1
152
+ export_force_local_attention .................... False
153
+ export_kd_cfg ................................... None
154
+ export_kd_teacher_ckpt_format ................... None
155
+ export_kd_teacher_load .......................... None
156
+ export_kv_cache_quant ........................... False
157
+ export_legacy_megatron .......................... False
158
+ export_model_type ............................... GPTModel
159
+ export_moe_apply_probs_on_input ................. False
160
+ export_qk_l2_norm ............................... False
161
+ export_quant_cfg ................................ None
162
+ export_real_quant_cfg ........................... None
163
+ export_te_mcore_model ........................... False
164
+ external_cuda_graph ............................. False
165
+ ffn_hidden_size ................................. 14336
166
+ finetune ........................................ False
167
+ finetune_data_split ............................. train
168
+ finetune_hf_dataset ............................. None
169
+ first_last_layers_bf16 .......................... False
170
+ flash_decode .................................... False
171
+ fp16 ............................................ False
172
+ fp16_lm_cross_entropy ........................... False
173
+ fp32_residual_connection ........................ False
174
+ fp8 ............................................. e4m3
175
+ fp8_amax_compute_algo ........................... most_recent
176
+ fp8_amax_history_len ............................ 1
177
+ fp8_interval .................................... 1
178
+ fp8_margin ...................................... 0
179
+ fp8_param_gather ................................ False
180
+ fp8_recipe ...................................... mxfp8
181
+ fp8_wgrad ....................................... True
182
+ fsdp_double_buffer .............................. False
183
+ global_batch_size ............................... 16
184
+ grad_reduce_in_bf16 ............................. False
185
+ gradient_accumulation_fusion .................... True
186
+ gradient_reduce_div_fusion ...................... True
187
+ group_query_attention ........................... True
188
+ head_lr_mult .................................... 1.0
189
+ heterogeneous_layers_config_encoded_json ........ None
190
+ heterogeneous_layers_config_path ................ None
191
+ hidden_dropout .................................. 0.0
192
+ hidden_size ..................................... 4096
193
+ hierarchical_context_parallel_sizes ............. None
194
+ high_priority_stream_groups ..................... []
195
+ hybrid_attention_ratio .......................... 0.0
196
+ hybrid_mlp_ratio ................................ 0.0
197
+ hybrid_override_pattern ......................... None
198
+ hysteresis ...................................... 2
199
+ ict_head_size ................................... None
200
+ ict_load ........................................ None
201
+ img_h ........................................... 224
202
+ img_w ........................................... 224
203
+ indexer_batch_size .............................. 128
204
+ indexer_log_interval ............................ 1000
205
+ inference_batch_times_seqlen_threshold .......... -1
206
+ inference_dynamic_batching ...................... False
207
+ inference_dynamic_batching_buffer_guaranteed_fraction 0.2
208
+ inference_dynamic_batching_buffer_overflow_factor None
209
+ inference_dynamic_batching_buffer_size_gb ....... 40.0
210
+ inference_dynamic_batching_chunk_size ........... 256
211
+ inference_dynamic_batching_max_requests_override None
212
+ inference_dynamic_batching_max_tokens_override .. None
213
+ inference_dynamic_batching_num_cuda_graphs ...... 16
214
+ inference_max_batch_size ........................ 8
215
+ inference_max_seq_length ........................ 2560
216
+ inference_rng_tracker ........................... False
217
+ init_method_std ................................. 0.01
218
+ init_method_xavier_uniform ...................... False
219
+ init_model_with_meta_device ..................... False
220
+ initial_loss_scale .............................. 4294967296
221
+ inprocess_active_world_size ..................... 8
222
+ inprocess_barrier_timeout ....................... 120
223
+ inprocess_completion_timeout .................... 120
224
+ inprocess_empty_cuda_cache ...................... False
225
+ inprocess_granularity ........................... node
226
+ inprocess_hard_timeout .......................... 90
227
+ inprocess_heartbeat_interval .................... 30
228
+ inprocess_heartbeat_timeout ..................... 60
229
+ inprocess_last_call_wait ........................ 1
230
+ inprocess_max_iterations ........................ None
231
+ inprocess_monitor_process_interval .............. 1.0
232
+ inprocess_monitor_thread_interval ............... 1.0
233
+ inprocess_progress_watchdog_interval ............ 1.0
234
+ inprocess_restart ............................... False
235
+ inprocess_soft_timeout .......................... 60
236
+ inprocess_termination_grace_time ................ 1
237
+ is_hybrid_model ................................. False
238
+ iter_per_epoch .................................. 1250
239
+ iterations_to_skip .............................. []
240
+ keep_fp8_transpose_cache_when_using_custom_fsdp . False
241
+ kitchen_config_file ............................. None
242
+ kitchen_recipe_number ........................... None
243
+ kv_channels ..................................... 128
244
+ kv_lora_rank .................................... 32
245
+ lazy_mpu_init ................................... None
246
+ load ............................................ None
247
+ load_main_params_from_ckpt ...................... None
248
+ load_model_opt_format ........................... False
249
+ local_rank ...................................... 0
250
+ log_energy ...................................... False
251
+ log_interval .................................... 1
252
+ log_loss_scale_to_tensorboard ................... True
253
+ log_memory_to_tensorboard ....................... False
254
+ log_num_zeros_in_grad ........................... False
255
+ log_params_norm ................................. False
256
+ log_progress .................................... False
257
+ log_straggler ................................... False
258
+ log_throughput .................................. True
259
+ log_timers_to_tensorboard ....................... False
260
+ log_validation_ppl_to_tensorboard ............... False
261
+ log_world_size_to_tensorboard ................... False
262
+ logging_level ................................... None
263
+ loss_scale ...................................... None
264
+ loss_scale_window ............................... 1000
265
+ lr .............................................. 0.0001
266
+ lr_decay_iters .................................. 8
267
+ lr_decay_samples ................................ None
268
+ lr_decay_style .................................. cosine
269
+ lr_warmup_fraction .............................. None
270
+ lr_warmup_init .................................. 0.0
271
+ lr_warmup_iters ................................. 5
272
+ lr_warmup_samples ............................... 0
273
+ lr_wsd_decay_iters .............................. None
274
+ lr_wsd_decay_samples ............................ None
275
+ lr_wsd_decay_style .............................. exponential
276
+ main_grads_dtype ................................ torch.float32
277
+ main_params_dtype ............................... torch.float32
278
+ make_vocab_size_divisible_by .................... 128
279
+ mamba_head_dim .................................. 64
280
+ mamba_num_groups ................................ 8
281
+ mamba_num_heads ................................. None
282
+ mamba_state_dim ................................. 128
283
+ manual_gc ....................................... False
284
+ manual_gc_eval .................................. True
285
+ manual_gc_interval .............................. 0
286
+ mask_factor ..................................... 1.0
287
+ mask_prob ....................................... 0.15
288
+ mask_type ....................................... random
289
+ masked_softmax_fusion ........................... False
290
+ max_position_embeddings ......................... 32768
291
+ max_tokens_to_oom ............................... 12000
292
+ memory_snapshot_path ............................ snapshot.pickle
293
+ merge_file ...................................... /workspace/data/wikitext/gpt2-merges.txt
294
+ micro_batch_size ................................ 1
295
+ microbatch_group_size_per_vp_stage .............. None
296
+ mid_level_dataset_surplus ....................... 0.005
297
+ min_loss_scale .................................. 1.0
298
+ min_lr .......................................... 1e-05
299
+ mlp_chunks_for_prefill .......................... 1
300
+ mmap_bin_files .................................. True
301
+ mock_data ....................................... False
302
+ moe_apply_probs_on_input ........................ False
303
+ moe_aux_loss_coeff .............................. 0.01
304
+ moe_capacity_monitor ............................ True
305
+ moe_deepep_num_sms .............................. 20
306
+ moe_echo_enable_random_offloading ............... False
307
+ moe_echo_expert_dispatch_overlap ................ False
308
+ moe_echo_expert_dispatcher_type ................. hybridep
309
+ moe_echo_recompute_expert_dispatch .............. False
310
+ moe_enable_deepep ............................... False
311
+ moe_enable_echo ................................. False
312
+ moe_enable_hybridep ............................. True
313
+ moe_enable_ultraep .............................. True
314
+ moe_expert_capacity_factor ...................... None
315
+ moe_extended_tp ................................. False
316
+ moe_ffn_hidden_size ............................. 14336
317
+ moe_grouped_gemm ................................ True
318
+ moe_hybridep_num_sms ............................ 16
319
+ moe_input_jitter_eps ............................ None
320
+ moe_layer_freq .................................. 1
321
+ moe_layer_recompute ............................. False
322
+ moe_num_echo_experts ............................ None
323
+ moe_num_redundant_experts_per_rank .............. 1
324
+ moe_pad_expert_input_to_capacity ................ False
325
+ moe_per_layer_logging ........................... False
326
+ moe_permute_fusion .............................. False
327
+ moe_received_token_capacity ..................... 2.40625
328
+ moe_router_bias_update_rate ..................... 0.001
329
+ moe_router_dtype ................................ fp32
330
+ moe_router_enable_expert_bias ................... False
331
+ moe_router_force_load_balancing ................. False
332
+ moe_router_group_topk ........................... None
333
+ moe_router_load_balancing_type .................. aux_loss
334
+ moe_router_num_groups ........................... None
335
+ moe_router_padding_for_fp8 ...................... False
336
+ moe_router_pre_softmax .......................... False
337
+ moe_router_score_function ....................... softmax
338
+ moe_router_topk ................................. 2
339
+ moe_router_topk_scaling_factor .................. None
340
+ moe_shared_expert_intermediate_size ............. None
341
+ moe_shared_expert_overlap ....................... False
342
+ moe_token_dispatcher_type ....................... flex
343
+ moe_token_drop_policy ........................... probs
344
+ moe_ultraep_max_microbatches .................... None
345
+ moe_ultraep_overlap_weight_sync ................. True
346
+ moe_ultraep_weight_sync_join .................... before_dispatch
347
+ moe_upcycling_granularity ....................... 1
348
+ moe_use_device_initiated_grouped_gemm ........... True
349
+ moe_use_legacy_grouped_gemm ..................... False
350
+ moe_use_upcycling ............................... False
351
+ moe_z_loss_coeff ................................ None
352
+ mrope_section ................................... None
353
+ mscale .......................................... 1.0
354
+ mscale_all_dim .................................. 0.0
355
+ mtp_loss_scaling_factor ......................... 0.1
356
+ mtp_num_layers .................................. None
357
+ multi_latent_attention .......................... False
358
+ nccl_all_reduce_for_prefill ..................... False
359
+ nccl_communicator_config_path ................... None
360
+ nccl_ub ......................................... False
361
+ no_load_optim ................................... None
362
+ no_load_rng ..................................... None
363
+ no_persist_layer_norm ........................... False
364
+ no_rope_freq .................................... None
365
+ no_save_optim ................................... None
366
+ no_save_rng ..................................... None
367
+ non_persistent_ckpt_type ........................ None
368
+ non_persistent_global_ckpt_dir .................. None
369
+ non_persistent_local_ckpt_algo .................. fully_parallel
370
+ non_persistent_local_ckpt_dir ................... None
371
+ non_persistent_save_interval .................... None
372
+ norm_epsilon .................................... 1e-05
373
+ normalization ................................... RMSNorm
374
+ num_attention_heads ............................. 32
375
+ num_channels .................................... 3
376
+ num_classes ..................................... 1000
377
+ num_dataset_builder_threads ..................... 1
378
+ num_distributed_optimizer_instances ............. 1
379
+ num_experts ..................................... 8
380
+ num_layers ...................................... 16
381
+ num_layers_at_end_in_bf16 ....................... 1
382
+ num_layers_at_start_in_bf16 ..................... 1
383
+ num_layers_per_virtual_pipeline_stage ........... None
384
+ num_query_groups ................................ 8
385
+ num_virtual_stages_per_pipeline_rank ............ None
386
+ num_workers ..................................... 2
387
+ object_storage_cache_path ....................... None
388
+ one_logger_async ................................ False
389
+ one_logger_project .............................. megatron-lm
390
+ one_logger_run_name ............................. None
391
+ onnx_safe ....................................... None
392
+ openai_gelu ..................................... False
393
+ optimizer ....................................... adam
394
+ optimizer_cpu_offload ........................... False
395
+ optimizer_offload_fraction ...................... 1.0
396
+ output_bert_embeddings .......................... False
397
+ overlap_cpu_optimizer_d2h_h2d ................... False
398
+ overlap_grad_reduce ............................. True
399
+ overlap_moe_expert_parallel_comm ................ False
400
+ overlap_p2p_comm ................................ False
401
+ overlap_p2p_comm_warmup_flush ................... False
402
+ overlap_param_gather ............................ True
403
+ overlap_param_gather_with_optimizer_step ........ False
404
+ override_opt_param_scheduler .................... False
405
+ params_dtype .................................... torch.bfloat16
406
+ patch_dim ....................................... 16
407
+ per_split_data_args_path ........................ None
408
+ perform_initialization .......................... True
409
+ pin_cpu_grads ................................... True
410
+ pin_cpu_params .................................. True
411
+ pipeline_model_parallel_comm_backend ............ None
412
+ pipeline_model_parallel_layout .................. None
413
+ pipeline_model_parallel_size .................... 1
414
+ position_embedding_type ......................... rope
415
+ pretrained_checkpoint ........................... None
416
+ profile ......................................... True
417
+ profile_ranks ................................... [0, 1, 2, 3, 4, 5, 6, 7]
418
+ profile_step_end ................................ 8
419
+ profile_step_start .............................. 6
420
+ q_lora_rank ..................................... None
421
+ qk_head_dim ..................................... 128
422
+ qk_l2_norm ...................................... False
423
+ qk_layernorm .................................... False
424
+ qk_pos_emb_head_dim ............................. 64
425
+ query_in_block_prob ............................. 0.1
426
+ rampup_batch_size ............................... None
427
+ rank ............................................ 0
428
+ recompute_granularity ........................... None
429
+ recompute_method ................................ None
430
+ recompute_modules ............................... None
431
+ recompute_num_layers ............................ None
432
+ record_memory_history ........................... False
433
+ relative_attention_max_distance ................. 128
434
+ relative_attention_num_buckets .................. 32
435
+ replication ..................................... False
436
+ replication_factor .............................. 2
437
+ replication_jump ................................ None
438
+ rerun_mode ...................................... validate_results
439
+ reset_attention_mask ............................ False
440
+ reset_position_ids .............................. False
441
+ result_rejected_tracker_filename ................ None
442
+ retriever_report_topk_accuracies ................ []
443
+ retriever_score_scaling ......................... False
444
+ retriever_seq_length ............................ 256
445
+ retro_add_retriever ............................. False
446
+ retro_attention_gate ............................ 1
447
+ retro_cyclic_train_iters ........................ None
448
+ retro_encoder_attention_dropout ................. 0.1
449
+ retro_encoder_hidden_dropout .................... 0.1
450
+ retro_encoder_layers ............................ 2
451
+ retro_num_neighbors ............................. 2
452
+ retro_num_retrieved_chunks ...................... 2
453
+ retro_project_dir ............................... None
454
+ retro_verify_neighbor_count ..................... True
455
+ reuse_grad_buf_for_mxfp8_param_ag ............... False
456
+ rope_scaling_factor ............................. 8.0
457
+ rope_type ....................................... None
458
+ rotary_base ..................................... 1000000
459
+ rotary_interleaved .............................. False
460
+ rotary_percent .................................. 1.0
461
+ rotary_scaling_factor ........................... 1.0
462
+ rotary_seq_len_interpolation_factor ............. None
463
+ run_workload_inspector_server ................... False
464
+ sample_rate ..................................... 1.0
465
+ save ............................................ None
466
+ save_interval ................................... None
467
+ save_retain_interval ............................ None
468
+ scatter_gather_tensors_in_pipeline .............. True
469
+ seed ............................................ 1234
470
+ seq_length ...................................... 4096
471
+ sequence_parallel ............................... False
472
+ sft ............................................. False
473
+ sft_tokenizer_prompt_format ..................... nemotron-h-aligned
474
+ sgd_momentum .................................... 0.9
475
+ short_seq_prob .................................. 0.1
476
+ skip_train ...................................... False
477
+ skipped_train_samples ........................... 0
478
+ spec ............................................ None
479
+ split ........................................... 100,0,0
480
+ squared_relu .................................... False
481
+ start_weight_decay .............................. 0.1
482
+ straggler_ctrlr_port ............................ 65535
483
+ straggler_minmax_count .......................... 1
484
+ suggested_communication_unit_size ............... None
485
+ swiglu .......................................... True
486
+ swin_backbone_type .............................. tiny
487
+ symmetric_ar_type ............................... None
488
+ te_rng_tracker .................................. True
489
+ tensor_model_parallel_size ...................... 1
490
+ tensorboard_dir ................................. None
491
+ tensorboard_log_interval ........................ 1
492
+ tensorboard_queue_size .......................... 1000
493
+ test_data_path .................................. None
494
+ test_mode ....................................... False
495
+ tiktoken_num_special_tokens ..................... 1000
496
+ tiktoken_pattern ................................ None
497
+ tiktoken_special_tokens ......................... None
498
+ timing_log_level ................................ 0
499
+ timing_log_option ............................... minmax
500
+ titles_data_path ................................ None
501
+ tokenizer_model ................................. None
502
+ tokenizer_type .................................. GPT2BPETokenizer
503
+ torch_fsdp2_reshard_after_forward ............... True
504
+ tp_comm_bootstrap_backend ....................... nccl
505
+ tp_comm_bulk_dgrad .............................. True
506
+ tp_comm_bulk_wgrad .............................. True
507
+ tp_comm_overlap ................................. False
508
+ tp_comm_overlap_ag .............................. True
509
+ tp_comm_overlap_cfg ............................. None
510
+ tp_comm_overlap_rs .............................. True
511
+ tp_comm_overlap_rs_dgrad ........................ False
512
+ tp_comm_split_ag ................................ True
513
+ tp_comm_split_rs ................................ True
514
+ train_data_path ................................. None
515
+ train_iters ..................................... 10
516
+ train_samples ................................... None
517
+ train_sync_interval ............................. None
518
+ transformer_impl ................................ transformer_engine
519
+ transformer_pipeline_model_parallel_size ........ 1
520
+ untie_embeddings_and_output_weights ............. True
521
+ use_checkpoint_args ............................. False
522
+ use_checkpoint_opt_param_scheduler .............. False
523
+ use_cpu_initialization .......................... None
524
+ use_custom_fsdp ................................. False
525
+ use_dist_ckpt ................................... True
526
+ use_dist_ckpt_deprecated ........................ False
527
+ use_distributed_optimizer ....................... True
528
+ use_flash_attn .................................. False
529
+ use_legacy_models ............................... False
530
+ use_mp_args_from_checkpoint_args ................ False
531
+ use_one_sent_docs ............................... False
532
+ use_persistent_ckpt_worker ...................... False
533
+ use_precision_aware_optimizer ................... False
534
+ use_pytorch_profiler ............................ False
535
+ use_ring_exchange_p2p ........................... False
536
+ use_rope_scaling ................................ False
537
+ use_rotary_position_embeddings .................. False
538
+ use_sharp ....................................... False
539
+ use_tokenizer_model_from_checkpoint_args ........ True
540
+ use_torch_fsdp2 ................................. False
541
+ use_torch_optimizer_for_cpu_offload ............. False
542
+ use_tp_pp_dp_mapping ............................ False
543
+ v_head_dim ...................................... 128
544
+ valid_data_path ................................. None
545
+ variable_seq_lengths ............................ False
546
+ virtual_pipeline_model_parallel_size ............ None
547
+ vision_backbone_type ............................ vit
548
+ vision_pretraining .............................. False
549
+ vision_pretraining_type ......................... classify
550
+ vocab_extra_ids ................................. 0
551
+ vocab_file ...................................... /workspace/data/wikitext/gpt2-vocab.json
552
+ vocab_size ...................................... None
553
+ wandb_exp_name ..................................
554
+ wandb_project ...................................
555
+ wandb_save_dir ..................................
556
+ weight_decay .................................... 0.1
557
+ weight_decay_incr_style ......................... constant
558
+ wgrad_deferral_limit ............................ 0
559
+ world_size ...................................... 8
560
+ yaml_cfg ........................................ None
561
+ -------------------- end of arguments ---------------------
562
+ INFO:megatron.core.num_microbatches_calculator:setting number of microbatches to constant 2
563
+ > building GPT2BPETokenizer tokenizer ...
564
+ > padded vocab (size: 50257) with 47 dummy tokens (new size: 50304)
565
+ WARNING:megatron.core.rerun_state_machine:RerunStateMachine initialized in mode RerunMode.VALIDATE_RESULTS
566
+ > initializing torch distributed ...
567
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
568
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
569
+ 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
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
+ /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
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
575
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
576
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
577
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
578
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
579
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
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 2 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
584
+ [Gloo] Rank 3 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
585
+ [Gloo] Rank 5 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
586
+ [Gloo] Rank 6 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
587
+ [Gloo] Rank 1 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
588
+ [Gloo] Rank 4 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 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 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 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[Gloo] Rank
608
+ 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
609
+ [rank2]:[W830 08:58:53.375778713 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
+ [rank4]:[W830 08:58:53.376827463 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.
611
+ [rank7]:[W830 08:58:53.376926331 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
+ [rank5]:[W830 08:58:53.377101205 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 08:58:53.377237313 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
+ [rank1]:[W830 08:58:53.377288427 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
+ [rank6]:[W830 08:58:53.377327904 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: 0.428 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 08:58:53.005670424 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: 3.724 seconds
625
+ time to initialize megatron (seconds): 9.407
626
+ [after megatron is initialized] datetime: 2026-08-30 08:58:59
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): 6720983040
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.self_attention.linear_qkv.weight
678
+ module.decoder.layers.14.pre_mlp_layernorm.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.pre_mlp_layernorm.weight
684
+ module.decoder.layers.13.self_attention.linear_qkv.weight
685
+ module.decoder.layers.13.mlp.router.weight
686
+ module.decoder.layers.13.self_attention.linear_proj.weight
687
+ module.decoder.layers.13.self_attention.linear_qkv.layer_norm_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_proj.weight
697
+ module.decoder.layers.11.self_attention.linear_qkv.layer_norm_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.pre_mlp_layernorm.weight
705
+ module.decoder.layers.10.self_attention.linear_qkv.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.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.self_attention.linear_qkv.weight
720
+ module.decoder.layers.7.pre_mlp_layernorm.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.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_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.self_attention.linear_proj.weight
758
+ module.decoder.layers.1.self_attention.linear_qkv.layer_norm_weight
759
+ module.decoder.layers.1.pre_mlp_layernorm.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_proj.weight
763
+ module.decoder.layers.0.self_attention.linear_qkv.weight
764
+ module.decoder.layers.0.pre_mlp_layernorm.weight
765
+ module.decoder.layers.0.self_attention.linear_qkv.layer_norm_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: 32
769
+ Params for bucket 1 (58720256 elements, 58720256 padded size):
770
+ module.decoder.layers.15.mlp.experts.linear_fc2.weight0
771
+ Params for bucket 2 (117440512 elements, 117440512 padded size):
772
+ module.decoder.layers.15.mlp.experts.linear_fc1.weight0
773
+ Params for bucket 3 (58720256 elements, 58720256 padded size):
774
+ module.decoder.layers.14.mlp.experts.linear_fc2.weight0
775
+ Params for bucket 4 (117440512 elements, 117440512 padded size):
776
+ module.decoder.layers.14.mlp.experts.linear_fc1.weight0
777
+ Params for bucket 5 (58720256 elements, 58720256 padded size):
778
+ module.decoder.layers.13.mlp.experts.linear_fc2.weight0
779
+ Params for bucket 6 (117440512 elements, 117440512 padded size):
780
+ module.decoder.layers.13.mlp.experts.linear_fc1.weight0
781
+ Params for bucket 7 (58720256 elements, 58720256 padded size):
782
+ module.decoder.layers.12.mlp.experts.linear_fc2.weight0
783
+ Params for bucket 8 (117440512 elements, 117440512 padded size):
784
+ module.decoder.layers.12.mlp.experts.linear_fc1.weight0
785
+ Params for bucket 9 (58720256 elements, 58720256 padded size):
786
+ module.decoder.layers.11.mlp.experts.linear_fc2.weight0
787
+ Params for bucket 10 (117440512 elements, 117440512 padded size):
788
+ module.decoder.layers.11.mlp.experts.linear_fc1.weight0
789
+ Params for bucket 11 (58720256 elements, 58720256 padded size):
790
+ module.decoder.layers.10.mlp.experts.linear_fc2.weight0
791
+ Params for bucket 12 (117440512 elements, 117440512 padded size):
792
+ module.decoder.layers.10.mlp.experts.linear_fc1.weight0
793
+ Params for bucket 13 (58720256 elements, 58720256 padded size):
794
+ module.decoder.layers.9.mlp.experts.linear_fc2.weight0
795
+ Params for bucket 14 (117440512 elements, 117440512 padded size):
796
+ module.decoder.layers.9.mlp.experts.linear_fc1.weight0
797
+ Params for bucket 15 (58720256 elements, 58720256 padded size):
798
+ module.decoder.layers.8.mlp.experts.linear_fc2.weight0
799
+ Params for bucket 16 (117440512 elements, 117440512 padded size):
800
+ module.decoder.layers.8.mlp.experts.linear_fc1.weight0
801
+ Params for bucket 17 (58720256 elements, 58720256 padded size):
802
+ module.decoder.layers.7.mlp.experts.linear_fc2.weight0
803
+ Params for bucket 18 (117440512 elements, 117440512 padded size):
804
+ module.decoder.layers.7.mlp.experts.linear_fc1.weight0
805
+ Params for bucket 19 (58720256 elements, 58720256 padded size):
806
+ module.decoder.layers.6.mlp.experts.linear_fc2.weight0
807
+ Params for bucket 20 (117440512 elements, 117440512 padded size):
808
+ module.decoder.layers.6.mlp.experts.linear_fc1.weight0
809
+ Params for bucket 21 (58720256 elements, 58720256 padded size):
810
+ module.decoder.layers.5.mlp.experts.linear_fc2.weight0
811
+ Params for bucket 22 (117440512 elements, 117440512 padded size):
812
+ module.decoder.layers.5.mlp.experts.linear_fc1.weight0
813
+ Params for bucket 23 (58720256 elements, 58720256 padded size):
814
+ module.decoder.layers.4.mlp.experts.linear_fc2.weight0
815
+ Params for bucket 24 (117440512 elements, 117440512 padded size):
816
+ module.decoder.layers.4.mlp.experts.linear_fc1.weight0
817
+ Params for bucket 25 (58720256 elements, 58720256 padded size):
818
+ module.decoder.layers.3.mlp.experts.linear_fc2.weight0
819
+ Params for bucket 26 (117440512 elements, 117440512 padded size):
820
+ module.decoder.layers.3.mlp.experts.linear_fc1.weight0
821
+ Params for bucket 27 (58720256 elements, 58720256 padded size):
822
+ module.decoder.layers.2.mlp.experts.linear_fc2.weight0
823
+ Params for bucket 28 (117440512 elements, 117440512 padded size):
824
+ module.decoder.layers.2.mlp.experts.linear_fc1.weight0
825
+ Params for bucket 29 (58720256 elements, 58720256 padded size):
826
+ module.decoder.layers.1.mlp.experts.linear_fc2.weight0
827
+ Params for bucket 30 (117440512 elements, 117440512 padded size):
828
+ module.decoder.layers.1.mlp.experts.linear_fc1.weight0
829
+ Params for bucket 31 (58720256 elements, 58720256 padded size):
830
+ module.decoder.layers.0.mlp.experts.linear_fc2.weight0
831
+ Params for bucket 32 (117440512 elements, 117440512 padded size):
832
+ module.decoder.layers.0.mlp.experts.linear_fc1.weight0
833
+ 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 0x7815b8c1df40>, config_logger_dir='')
834
+ INFO:megatron.core.optimizer_param_scheduler:> learning rate decay style: cosine
835
+ [after model, optimizer, and learning rate scheduler are built] datetime: 2026-08-30 08:59:02
836
+ > building train, validation, and test datasets ...
837
+ > datasets target sizes (minimum size):
838
+ train: 160
839
+ validation: 0
840
+ test: 0
841
+ INFO:megatron.core.datasets.blended_megatron_dataset_config:Let split_matrix = [(0, 1.0), None, None]
842
+ > building train, validation, and test datasets for GPT ...
843
+ INFO:megatron.core.datasets.blended_megatron_dataset_builder:Building GPTDataset splits with sizes=(160, 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 0x7815b9b9f800>, 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)
844
+ INFO:megatron.core.datasets.indexed_dataset:Load the _IndexReader from /workspace/data/wikitext/wikitext103_gpt2_text_document.idx
845
+ INFO:megatron.core.datasets.indexed_dataset: Extract the sequence lengths
846
+ INFO:megatron.core.datasets.indexed_dataset: Extract the sequence pointers
847
+ INFO:megatron.core.datasets.indexed_dataset: Extract the document indices
848
+ INFO:megatron.core.datasets.indexed_dataset:> total number of sequences: 1165029
849
+ INFO:megatron.core.datasets.indexed_dataset:> total number of documents: 1801350
850
+ INFO:megatron.core.datasets.gpt_dataset:Load the GPTDataset train indices
851
+ INFO:megatron.core.datasets.gpt_dataset: Load the document index from 68684a515e118253a8572c4517cf2ffe-GPTDataset-train-document_index.npy
852
+ INFO:megatron.core.datasets.gpt_dataset: Load the sample index from 68684a515e118253a8572c4517cf2ffe-GPTDataset-train-sample_index.npy
853
+ INFO:megatron.core.datasets.gpt_dataset: Load the shuffle index from 68684a515e118253a8572c4517cf2ffe-GPTDataset-train-shuffle_index.npy
854
+ INFO:megatron.core.datasets.gpt_dataset:> total number of samples: 29073
855
+ > finished creating GPT datasets ...
856
+ [after dataloaders are built] datetime: 2026-08-30 08:59:03
857
+ done with setup ...
858
+ training ...
859
+ Overwriting rerun_state_machine.current_iteration from -1 to 0...
860
+ (min, max) time across ranks (ms):
861
+ model-and-optimizer-setup ......................: (3267.73, 3269.75)
862
+ train/valid/test-data-iterators-setup ..........: (92.04, 116.78)
863
+ [before the start of training step] datetime: 2026-08-30 08:59:03
864
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
865
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
866
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
867
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
868
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
869
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
870
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
871
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
872
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
873
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
874
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
875
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
876
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
877
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
878
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
879
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
880
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
881
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
882
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
883
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
884
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
885
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
886
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
887
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
888
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
889
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
890
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
891
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
892
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
893
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
894
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
895
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
896
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
897
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
898
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
899
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
900
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
901
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
902
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
903
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
904
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
905
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
906
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
907
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
908
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
909
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
910
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
911
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
912
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
913
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
914
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
915
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
916
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
917
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
918
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
919
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
920
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
921
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
922
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
923
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
924
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
925
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
926
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
927
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
928
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
929
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
930
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
931
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
932
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
933
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
934
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
935
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
936
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
937
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
938
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
939
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
940
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
941
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
942
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
943
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
944
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
945
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
946
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
947
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
948
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
949
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
950
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
951
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
952
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
953
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
954
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
955
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
956
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
957
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
958
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
959
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
960
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
961
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
962
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
963
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
964
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
965
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
966
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
967
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
968
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
969
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
970
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
971
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
972
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
973
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
974
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
975
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
976
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
977
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
978
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
979
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
980
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
981
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
982
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
983
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
984
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
985
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
986
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
987
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
988
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
989
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
990
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
991
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
992
+ [2026-08-30 08:59:32] iteration 1/ 10 | consumed samples: 16 | elapsed time per iteration (ms): 29209.8 | throughput per GPU (TFLOP/s/GPU): 11.4 | learning rate: 2.000000E-05 | global batch size: 16 | lm loss: 1.104376E+01 | loss scale: 1.0 | grad norm: 35.710 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 13312 | min safe capacity: 1.6250 |
993
+ Number of parameters in transformer block in billions: 23.22
994
+ Number of parameters in embedding layers in billions: 0.41
995
+ Total number of parameters in billions: 23.63
996
+ Number of parameters in most loaded shard in billions: 23.6320
997
+ Theoretical memory footprints: weight and optimizer=169029.43 MB
998
+ [Rank 0] (after 1 iterations) memory (MB) | allocated: 58364.49658203125 | max allocated: 85653.79931640625 | reserved: 87692.0 | max reserved: 87692.0
999
+ [2026-08-30 08:59:33] iteration 2/ 10 | consumed samples: 32 | elapsed time per iteration (ms): 831.9 | throughput per GPU (TFLOP/s/GPU): 400.8 | learning rate: 4.000000E-05 | global batch size: 16 | lm loss: 1.104935E+01 | loss scale: 1.0 | grad norm: 37.626 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 13312 | min safe capacity: 1.6250 |
1000
+ [2026-08-30 08:59:34] iteration 3/ 10 | consumed samples: 48 | elapsed time per iteration (ms): 876.5 | throughput per GPU (TFLOP/s/GPU): 380.3 | learning rate: 6.000000E-05 | global batch size: 16 | lm loss: 9.679840E+00 | loss scale: 1.0 | grad norm: 42.123 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 13312 | min safe capacity: 1.6250 |
1001
+ Capture CUDA graph for training!!!
1002
+ CUDA graph capture done!!!
1003
+ [2026-08-30 08:59:36] iteration 4/ 10 | consumed samples: 64 | elapsed time per iteration (ms): 1943.8 | throughput per GPU (TFLOP/s/GPU): 171.5 | learning rate: 8.000000E-05 | global batch size: 16 | lm loss: 1.120076E+01 | loss scale: 1.0 | grad norm: 89.030 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 13312 | min safe capacity: 1.6250 |
1004
+ [2026-08-30 08:59:36] iteration 5/ 10 | consumed samples: 80 | elapsed time per iteration (ms): 878.6 | throughput per GPU (TFLOP/s/GPU): 379.4 | learning rate: 1.000000E-04 | global batch size: 16 | lm loss: 1.016875E+01 | loss scale: 1.0 | grad norm: 15.418 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 13312 | min safe capacity: 1.6250 |
1005
+ [2026-08-30 08:59:37] iteration 6/ 10 | consumed samples: 96 | elapsed time per iteration (ms): 913.6 | throughput per GPU (TFLOP/s/GPU): 364.9 | learning rate: 7.750000E-05 | global batch size: 16 | lm loss: 1.012656E+01 | loss scale: 1.0 | grad norm: 43.758 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 13312 | min safe capacity: 1.6250 |
1006
+ [2026-08-30 08:59:41] iteration 7/ 10 | consumed samples: 112 | elapsed time per iteration (ms): 3670.5 | throughput per GPU (TFLOP/s/GPU): 90.8 | learning rate: 3.250000E-05 | global batch size: 16 | lm loss: 9.868882E+00 | loss scale: 1.0 | grad norm: 8.163 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 13312 | min safe capacity: 1.6250 |
1007
+ [2026-08-30 08:59:42] iteration 8/ 10 | consumed samples: 128 | elapsed time per iteration (ms): 903.2 | throughput per GPU (TFLOP/s/GPU): 369.1 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 9.027750E+00 | loss scale: 1.0 | grad norm: 7.091 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 13312 | min safe capacity: 1.6250 |
1008
+ Capture range started in the application.
1009
+ Capture range ended in the application.
1010
+ Generating '/tmp/nsys-report-be84.qdstrm'
1011
+
1012
+
1013
+ [2026-08-30 09:00:25] iteration 9/ 10 | consumed samples: 144 | elapsed time per iteration (ms): 42930.6 | throughput per GPU (TFLOP/s/GPU): 7.8 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.669333E+00 | loss scale: 1.0 | grad norm: 6.362 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 13312 | min safe capacity: 1.6250 |
1014
+ [2026-08-30 09:00:26] iteration 10/ 10 | consumed samples: 160 | elapsed time per iteration (ms): 904.9 | throughput per GPU (TFLOP/s/GPU): 368.4 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.479151E+00 | loss scale: 1.0 | grad norm: 5.889 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 13312 | min safe capacity: 1.6250 |
1015
+ [after training is done] datetime: 2026-08-30 09:00:26
1016
+ [rank0]:[W830 09:00:27.758684095 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())
1017
+ [rank2]:[W830 09:00:28.004481366 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())
1018
+ [rank1]:[W830 09:00:28.017292076 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())
1019
+ [rank6]:[W830 09:00:29.287991418 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())
1020
+ [rank5]:[W830 09:00:29.403914015 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())
1021
+ [rank4]:[W830 09:00:29.405887893 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())
1022
+ [rank7]:[W830 09:00:29.560216757 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())
1023
+ [rank3]:[W830 09:00:29.562397666 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())
1024
+ [rank0]:[W830 09:00:31.932837925 observer.cpp:116] Warning: Leaked callback handle: 1 (function operator())
1025
+ pure virtual method called
1026
+ terminate called without an active exception
1027
+ [rank1]:[W830 09:00:31.020246606 observer.cpp:116] Warning: Leaked callback handle: 1 (function operator())
1028
+ pure virtual method called
1029
+ terminate called without an active exception
1030
+ [rank5]:[W830 09:00:31.089187142 observer.cpp:116] Warning: Leaked callback handle: 1 (function operator())
1031
+ pure virtual method called
1032
+ terminate called without an active exception
1033
+ Collection stop is not allowed in this state.
1034
+ W0830 09:00:32.275000 903525 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 903637 closing signal SIGTERM
1035
+ W0830 09:00:32.276000 903525 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 903638 closing signal SIGTERM
1036
+ W0830 09:00:32.276000 903525 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 903639 closing signal SIGTERM
1037
+ W0830 09:00:32.277000 903525 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 903640 closing signal SIGTERM
1038
+ W0830 09:00:32.278000 903525 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 903641 closing signal SIGTERM
1039
+ W0830 09:00:32.279000 903525 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 903642 closing signal SIGTERM
1040
+ W0830 09:00:32.280000 903525 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 903643 closing signal SIGTERM
1041
+ E0830 09:00:33.875000 903525 torch/distributed/elastic/multiprocessing/api.py:874] failed (exitcode: -6) local_rank: 0 (pid: 903636) of binary: /usr/bin/python
1042
+ Traceback (most recent call last):
1043
+ File "/usr/local/bin/torchrun", line 8, in <module>
1044
+ sys.exit(main())
1045
+ ^^^^^^
1046
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/multiprocessing/errors/__init__.py", line 355, in wrapper
1047
+ return f(*args, **kwargs)
1048
+ ^^^^^^^^^^^^^^^^^^
1049
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/run.py", line 892, in main
1050
+ run(args)
1051
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/run.py", line 883, in run
1052
+ elastic_launch(
1053
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/launcher/api.py", line 139, in __call__
1054
+ return launch_agent(self._config, self._entrypoint, list(args))
1055
+ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1056
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/launcher/api.py", line 270, in launch_agent
1057
+ raise ChildFailedError(
1058
+ torch.distributed.elastic.multiprocessing.errors.ChildFailedError:
1059
+ ============================================================
1060
+ pretrain_gpt.py FAILED
1061
+ ------------------------------------------------------------
1062
+ Failures:
1063
+ <NO_OTHER_FAILURES>
1064
+ ------------------------------------------------------------
1065
+ Root Cause (first observed failure):
1066
+ [0]:
1067
+ time : 2026-08-30_09:00:32
1068
+ host : sp-aiinfra-interactive-aghosh-7jtjl-8xb200-moe-cg-0
1069
+ rank : 0 (local_rank: 0)
1070
+ exitcode : -6 (pid: 903636)
1071
+ error_file: <N/A>
1072
+ traceback : Signal 6 (SIGABRT) received by PID 903636
1073
+ ============================================================
1074
+ Generated:
1075
+ /workspace/tiera-rerun-fixed-moonep/ultraep/tierbtrace/tierbtrace_gbs16/G1__ultraep__matched__seq4096__gbs16__cap2p40625/nsys/G1__ultraep__matched__seq4096__gbs16__cap2p40625.nsys-rep
tiera-rerun-fixed-moonep/ultraep/tierbtrace/tierbtrace_gbs16/G1__ultraep__matched__seq4096__gbs16__cap2p40625/nsys/verify_trace_ranks.log ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ report : /workspace/tiera-rerun-fixed-moonep/ultraep/tierbtrace/tierbtrace_gbs16/G1__ultraep__matched__seq4096__gbs16__cap2p40625/nsys/G1__ultraep__matched__seq4096__gbs16__cap2p40625.nsys-rep
2
+ kernels : 132833
3
+ devices : 8 of 8 -> [0, 1, 2, 3, 4, 5, 6, 7]
4
+ processes: 1
5
+ markers : 26624 ultraep_trace__ kernels on devices [0, 1, 2, 3, 4, 5, 6, 7]
6
+ PASS: all ranks present and backend markers emitted
tiera-rerun-fixed-moonep/ultraep/tierbtrace/tierbtrace_gbs16/G2__ultraep__matched__seq4096__gbs16__cap1p08203125/nsys/G2__ultraep__matched__seq4096__gbs16__cap1p08203125.nsys-rep ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2e93cd9399483376f45427796547ecc4d9572d07d300ac33e4e39f7fdacc4b84
3
+ size 10947387
tiera-rerun-fixed-moonep/ultraep/tierbtrace/tierbtrace_gbs16/G2__ultraep__matched__seq4096__gbs16__cap1p08203125/nsys/run.env ADDED
@@ -0,0 +1,70 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ backend=ultraep
2
+ config=G2
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=3584
11
+ num_experts=32
12
+ topk=8
13
+ experts_per_rank=4
14
+ expert_params=44040192
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=32768
40
+ mean_tokens_per_expert=8192
41
+ slot_policy=matched
42
+ slots_per_rank=4
43
+ capacity_mode=static
44
+ capacity=1.08203125
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=10
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=1
61
+ kernel_trace=1
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/tierbtrace/tierbtrace_gbs16/G2__ultraep__matched__seq4096__gbs16__cap1p08203125/nsys/train.log ADDED
@@ -0,0 +1,1132 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ==============================================================
2
+ G2 / ultraep (slots/rank=4 policy=matched)
3
+ E=32 K=8 H'=3584 m=4
4
+ L=16 seq=4096 gbs=16 (2 microbatches)
5
+ cap_mode=static cap=1.08203125 cg=1 recompute=none
6
+ memory_dump=none snapshots=/workspace/tiera-rerun-fixed-moonep/ultraep/tierbtrace/tierbtrace_gbs16/G2__ultraep__matched__seq4096__gbs16__cap1p08203125/nsys/memory_snapshots/snapshot.pickle
7
+ repo=/workspace/Megatron-UltraEP @ 03d57ef7
8
+ ==============================================================
9
+ W0830 09:01:11.777000 910350 torch/distributed/run.py:766]
10
+ W0830 09:01:11.777000 910350 torch/distributed/run.py:766] *****************************************
11
+ W0830 09:01:11.777000 910350 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 09:01:11.777000 910350 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
+ 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
17
+ WARNING: overriding default arguments for tokenizer_type:GPT2BPETokenizer with tokenizer_type:GPT2BPETokenizer
18
+ Number of virtual stages per pipeline stage: None
19
+ accumulate and all-reduce gradients in fp32 for bfloat16 data type.
20
+ using torch.bfloat16 for parameters ...
21
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
22
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
23
+ ------------------------ arguments ------------------------
24
+ account_for_embedding_in_pipeline_split ......... False
25
+ account_for_loss_in_pipeline_split .............. False
26
+ accumulate_allreduce_grads_in_fp32 .............. True
27
+ adam_beta1 ...................................... 0.9
28
+ adam_beta2 ...................................... 0.999
29
+ adam_eps ........................................ 1e-08
30
+ add_bias_linear ................................. False
31
+ add_position_embedding .......................... False
32
+ add_qkv_bias .................................... False
33
+ adlr_autoresume ................................. False
34
+ adlr_autoresume_interval ........................ 1000
35
+ align_grad_reduce ............................... True
36
+ align_param_gather .............................. False
37
+ app_tag_run_name ................................ None
38
+ app_tag_run_version ............................. 0.0.0
39
+ apply_layernorm_1p .............................. False
40
+ apply_query_key_layer_scaling ................... False
41
+ apply_residual_connection_post_layernorm ........ False
42
+ apply_rope_fusion ............................... True
43
+ async_save ...................................... None
44
+ async_tensor_model_parallel_allreduce ........... True
45
+ attention_backend ............................... AttnBackend.auto
46
+ attention_dropout ............................... 0.0
47
+ attention_softmax_in_fp32 ....................... False
48
+ auto_detect_ckpt_format ......................... False
49
+ barrier_with_L1_time ............................ True
50
+ bert_binary_head ................................ True
51
+ bert_embedder_type .............................. megatron
52
+ bert_load ....................................... None
53
+ bf16 ............................................ True
54
+ bias_dropout_fusion ............................. True
55
+ bias_gelu_fusion ................................ False
56
+ bias_swiglu_fusion .............................. True
57
+ biencoder_projection_dim ........................ 0
58
+ biencoder_shared_query_context_model ............ False
59
+ block_data_path ................................. None
60
+ calc_ft_timeouts ................................ False
61
+ calculate_per_token_loss ........................ False
62
+ check_for_large_grads ........................... False
63
+ check_for_nan_in_loss_and_grad .................. False
64
+ check_for_spiky_loss ............................ False
65
+ check_weight_hash_across_dp_replicas_interval ... None
66
+ ckpt_assume_constant_structure .................. False
67
+ ckpt_convert_format ............................. None
68
+ ckpt_convert_save ............................... None
69
+ ckpt_convert_update_legacy_dist_opt_format ...... False
70
+ ckpt_format ..................................... torch_dist
71
+ ckpt_fully_parallel_load ........................ False
72
+ ckpt_fully_parallel_save ........................ True
73
+ ckpt_fully_parallel_save_deprecated ............. False
74
+ ckpt_step ....................................... None
75
+ classes_fraction ................................ 1.0
76
+ clip_grad ....................................... 1.0
77
+ clone_scatter_output_in_embedding ............... True
78
+ config_logger_dir ...............................
79
+ consumed_train_samples .......................... 0
80
+ consumed_valid_samples .......................... 0
81
+ context_parallel_size ........................... 1
82
+ cp_comm_type .................................... ['p2p']
83
+ create_attention_mask_in_dataloader ............. True
84
+ cross_entropy_fusion_impl ....................... native
85
+ cross_entropy_loss_fusion ....................... False
86
+ cuda_graph_scope ................................ full_iteration
87
+ cuda_graph_warmup_steps ......................... 3
88
+ data_args_path .................................. None
89
+ data_cache_path ................................. None
90
+ data_parallel_random_init ....................... False
91
+ data_parallel_sharding_strategy ................. no_shard
92
+ data_parallel_size .............................. 8
93
+ data_path ....................................... ['/workspace/data/wikitext/wikitext103_gpt2_text_document']
94
+ data_per_class_fraction ......................... 1.0
95
+ data_sharding ................................... True
96
+ dataloader_type ................................. single
97
+ ddp_average_in_collective ....................... False
98
+ ddp_bucket_size ................................. None
99
+ ddp_num_buckets ................................. None
100
+ ddp_pad_buckets_for_high_nccl_busbw ............. False
101
+ decoder_first_pipeline_num_layers ............... None
102
+ decoder_last_pipeline_num_layers ................ None
103
+ decoder_num_layers .............................. None
104
+ decoder_seq_length .............................. None
105
+ decoupled_lr .................................... None
106
+ decoupled_min_lr ................................ None
107
+ decrease_batch_size_if_needed ................... False
108
+ defer_embedding_wgrad_compute ................... False
109
+ delay_wgrad_compute ............................. False
110
+ deprecated_use_mcore_models ..................... True
111
+ deterministic_mode .............................. False
112
+ dino_bottleneck_size ............................ 256
113
+ dino_freeze_last_layer .......................... 1
114
+ dino_head_hidden_size ........................... 2048
115
+ dino_local_crops_number ......................... 10
116
+ dino_local_img_size ............................. 96
117
+ dino_norm_last_layer ............................ False
118
+ dino_teacher_temp ............................... 0.07
119
+ dino_warmup_teacher_temp ........................ 0.04
120
+ dino_warmup_teacher_temp_epochs ................. 30
121
+ disable_bf16_reduced_precision_matmul ........... False
122
+ disable_mamba_mem_eff_path ...................... False
123
+ disable_straggler_on_startup .................... False
124
+ dist_ckpt_format_deprecated ..................... None
125
+ dist_ckpt_strictness ............................ assume_ok_unexpected
126
+ distribute_saved_activations .................... False
127
+ distributed_backend ............................. nccl
128
+ distributed_timeout_minutes ..................... 10
129
+ embedding_init_method_std ....................... None
130
+ embedding_path .................................. None
131
+ empty_unused_memory_level ....................... 0
132
+ enable_cuda_graph ............................... True
133
+ enable_experimental ............................. False
134
+ enable_ft_package ............................... False
135
+ enable_gloo_process_groups ...................... True
136
+ enable_msc ...................................... True
137
+ enable_one_logger ............................... True
138
+ encoder_num_layers .............................. 16
139
+ encoder_seq_length .............................. 4096
140
+ end_weight_decay ................................ 0.1
141
+ eod_mask_loss ................................... False
142
+ error_injection_rate ............................ 0
143
+ error_injection_type ............................ transient_error
144
+ eval_interval ................................... 1000000
145
+ eval_iters ...................................... 0
146
+ evidence_data_path .............................. None
147
+ exit_duration_in_mins ........................... None
148
+ exit_interval ................................... None
149
+ exit_on_missing_checkpoint ...................... False
150
+ exit_signal_handler ............................. False
151
+ exp_avg_dtype ................................... torch.float32
152
+ exp_avg_sq_dtype ................................ torch.float32
153
+ expert_model_parallel_size ...................... 8
154
+ expert_tensor_parallel_size ..................... 1
155
+ export_force_local_attention .................... False
156
+ export_kd_cfg ................................... None
157
+ export_kd_teacher_ckpt_format ................... None
158
+ export_kd_teacher_load .......................... None
159
+ export_kv_cache_quant ........................... False
160
+ export_legacy_megatron .......................... False
161
+ export_model_type ............................... GPTModel
162
+ export_moe_apply_probs_on_input ................. False
163
+ export_qk_l2_norm ............................... False
164
+ export_quant_cfg ................................ None
165
+ export_real_quant_cfg ........................... None
166
+ export_te_mcore_model ........................... False
167
+ external_cuda_graph ............................. False
168
+ ffn_hidden_size ................................. 14336
169
+ finetune ........................................ False
170
+ finetune_data_split ............................. train
171
+ finetune_hf_dataset ............................. None
172
+ first_last_layers_bf16 .......................... False
173
+ flash_decode .................................... False
174
+ fp16 ............................................ False
175
+ fp16_lm_cross_entropy ........................... False
176
+ fp32_residual_connection ........................ False
177
+ fp8 ............................................. e4m3
178
+ fp8_amax_compute_algo ........................... most_recent
179
+ fp8_amax_history_len ............................ 1
180
+ fp8_interval .................................... 1
181
+ fp8_margin ...................................... 0
182
+ fp8_param_gather ................................ False
183
+ fp8_recipe ...................................... mxfp8
184
+ fp8_wgrad ....................................... True
185
+ fsdp_double_buffer .............................. False
186
+ global_batch_size ............................... 16
187
+ grad_reduce_in_bf16 ............................. False
188
+ gradient_accumulation_fusion .................... True
189
+ gradient_reduce_div_fusion ...................... True
190
+ group_query_attention ........................... True
191
+ head_lr_mult .................................... 1.0
192
+ heterogeneous_layers_config_encoded_json ........ None
193
+ heterogeneous_layers_config_path ................ None
194
+ hidden_dropout .................................. 0.0
195
+ hidden_size ..................................... 4096
196
+ hierarchical_context_parallel_sizes ............. None
197
+ high_priority_stream_groups ..................... []
198
+ hybrid_attention_ratio .......................... 0.0
199
+ hybrid_mlp_ratio ................................ 0.0
200
+ hybrid_override_pattern ......................... None
201
+ hysteresis ...................................... 2
202
+ ict_head_size ................................... None
203
+ ict_load ........................................ None
204
+ img_h ........................................... 224
205
+ img_w ........................................... 224
206
+ indexer_batch_size .............................. 128
207
+ indexer_log_interval ............................ 1000
208
+ inference_batch_times_seqlen_threshold .......... -1
209
+ inference_dynamic_batching ...................... False
210
+ inference_dynamic_batching_buffer_guaranteed_fraction 0.2
211
+ inference_dynamic_batching_buffer_overflow_factor None
212
+ inference_dynamic_batching_buffer_size_gb ....... 40.0
213
+ inference_dynamic_batching_chunk_size ........... 256
214
+ inference_dynamic_batching_max_requests_override None
215
+ inference_dynamic_batching_max_tokens_override .. None
216
+ inference_dynamic_batching_num_cuda_graphs ...... 16
217
+ inference_max_batch_size ........................ 8
218
+ inference_max_seq_length ........................ 2560
219
+ inference_rng_tracker ........................... False
220
+ init_method_std ................................. 0.01
221
+ init_method_xavier_uniform ...................... False
222
+ init_model_with_meta_device ..................... False
223
+ initial_loss_scale .............................. 4294967296
224
+ inprocess_active_world_size ..................... 8
225
+ inprocess_barrier_timeout ....................... 120
226
+ inprocess_completion_timeout .................... 120
227
+ inprocess_empty_cuda_cache ...................... False
228
+ inprocess_granularity ........................... node
229
+ inprocess_hard_timeout .......................... 90
230
+ inprocess_heartbeat_interval .................... 30
231
+ inprocess_heartbeat_timeout ..................... 60
232
+ inprocess_last_call_wait ........................ 1
233
+ inprocess_max_iterations ........................ None
234
+ inprocess_monitor_process_interval .............. 1.0
235
+ inprocess_monitor_thread_interval ............... 1.0
236
+ inprocess_progress_watchdog_interval ............ 1.0
237
+ inprocess_restart ............................... False
238
+ inprocess_soft_timeout .......................... 60
239
+ inprocess_termination_grace_time ................ 1
240
+ is_hybrid_model ................................. False
241
+ iter_per_epoch .................................. 1250
242
+ iterations_to_skip .............................. []
243
+ keep_fp8_transpose_cache_when_using_custom_fsdp . False
244
+ kitchen_config_file ............................. None
245
+ kitchen_recipe_number ........................... None
246
+ kv_channels ..................................... 128
247
+ kv_lora_rank .................................... 32
248
+ lazy_mpu_init ................................... None
249
+ load ............................................ None
250
+ load_main_params_from_ckpt ...................... None
251
+ load_model_opt_format ........................... False
252
+ local_rank ...................................... 0
253
+ log_energy ...................................... False
254
+ log_interval .................................... 1
255
+ log_loss_scale_to_tensorboard ................... True
256
+ log_memory_to_tensorboard ....................... False
257
+ log_num_zeros_in_grad ........................... False
258
+ log_params_norm ................................. False
259
+ log_progress .................................... False
260
+ log_straggler ................................... False
261
+ log_throughput .................................. True
262
+ log_timers_to_tensorboard ....................... False
263
+ log_validation_ppl_to_tensorboard ............... False
264
+ log_world_size_to_tensorboard ................... False
265
+ logging_level ................................... None
266
+ loss_scale ...................................... None
267
+ loss_scale_window ............................... 1000
268
+ lr .............................................. 0.0001
269
+ lr_decay_iters .................................. 8
270
+ lr_decay_samples ................................ None
271
+ lr_decay_style .................................. cosine
272
+ lr_warmup_fraction .............................. None
273
+ lr_warmup_init .................................. 0.0
274
+ lr_warmup_iters ................................. 5
275
+ lr_warmup_samples ............................... 0
276
+ lr_wsd_decay_iters .............................. None
277
+ lr_wsd_decay_samples ............................ None
278
+ lr_wsd_decay_style .............................. exponential
279
+ main_grads_dtype ................................ torch.float32
280
+ main_params_dtype ............................... torch.float32
281
+ make_vocab_size_divisible_by .................... 128
282
+ mamba_head_dim .................................. 64
283
+ mamba_num_groups ................................ 8
284
+ mamba_num_heads ................................. None
285
+ mamba_state_dim ................................. 128
286
+ manual_gc ....................................... False
287
+ manual_gc_eval .................................. True
288
+ manual_gc_interval .............................. 0
289
+ mask_factor ..................................... 1.0
290
+ mask_prob ....................................... 0.15
291
+ mask_type ....................................... random
292
+ masked_softmax_fusion ........................... False
293
+ max_position_embeddings ......................... 32768
294
+ max_tokens_to_oom ............................... 12000
295
+ memory_snapshot_path ............................ snapshot.pickle
296
+ merge_file ...................................... /workspace/data/wikitext/gpt2-merges.txt
297
+ micro_batch_size ................................ 1
298
+ microbatch_group_size_per_vp_stage .............. None
299
+ mid_level_dataset_surplus ....................... 0.005
300
+ min_loss_scale .................................. 1.0
301
+ min_lr .......................................... 1e-05
302
+ mlp_chunks_for_prefill .......................... 1
303
+ mmap_bin_files .................................. True
304
+ mock_data ....................................... False
305
+ moe_apply_probs_on_input ........................ False
306
+ moe_aux_loss_coeff .............................. 0.01
307
+ moe_capacity_monitor ............................ True
308
+ moe_deepep_num_sms .............................. 20
309
+ moe_echo_enable_random_offloading ............... False
310
+ moe_echo_expert_dispatch_overlap ................ False
311
+ moe_echo_expert_dispatcher_type ................. hybridep
312
+ moe_echo_recompute_expert_dispatch .............. False
313
+ moe_enable_deepep ............................... False
314
+ moe_enable_echo ................................. False
315
+ moe_enable_hybridep ............................. True
316
+ moe_enable_ultraep .............................. True
317
+ moe_expert_capacity_factor ...................... None
318
+ moe_extended_tp ................................. False
319
+ moe_ffn_hidden_size ............................. 3584
320
+ moe_grouped_gemm ................................ True
321
+ moe_hybridep_num_sms ............................ 16
322
+ moe_input_jitter_eps ............................ None
323
+ moe_layer_freq .................................. 1
324
+ moe_layer_recompute ............................. False
325
+ moe_num_echo_experts ............................ None
326
+ moe_num_redundant_experts_per_rank .............. 4
327
+ moe_pad_expert_input_to_capacity ................ False
328
+ moe_per_layer_logging ........................... False
329
+ moe_permute_fusion .............................. False
330
+ moe_received_token_capacity ..................... 1.08203125
331
+ moe_router_bias_update_rate ..................... 0.001
332
+ moe_router_dtype ................................ fp32
333
+ moe_router_enable_expert_bias ................... False
334
+ moe_router_force_load_balancing ................. False
335
+ moe_router_group_topk ........................... None
336
+ moe_router_load_balancing_type .................. aux_loss
337
+ moe_router_num_groups ........................... None
338
+ moe_router_padding_for_fp8 ...................... False
339
+ moe_router_pre_softmax .......................... False
340
+ moe_router_score_function ....................... softmax
341
+ moe_router_topk ................................. 8
342
+ moe_router_topk_scaling_factor .................. None
343
+ moe_shared_expert_intermediate_size ............. None
344
+ moe_shared_expert_overlap ....................... False
345
+ moe_token_dispatcher_type ....................... flex
346
+ moe_token_drop_policy ........................... probs
347
+ moe_ultraep_max_microbatches .................... None
348
+ moe_ultraep_overlap_weight_sync ................. True
349
+ moe_ultraep_weight_sync_join .................... before_dispatch
350
+ moe_upcycling_granularity ....................... 1
351
+ moe_use_device_initiated_grouped_gemm ........... True
352
+ moe_use_legacy_grouped_gemm ..................... False
353
+ moe_use_upcycling ............................... False
354
+ moe_z_loss_coeff ................................ None
355
+ mrope_section ................................... None
356
+ mscale .......................................... 1.0
357
+ mscale_all_dim .................................. 0.0
358
+ mtp_loss_scaling_factor ......................... 0.1
359
+ mtp_num_layers .................................. None
360
+ multi_latent_attention .......................... False
361
+ nccl_all_reduce_for_prefill ..................... False
362
+ nccl_communicator_config_path ................... None
363
+ nccl_ub ......................................... False
364
+ no_load_optim ................................... None
365
+ no_load_rng ..................................... None
366
+ no_persist_layer_norm ........................... False
367
+ no_rope_freq .................................... None
368
+ no_save_optim ................................... None
369
+ no_save_rng ..................................... None
370
+ non_persistent_ckpt_type ........................ None
371
+ non_persistent_global_ckpt_dir .................. None
372
+ non_persistent_local_ckpt_algo .................. fully_parallel
373
+ non_persistent_local_ckpt_dir ................... None
374
+ non_persistent_save_interval .................... None
375
+ norm_epsilon .................................... 1e-05
376
+ normalization ................................... RMSNorm
377
+ num_attention_heads ............................. 32
378
+ num_channels .................................... 3
379
+ num_classes ..................................... 1000
380
+ num_dataset_builder_threads ..................... 1
381
+ num_distributed_optimizer_instances ............. 1
382
+ num_experts ..................................... 32
383
+ num_layers ...................................... 16
384
+ num_layers_at_end_in_bf16 ....................... 1
385
+ num_layers_at_start_in_bf16 ..................... 1
386
+ num_layers_per_virtual_pipeline_stage ........... None
387
+ num_query_groups ................................ 8
388
+ num_virtual_stages_per_pipeline_rank ............ None
389
+ num_workers ..................................... 2
390
+ object_storage_cache_path ....................... None
391
+ one_logger_async ................................ False
392
+ one_logger_project .............................. megatron-lm
393
+ one_logger_run_name ............................. None
394
+ onnx_safe ....................................... None
395
+ openai_gelu ..................................... False
396
+ optimizer ....................................... adam
397
+ optimizer_cpu_offload ........................... False
398
+ optimizer_offload_fraction ...................... 1.0
399
+ output_bert_embeddings .......................... False
400
+ overlap_cpu_optimizer_d2h_h2d ................... False
401
+ overlap_grad_reduce ............................. True
402
+ overlap_moe_expert_parallel_comm ................ False
403
+ overlap_p2p_comm ................................ False
404
+ overlap_p2p_comm_warmup_flush ................... False
405
+ overlap_param_gather ............................ True
406
+ overlap_param_gather_with_optimizer_step ........ False
407
+ override_opt_param_scheduler .................... False
408
+ params_dtype .................................... torch.bfloat16
409
+ patch_dim ....................................... 16
410
+ per_split_data_args_path ........................ None
411
+ perform_initialization .......................... True
412
+ pin_cpu_grads ................................... True
413
+ pin_cpu_params .................................. True
414
+ pipeline_model_parallel_comm_backend ............ None
415
+ pipeline_model_parallel_layout .................. None
416
+ pipeline_model_parallel_size .................... 1
417
+ position_embedding_type ......................... rope
418
+ pretrained_checkpoint ........................... None
419
+ profile ......................................... True
420
+ profile_ranks ................................... [0, 1, 2, 3, 4, 5, 6, 7]
421
+ profile_step_end ................................ 8
422
+ profile_step_start .............................. 6
423
+ q_lora_rank ..................................... None
424
+ qk_head_dim ..................................... 128
425
+ qk_l2_norm ...................................... False
426
+ qk_layernorm .................................... False
427
+ qk_pos_emb_head_dim ............................. 64
428
+ query_in_block_prob ............................. 0.1
429
+ rampup_batch_size ............................... None
430
+ rank ............................................ 0
431
+ recompute_granularity ........................... None
432
+ recompute_method ................................ None
433
+ recompute_modules ............................... None
434
+ recompute_num_layers ............................ None
435
+ record_memory_history ........................... False
436
+ relative_attention_max_distance ................. 128
437
+ relative_attention_num_buckets .................. 32
438
+ replication ..................................... False
439
+ replication_factor .............................. 2
440
+ replication_jump ................................ None
441
+ rerun_mode ...................................... validate_results
442
+ reset_attention_mask ............................ False
443
+ reset_position_ids .............................. False
444
+ result_rejected_tracker_filename ................ None
445
+ retriever_report_topk_accuracies ................ []
446
+ retriever_score_scaling ......................... False
447
+ retriever_seq_length ............................ 256
448
+ retro_add_retriever ............................. False
449
+ retro_attention_gate ............................ 1
450
+ retro_cyclic_train_iters ........................ None
451
+ retro_encoder_attention_dropout ................. 0.1
452
+ retro_encoder_hidden_dropout .................... 0.1
453
+ retro_encoder_layers ............................ 2
454
+ retro_num_neighbors ............................. 2
455
+ retro_num_retrieved_chunks ...................... 2
456
+ retro_project_dir ............................... None
457
+ retro_verify_neighbor_count ..................... True
458
+ reuse_grad_buf_for_mxfp8_param_ag ............... False
459
+ rope_scaling_factor ............................. 8.0
460
+ rope_type ....................................... None
461
+ rotary_base ..................................... 1000000
462
+ rotary_interleaved .............................. False
463
+ rotary_percent .................................. 1.0
464
+ rotary_scaling_factor ........................... 1.0
465
+ rotary_seq_len_interpolation_factor ............. None
466
+ run_workload_inspector_server ................... False
467
+ sample_rate ..................................... 1.0
468
+ save ............................................ None
469
+ save_interval ................................... None
470
+ save_retain_interval ............................ None
471
+ scatter_gather_tensors_in_pipeline .............. True
472
+ seed ............................................ 1234
473
+ seq_length ...................................... 4096
474
+ sequence_parallel ............................... False
475
+ sft ............................................. False
476
+ sft_tokenizer_prompt_format ..................... nemotron-h-aligned
477
+ sgd_momentum .................................... 0.9
478
+ short_seq_prob .................................. 0.1
479
+ skip_train ...................................... False
480
+ skipped_train_samples ........................... 0
481
+ spec ............................................ None
482
+ split ........................................... 100,0,0
483
+ squared_relu .................................... False
484
+ start_weight_decay .............................. 0.1
485
+ straggler_ctrlr_port ............................ 65535
486
+ straggler_minmax_count .......................... 1
487
+ suggested_communication_unit_size ............... None
488
+ swiglu .......................................... True
489
+ swin_backbone_type .............................. tiny
490
+ symmetric_ar_type ............................... None
491
+ te_rng_tracker .................................. True
492
+ tensor_model_parallel_size ...................... 1
493
+ tensorboard_dir ................................. None
494
+ tensorboard_log_interval ........................ 1
495
+ tensorboard_queue_size .......................... 1000
496
+ test_data_path .................................. None
497
+ test_mode ....................................... False
498
+ tiktoken_num_special_tokens ..................... 1000
499
+ tiktoken_pattern ................................ None
500
+ tiktoken_special_tokens ......................... None
501
+ timing_log_level ................................ 0
502
+ timing_log_option ............................... minmax
503
+ titles_data_path ................................ None
504
+ tokenizer_model ................................. None
505
+ tokenizer_type .................................. GPT2BPETokenizer
506
+ torch_fsdp2_reshard_after_forward ............... True
507
+ tp_comm_bootstrap_backend ....................... nccl
508
+ tp_comm_bulk_dgrad .............................. True
509
+ tp_comm_bulk_wgrad .............................. True
510
+ tp_comm_overlap ................................. False
511
+ tp_comm_overlap_ag .............................. True
512
+ tp_comm_overlap_cfg ............................. None
513
+ tp_comm_overlap_rs .............................. True
514
+ tp_comm_overlap_rs_dgrad ........................ False
515
+ tp_comm_split_ag ................................ True
516
+ tp_comm_split_rs ................................ True
517
+ train_data_path ................................. None
518
+ train_iters ..................................... 10
519
+ train_samples ................................... None
520
+ train_sync_interval ............................. None
521
+ transformer_impl ................................ transformer_engine
522
+ transformer_pipeline_model_parallel_size ........ 1
523
+ untie_embeddings_and_output_weights ............. True
524
+ use_checkpoint_args ............................. False
525
+ use_checkpoint_opt_param_scheduler .............. False
526
+ use_cpu_initialization .......................... None
527
+ use_custom_fsdp ................................. False
528
+ use_dist_ckpt ................................... True
529
+ use_dist_ckpt_deprecated ........................ False
530
+ use_distributed_optimizer ....................... True
531
+ use_flash_attn .................................. False
532
+ use_legacy_models ............................... False
533
+ use_mp_args_from_checkpoint_args ................ False
534
+ use_one_sent_docs ............................... False
535
+ use_persistent_ckpt_worker ...................... False
536
+ use_precision_aware_optimizer ................... False
537
+ use_pytorch_profiler ............................ False
538
+ use_ring_exchange_p2p ........................... False
539
+ use_rope_scaling ................................ False
540
+ use_rotary_position_embeddings .................. False
541
+ use_sharp ....................................... False
542
+ use_tokenizer_model_from_checkpoint_args ........ True
543
+ use_torch_fsdp2 ................................. False
544
+ use_torch_optimizer_for_cpu_offload ............. False
545
+ use_tp_pp_dp_mapping ............................ False
546
+ v_head_dim ...................................... 128
547
+ valid_data_path ................................. None
548
+ variable_seq_lengths ............................ False
549
+ virtual_pipeline_model_parallel_size ............ None
550
+ vision_backbone_type ............................ vit
551
+ vision_pretraining .............................. False
552
+ vision_pretraining_type ......................... classify
553
+ vocab_extra_ids ................................. 0
554
+ vocab_file ...................................... /workspace/data/wikitext/gpt2-vocab.json
555
+ vocab_size ...................................... None
556
+ wandb_exp_name ..................................
557
+ wandb_project ...................................
558
+ wandb_save_dir ..................................
559
+ weight_decay .................................... 0.1
560
+ weight_decay_incr_style ......................... constant
561
+ wgrad_deferral_limit ............................ 0
562
+ world_size ...................................... 8
563
+ yaml_cfg ........................................ None
564
+ -------------------- end of arguments ---------------------
565
+ INFO:megatron.core.num_microbatches_calculator:setting number of microbatches to constant 2
566
+ > building GPT2BPETokenizer tokenizer ...
567
+ > padded vocab (size: 50257) with 47 dummy tokens (new size: 50304)
568
+ WARNING:megatron.core.rerun_state_machine:RerunStateMachine initialized in mode RerunMode.VALIDATE_RESULTS
569
+ > initializing torch distributed ...
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
+ /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
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
575
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
576
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
577
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
578
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
579
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
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 1 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 3 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 6 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
587
+ [Gloo] Rank 4 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
588
+ [Gloo] Rank 7 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
589
+ [Gloo] Rank 5 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
590
+ [Gloo] Rank 3 is connected to 7 peer ranks. [Gloo] Rank Expected number of connected peer ranks is : 72
591
+ 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 6 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
594
+ [Gloo] Rank 7 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
595
+ [Gloo] Rank 1 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 4 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
+ [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
600
+
601
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
602
+ > initialized tensor model parallel with size 1
603
+ > initialized pipeline model parallel with size 1
604
+ > setting random seeds to 1234 ...
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
+ [rank2]:[W830 09:01:25.675613308 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 09:01:25.675808796 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 09:01:25.680622701 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
+ [rank7]:[W830 09:01:25.681161364 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.
613
+ [rank4]:[W830 09:01:25.681187865 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.
614
+ [rank6]:[W830 09:01:25.681485574 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
+ [rank5]:[W830 09:01:25.681607216 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.
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: 0.427 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 09:01:26.295558151 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: 3.590 seconds
625
+ time to initialize megatron (seconds): 9.488
626
+ [after megatron is initialized] datetime: 2026-08-30 09:01:31
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): 6722555904
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 (42086400 elements, 42086400 padded size):
670
+ module.decoder.layers.15.pre_mlp_layernorm.weight
671
+ module.decoder.layers.15.self_attention.linear_qkv.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 (42082304 elements, 42082304 padded size):
677
+ module.decoder.layers.14.self_attention.linear_qkv.weight
678
+ module.decoder.layers.14.pre_mlp_layernorm.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 (42082304 elements, 42082304 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 (42082304 elements, 42082304 padded size):
689
+ module.decoder.layers.12.self_attention.linear_qkv.weight
690
+ module.decoder.layers.12.pre_mlp_layernorm.weight
691
+ module.decoder.layers.12.mlp.router.weight
692
+ module.decoder.layers.12.self_attention.linear_qkv.layer_norm_weight
693
+ module.decoder.layers.12.self_attention.linear_proj.weight
694
+ Params for bucket 6 (42082304 elements, 42082304 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 (42082304 elements, 42082304 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.pre_mlp_layernorm.weight
705
+ module.decoder.layers.10.self_attention.linear_qkv.weight
706
+ Params for bucket 8 (42082304 elements, 42082304 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_qkv.layer_norm_weight
711
+ module.decoder.layers.9.self_attention.linear_proj.weight
712
+ Params for bucket 9 (42082304 elements, 42082304 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 (42082304 elements, 42082304 padded size):
719
+ module.decoder.layers.7.self_attention.linear_qkv.weight
720
+ module.decoder.layers.7.pre_mlp_layernorm.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 (42082304 elements, 42082304 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 (42082304 elements, 42082304 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 (42082304 elements, 42082304 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.self_attention.linear_qkv.weight
741
+ module.decoder.layers.4.pre_mlp_layernorm.weight
742
+ Params for bucket 14 (42082304 elements, 42082304 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 (42082304 elements, 42082304 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 (42082304 elements, 42082304 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.self_attention.linear_qkv.weight
759
+ module.decoder.layers.1.pre_mlp_layernorm.weight
760
+ Params for bucket 17 (42082304 elements, 42082304 padded size):
761
+ module.decoder.layers.0.self_attention.linear_proj.weight
762
+ module.decoder.layers.0.mlp.router.weight
763
+ module.decoder.layers.0.self_attention.linear_qkv.weight
764
+ module.decoder.layers.0.pre_mlp_layernorm.weight
765
+ module.decoder.layers.0.self_attention.linear_qkv.layer_norm_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: 56
769
+ Params for bucket 1 (44040192 elements, 44040192 padded size):
770
+ module.decoder.layers.15.mlp.experts.linear_fc2.weight3
771
+ module.decoder.layers.15.mlp.experts.linear_fc2.weight2
772
+ module.decoder.layers.15.mlp.experts.linear_fc2.weight1
773
+ Params for bucket 2 (44040192 elements, 44040192 padded size):
774
+ module.decoder.layers.15.mlp.experts.linear_fc2.weight0
775
+ module.decoder.layers.15.mlp.experts.linear_fc1.weight3
776
+ Params for bucket 3 (58720256 elements, 58720256 padded size):
777
+ module.decoder.layers.15.mlp.experts.linear_fc1.weight2
778
+ module.decoder.layers.15.mlp.experts.linear_fc1.weight1
779
+ Params for bucket 4 (44040192 elements, 44040192 padded size):
780
+ module.decoder.layers.15.mlp.experts.linear_fc1.weight0
781
+ module.decoder.layers.14.mlp.experts.linear_fc2.weight3
782
+ Params for bucket 5 (44040192 elements, 44040192 padded size):
783
+ module.decoder.layers.14.mlp.experts.linear_fc2.weight2
784
+ module.decoder.layers.14.mlp.experts.linear_fc2.weight0
785
+ module.decoder.layers.14.mlp.experts.linear_fc2.weight1
786
+ Params for bucket 6 (58720256 elements, 58720256 padded size):
787
+ module.decoder.layers.14.mlp.experts.linear_fc1.weight3
788
+ module.decoder.layers.14.mlp.experts.linear_fc1.weight2
789
+ Params for bucket 7 (58720256 elements, 58720256 padded size):
790
+ module.decoder.layers.14.mlp.experts.linear_fc1.weight1
791
+ module.decoder.layers.14.mlp.experts.linear_fc1.weight0
792
+ Params for bucket 8 (44040192 elements, 44040192 padded size):
793
+ module.decoder.layers.13.mlp.experts.linear_fc2.weight3
794
+ module.decoder.layers.13.mlp.experts.linear_fc2.weight2
795
+ module.decoder.layers.13.mlp.experts.linear_fc2.weight1
796
+ Params for bucket 9 (44040192 elements, 44040192 padded size):
797
+ module.decoder.layers.13.mlp.experts.linear_fc2.weight0
798
+ module.decoder.layers.13.mlp.experts.linear_fc1.weight3
799
+ Params for bucket 10 (58720256 elements, 58720256 padded size):
800
+ module.decoder.layers.13.mlp.experts.linear_fc1.weight2
801
+ module.decoder.layers.13.mlp.experts.linear_fc1.weight1
802
+ Params for bucket 11 (44040192 elements, 44040192 padded size):
803
+ module.decoder.layers.13.mlp.experts.linear_fc1.weight0
804
+ module.decoder.layers.12.mlp.experts.linear_fc2.weight3
805
+ Params for bucket 12 (44040192 elements, 44040192 padded size):
806
+ module.decoder.layers.12.mlp.experts.linear_fc2.weight2
807
+ module.decoder.layers.12.mlp.experts.linear_fc2.weight1
808
+ module.decoder.layers.12.mlp.experts.linear_fc2.weight0
809
+ Params for bucket 13 (58720256 elements, 58720256 padded size):
810
+ module.decoder.layers.12.mlp.experts.linear_fc1.weight3
811
+ module.decoder.layers.12.mlp.experts.linear_fc1.weight2
812
+ Params for bucket 14 (58720256 elements, 58720256 padded size):
813
+ module.decoder.layers.12.mlp.experts.linear_fc1.weight1
814
+ module.decoder.layers.12.mlp.experts.linear_fc1.weight0
815
+ Params for bucket 15 (44040192 elements, 44040192 padded size):
816
+ module.decoder.layers.11.mlp.experts.linear_fc2.weight3
817
+ module.decoder.layers.11.mlp.experts.linear_fc2.weight1
818
+ module.decoder.layers.11.mlp.experts.linear_fc2.weight2
819
+ Params for bucket 16 (44040192 elements, 44040192 padded size):
820
+ module.decoder.layers.11.mlp.experts.linear_fc2.weight0
821
+ module.decoder.layers.11.mlp.experts.linear_fc1.weight3
822
+ Params for bucket 17 (58720256 elements, 58720256 padded size):
823
+ module.decoder.layers.11.mlp.experts.linear_fc1.weight2
824
+ module.decoder.layers.11.mlp.experts.linear_fc1.weight1
825
+ Params for bucket 18 (44040192 elements, 44040192 padded size):
826
+ module.decoder.layers.11.mlp.experts.linear_fc1.weight0
827
+ module.decoder.layers.10.mlp.experts.linear_fc2.weight3
828
+ Params for bucket 19 (44040192 elements, 44040192 padded size):
829
+ module.decoder.layers.10.mlp.experts.linear_fc2.weight2
830
+ module.decoder.layers.10.mlp.experts.linear_fc2.weight0
831
+ module.decoder.layers.10.mlp.experts.linear_fc2.weight1
832
+ Params for bucket 20 (58720256 elements, 58720256 padded size):
833
+ module.decoder.layers.10.mlp.experts.linear_fc1.weight3
834
+ module.decoder.layers.10.mlp.experts.linear_fc1.weight2
835
+ Params for bucket 21 (58720256 elements, 58720256 padded size):
836
+ module.decoder.layers.10.mlp.experts.linear_fc1.weight1
837
+ module.decoder.layers.10.mlp.experts.linear_fc1.weight0
838
+ Params for bucket 22 (44040192 elements, 44040192 padded size):
839
+ module.decoder.layers.9.mlp.experts.linear_fc2.weight3
840
+ module.decoder.layers.9.mlp.experts.linear_fc2.weight2
841
+ module.decoder.layers.9.mlp.experts.linear_fc2.weight1
842
+ Params for bucket 23 (44040192 elements, 44040192 padded size):
843
+ module.decoder.layers.9.mlp.experts.linear_fc2.weight0
844
+ module.decoder.layers.9.mlp.experts.linear_fc1.weight3
845
+ Params for bucket 24 (58720256 elements, 58720256 padded size):
846
+ module.decoder.layers.9.mlp.experts.linear_fc1.weight2
847
+ module.decoder.layers.9.mlp.experts.linear_fc1.weight1
848
+ Params for bucket 25 (44040192 elements, 44040192 padded size):
849
+ module.decoder.layers.9.mlp.experts.linear_fc1.weight0
850
+ module.decoder.layers.8.mlp.experts.linear_fc2.weight3
851
+ Params for bucket 26 (44040192 elements, 44040192 padded size):
852
+ module.decoder.layers.8.mlp.experts.linear_fc2.weight2
853
+ module.decoder.layers.8.mlp.experts.linear_fc2.weight0
854
+ module.decoder.layers.8.mlp.experts.linear_fc2.weight1
855
+ Params for bucket 27 (58720256 elements, 58720256 padded size):
856
+ module.decoder.layers.8.mlp.experts.linear_fc1.weight3
857
+ module.decoder.layers.8.mlp.experts.linear_fc1.weight2
858
+ Params for bucket 28 (58720256 elements, 58720256 padded size):
859
+ module.decoder.layers.8.mlp.experts.linear_fc1.weight1
860
+ module.decoder.layers.8.mlp.experts.linear_fc1.weight0
861
+ Params for bucket 29 (44040192 elements, 44040192 padded size):
862
+ module.decoder.layers.7.mlp.experts.linear_fc2.weight3
863
+ module.decoder.layers.7.mlp.experts.linear_fc2.weight1
864
+ module.decoder.layers.7.mlp.experts.linear_fc2.weight2
865
+ Params for bucket 30 (44040192 elements, 44040192 padded size):
866
+ module.decoder.layers.7.mlp.experts.linear_fc2.weight0
867
+ module.decoder.layers.7.mlp.experts.linear_fc1.weight3
868
+ Params for bucket 31 (58720256 elements, 58720256 padded size):
869
+ module.decoder.layers.7.mlp.experts.linear_fc1.weight2
870
+ module.decoder.layers.7.mlp.experts.linear_fc1.weight1
871
+ Params for bucket 32 (44040192 elements, 44040192 padded size):
872
+ module.decoder.layers.7.mlp.experts.linear_fc1.weight0
873
+ module.decoder.layers.6.mlp.experts.linear_fc2.weight3
874
+ Params for bucket 33 (44040192 elements, 44040192 padded size):
875
+ module.decoder.layers.6.mlp.experts.linear_fc2.weight2
876
+ module.decoder.layers.6.mlp.experts.linear_fc2.weight1
877
+ module.decoder.layers.6.mlp.experts.linear_fc2.weight0
878
+ Params for bucket 34 (58720256 elements, 58720256 padded size):
879
+ module.decoder.layers.6.mlp.experts.linear_fc1.weight3
880
+ module.decoder.layers.6.mlp.experts.linear_fc1.weight2
881
+ Params for bucket 35 (58720256 elements, 58720256 padded size):
882
+ module.decoder.layers.6.mlp.experts.linear_fc1.weight1
883
+ module.decoder.layers.6.mlp.experts.linear_fc1.weight0
884
+ Params for bucket 36 (44040192 elements, 44040192 padded size):
885
+ module.decoder.layers.5.mlp.experts.linear_fc2.weight3
886
+ module.decoder.layers.5.mlp.experts.linear_fc2.weight2
887
+ module.decoder.layers.5.mlp.experts.linear_fc2.weight1
888
+ Params for bucket 37 (44040192 elements, 44040192 padded size):
889
+ module.decoder.layers.5.mlp.experts.linear_fc2.weight0
890
+ module.decoder.layers.5.mlp.experts.linear_fc1.weight3
891
+ Params for bucket 38 (58720256 elements, 58720256 padded size):
892
+ module.decoder.layers.5.mlp.experts.linear_fc1.weight2
893
+ module.decoder.layers.5.mlp.experts.linear_fc1.weight1
894
+ Params for bucket 39 (44040192 elements, 44040192 padded size):
895
+ module.decoder.layers.5.mlp.experts.linear_fc1.weight0
896
+ module.decoder.layers.4.mlp.experts.linear_fc2.weight3
897
+ Params for bucket 40 (44040192 elements, 44040192 padded size):
898
+ module.decoder.layers.4.mlp.experts.linear_fc2.weight2
899
+ module.decoder.layers.4.mlp.experts.linear_fc2.weight0
900
+ module.decoder.layers.4.mlp.experts.linear_fc2.weight1
901
+ Params for bucket 41 (58720256 elements, 58720256 padded size):
902
+ module.decoder.layers.4.mlp.experts.linear_fc1.weight3
903
+ module.decoder.layers.4.mlp.experts.linear_fc1.weight2
904
+ Params for bucket 42 (58720256 elements, 58720256 padded size):
905
+ module.decoder.layers.4.mlp.experts.linear_fc1.weight1
906
+ module.decoder.layers.4.mlp.experts.linear_fc1.weight0
907
+ Params for bucket 43 (44040192 elements, 44040192 padded size):
908
+ module.decoder.layers.3.mlp.experts.linear_fc2.weight3
909
+ module.decoder.layers.3.mlp.experts.linear_fc2.weight1
910
+ module.decoder.layers.3.mlp.experts.linear_fc2.weight2
911
+ Params for bucket 44 (44040192 elements, 44040192 padded size):
912
+ module.decoder.layers.3.mlp.experts.linear_fc2.weight0
913
+ module.decoder.layers.3.mlp.experts.linear_fc1.weight3
914
+ Params for bucket 45 (58720256 elements, 58720256 padded size):
915
+ module.decoder.layers.3.mlp.experts.linear_fc1.weight2
916
+ module.decoder.layers.3.mlp.experts.linear_fc1.weight1
917
+ Params for bucket 46 (44040192 elements, 44040192 padded size):
918
+ module.decoder.layers.3.mlp.experts.linear_fc1.weight0
919
+ module.decoder.layers.2.mlp.experts.linear_fc2.weight3
920
+ Params for bucket 47 (44040192 elements, 44040192 padded size):
921
+ module.decoder.layers.2.mlp.experts.linear_fc2.weight2
922
+ module.decoder.layers.2.mlp.experts.linear_fc2.weight1
923
+ module.decoder.layers.2.mlp.experts.linear_fc2.weight0
924
+ Params for bucket 48 (58720256 elements, 58720256 padded size):
925
+ module.decoder.layers.2.mlp.experts.linear_fc1.weight3
926
+ module.decoder.layers.2.mlp.experts.linear_fc1.weight2
927
+ Params for bucket 49 (58720256 elements, 58720256 padded size):
928
+ module.decoder.layers.2.mlp.experts.linear_fc1.weight1
929
+ module.decoder.layers.2.mlp.experts.linear_fc1.weight0
930
+ Params for bucket 50 (44040192 elements, 44040192 padded size):
931
+ module.decoder.layers.1.mlp.experts.linear_fc2.weight3
932
+ module.decoder.layers.1.mlp.experts.linear_fc2.weight1
933
+ module.decoder.layers.1.mlp.experts.linear_fc2.weight2
934
+ Params for bucket 51 (44040192 elements, 44040192 padded size):
935
+ module.decoder.layers.1.mlp.experts.linear_fc2.weight0
936
+ module.decoder.layers.1.mlp.experts.linear_fc1.weight3
937
+ Params for bucket 52 (58720256 elements, 58720256 padded size):
938
+ module.decoder.layers.1.mlp.experts.linear_fc1.weight2
939
+ module.decoder.layers.1.mlp.experts.linear_fc1.weight1
940
+ Params for bucket 53 (44040192 elements, 44040192 padded size):
941
+ module.decoder.layers.1.mlp.experts.linear_fc1.weight0
942
+ module.decoder.layers.0.mlp.experts.linear_fc2.weight3
943
+ Params for bucket 54 (44040192 elements, 44040192 padded size):
944
+ module.decoder.layers.0.mlp.experts.linear_fc2.weight2
945
+ module.decoder.layers.0.mlp.experts.linear_fc2.weight0
946
+ module.decoder.layers.0.mlp.experts.linear_fc2.weight1
947
+ Params for bucket 55 (58720256 elements, 58720256 padded size):
948
+ module.decoder.layers.0.mlp.experts.linear_fc1.weight3
949
+ module.decoder.layers.0.mlp.experts.linear_fc1.weight2
950
+ Params for bucket 56 (58720256 elements, 58720256 padded size):
951
+ module.decoder.layers.0.mlp.experts.linear_fc1.weight1
952
+ module.decoder.layers.0.mlp.experts.linear_fc1.weight0
953
+ 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 0x702fe1db3830>, config_logger_dir='')
954
+ INFO:megatron.core.optimizer_param_scheduler:> learning rate decay style: cosine
955
+ [after model, optimizer, and learning rate scheduler are built] datetime: 2026-08-30 09:01:35
956
+ > building train, validation, and test datasets ...
957
+ > datasets target sizes (minimum size):
958
+ train: 160
959
+ validation: 0
960
+ test: 0
961
+ INFO:megatron.core.datasets.blended_megatron_dataset_config:Let split_matrix = [(0, 1.0), None, None]
962
+ > building train, validation, and test datasets for GPT ...
963
+ INFO:megatron.core.datasets.blended_megatron_dataset_builder:Building GPTDataset splits with sizes=(160, 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 0x702fe1f0d1f0>, 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)
964
+ INFO:megatron.core.datasets.indexed_dataset:Load the _IndexReader from /workspace/data/wikitext/wikitext103_gpt2_text_document.idx
965
+ INFO:megatron.core.datasets.indexed_dataset: Extract the sequence lengths
966
+ INFO:megatron.core.datasets.indexed_dataset: Extract the sequence pointers
967
+ INFO:megatron.core.datasets.indexed_dataset: Extract the document indices
968
+ INFO:megatron.core.datasets.indexed_dataset:> total number of sequences: 1165029
969
+ INFO:megatron.core.datasets.indexed_dataset:> total number of documents: 1801350
970
+ INFO:megatron.core.datasets.gpt_dataset:Load the GPTDataset train indices
971
+ INFO:megatron.core.datasets.gpt_dataset: Load the document index from 68684a515e118253a8572c4517cf2ffe-GPTDataset-train-document_index.npy
972
+ INFO:megatron.core.datasets.gpt_dataset: Load the sample index from 68684a515e118253a8572c4517cf2ffe-GPTDataset-train-sample_index.npy
973
+ INFO:megatron.core.datasets.gpt_dataset: Load the shuffle index from 68684a515e118253a8572c4517cf2ffe-GPTDataset-train-shuffle_index.npy
974
+ INFO:megatron.core.datasets.gpt_dataset:> total number of samples: 29073
975
+ > finished creating GPT datasets ...
976
+ [after dataloaders are built] datetime: 2026-08-30 09:01:35
977
+ done with setup ...
978
+ (min, max) time across ranks (ms):
979
+ model-and-optimizer-setup ......................: (3394.48, 3400.83)
980
+ train/valid/test-data-iterators-setup ..........: (96.84, 132.71)
981
+ training ...
982
+ Overwriting rerun_state_machine.current_iteration from -1 to 0...
983
+ [before the start of training step] datetime: 2026-08-30 09:01:35
984
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
985
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
986
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
987
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
988
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
989
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
990
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
991
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
992
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
993
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
994
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
995
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
996
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
997
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
998
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
999
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
1000
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1001
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1002
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1003
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1004
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
1005
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
1006
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
1007
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
1008
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1009
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1010
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
1011
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
1012
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1013
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1014
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
1015
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
1016
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1017
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1018
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1019
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1020
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1021
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1022
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1023
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1024
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1025
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1026
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1027
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1028
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1029
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1030
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1031
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1032
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1033
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1034
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
1035
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
1036
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1037
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1038
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1039
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1040
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1041
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1042
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
1043
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
1044
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
1045
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
1046
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
1047
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
1048
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1049
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1050
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
1051
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
1052
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1053
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1054
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
1055
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
1056
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1057
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1058
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
1059
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
1060
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1061
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1062
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
1063
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
1064
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1065
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1066
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1067
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1068
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1069
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1070
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1071
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1072
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1073
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1074
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1075
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1076
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1077
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1078
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1079
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1080
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1081
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1082
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1083
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1084
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1085
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1086
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1087
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1088
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1089
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1090
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1091
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1092
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1093
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1094
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1095
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1096
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
1097
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
1098
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
1099
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
1100
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
1101
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
1102
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
1103
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
1104
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
1105
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
1106
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
1107
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
1108
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
1109
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
1110
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
1111
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
1112
+ [2026-08-30 09:02:05] iteration 1/ 10 | consumed samples: 16 | elapsed time per iteration (ms): 30291.8 | throughput per GPU (TFLOP/s/GPU): 11.0 | learning rate: 2.000000E-05 | global batch size: 16 | lm loss: 1.102622E+01 | loss scale: 1.0 | grad norm: 44.464 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 33536 | min safe capacity: 1.0234 |
1113
+ Number of parameters in transformer block in billions: 23.22
1114
+ Number of parameters in embedding layers in billions: 0.41
1115
+ Total number of parameters in billions: 23.63
1116
+ Number of parameters in most loaded shard in billions: 23.6320
1117
+ Theoretical memory footprints: weight and optimizer=169029.43 MB
1118
+ [Rank 0] (after 1 iterations) memory (MB) | allocated: 58124.0458984375 | max allocated: 74507.9609375 | reserved: 76502.0 | max reserved: 76502.0
1119
+ [2026-08-30 09:02:06] iteration 2/ 10 | consumed samples: 32 | elapsed time per iteration (ms): 641.7 | throughput per GPU (TFLOP/s/GPU): 519.5 | learning rate: 4.000000E-05 | global batch size: 16 | lm loss: 1.101621E+01 | loss scale: 1.0 | grad norm: 45.877 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 33664 | min safe capacity: 1.0273 |
1120
+ [2026-08-30 09:02:06] iteration 3/ 10 | consumed samples: 48 | elapsed time per iteration (ms): 550.8 | throughput per GPU (TFLOP/s/GPU): 605.2 | learning rate: 6.000000E-05 | global batch size: 16 | lm loss: 1.028951E+01 | loss scale: 1.0 | grad norm: 201.661 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 33664 | min safe capacity: 1.0273 |
1121
+ Capture CUDA graph for training!!!
1122
+ CUDA graph capture done!!!
1123
+ [2026-08-30 09:02:09] iteration 4/ 10 | consumed samples: 64 | elapsed time per iteration (ms): 2792.2 | throughput per GPU (TFLOP/s/GPU): 119.4 | learning rate: 8.000000E-05 | global batch size: 16 | lm loss: 1.322907E+01 | loss scale: 1.0 | grad norm: 31.648 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 33664 | min safe capacity: 1.0273 |
1124
+ [2026-08-30 09:02:10] iteration 5/ 10 | consumed samples: 80 | elapsed time per iteration (ms): 471.5 | throughput per GPU (TFLOP/s/GPU): 707.1 | learning rate: 1.000000E-04 | global batch size: 16 | lm loss: 1.039297E+01 | loss scale: 1.0 | grad norm: 16.304 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 33664 | min safe capacity: 1.0273 |
1125
+ [2026-08-30 09:02:10] iteration 6/ 10 | consumed samples: 96 | elapsed time per iteration (ms): 466.9 | throughput per GPU (TFLOP/s/GPU): 714.0 | learning rate: 7.750000E-05 | global batch size: 16 | lm loss: 1.047586E+01 | loss scale: 1.0 | grad norm: 38.779 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 33664 | min safe capacity: 1.0273 |
1126
+ [2026-08-30 09:02:14] iteration 7/ 10 | consumed samples: 112 | elapsed time per iteration (ms): 3365.8 | throughput per GPU (TFLOP/s/GPU): 99.0 | learning rate: 3.250000E-05 | global batch size: 16 | lm loss: 1.013083E+01 | loss scale: 1.0 | grad norm: 7.720 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 33664 | min safe capacity: 1.0273 |
1127
+ [2026-08-30 09:02:14] iteration 8/ 10 | consumed samples: 128 | elapsed time per iteration (ms): 488.1 | throughput per GPU (TFLOP/s/GPU): 683.0 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 9.558809E+00 | loss scale: 1.0 | grad norm: 5.844 | number of skipped iterations: 0 | number of nan iterations: 0 | peak recv tokens: 33664 | min safe capacity: 1.0273 |
1128
+ Capture range started in the application.
1129
+ Capture range ended in the application.
1130
+ Generating '/tmp/nsys-report-f1d3.qdstrm'
1131
+
1132
+
tiera-rerun-fixed-moonep/ultraep/tierbtrace/tierbtrace_gbs16/results.tsv ADDED
@@ -0,0 +1 @@
 
 
1
+ config backend policy slots seq gbs cap recompute rep status iter_ms_mean iter_ms_min peak_mib run_dir
tiera-rerun-fixed-moonep/ultraep/tierbtrace/tierbtrace_gbs16/trace_extension_build.log ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Could not find an up-to-date installation of `packaging`. License expressions might not be validated. To enforce validation, please install `packaging>=24.2`.
2
+ /usr/local/lib/python3.12/dist-packages/setuptools/config/_apply_pyprojecttoml.py:82: SetuptoolsDeprecationWarning: `project.license` as a TOML table is deprecated
3
+ !!
4
+
5
+ ********************************************************************************
6
+ Please use a simple string containing a SPDX expression for `project.license`. You can also use `project.license-files`. (Both options available on setuptools>=77.0.0).
7
+
8
+ This deprecation is overdue, please update your project and remove deprecated
9
+ calls to avoid build errors in the future.
10
+
11
+ See https://packaging.python.org/en/latest/guides/writing-pyproject-toml/#license for details.
12
+ ********************************************************************************
13
+
14
+ !!
15
+ corresp(dist, value, root_dir)
16
+ /usr/local/lib/python3.12/dist-packages/setuptools/config/_apply_pyprojecttoml.py:61: SetuptoolsDeprecationWarning: License classifiers are deprecated.
17
+ !!
18
+
19
+ ********************************************************************************
20
+ Please consider removing the following classifiers in favor of a SPDX license expression:
21
+
22
+ License :: OSI Approved :: BSD License
23
+
24
+ See https://packaging.python.org/en/latest/guides/writing-pyproject-toml/#license for details.
25
+ ********************************************************************************
26
+
27
+ !!
28
+ dist._finalize_license_expression()
29
+ /usr/local/lib/python3.12/dist-packages/setuptools/dist.py:759: SetuptoolsDeprecationWarning: License classifiers are deprecated.
30
+ !!
31
+
32
+ ********************************************************************************
33
+ Please consider removing the following classifiers in favor of a SPDX license expression:
34
+
35
+ License :: OSI Approved :: BSD License
36
+
37
+ See https://packaging.python.org/en/latest/guides/writing-pyproject-toml/#license for details.
38
+ ********************************************************************************
39
+
40
+ !!
41
+ self._finalize_license_expression()
42
+ running build_ext
43
+ building 'megatron.core.datasets.helpers_cpp' extension
44
+ Emitting ninja build file /workspace/Megatron-UltraEP/build/temp.linux-x86_64-cpython-312/build.ninja...
45
+ Compiling objects...
46
+ Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
47
+ ninja: no work to do.
48
+ x86_64-linux-gnu-g++ -fno-strict-overflow -Wsign-compare -DNDEBUG -g -O2 -Wall -shared -Wl,-O1 -Wl,-Bsymbolic-functions /workspace/Megatron-UltraEP/build/temp.linux-x86_64-cpython-312/megatron/core/datasets/helpers.o -L/usr/lib/x86_64-linux-gnu -o build/lib.linux-x86_64-cpython-312/megatron/core/datasets/helpers_cpp.cpython-312-x86_64-linux-gnu.so
49
+ building 'megatron.core.transformer.moe._ultraep_trace_C' extension
50
+ Emitting ninja build file /workspace/Megatron-UltraEP/build/temp.linux-x86_64-cpython-312/build.ninja...
51
+ Compiling objects...
52
+ Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
53
+ [1/2] c++ -MMD -MF /workspace/Megatron-UltraEP/build/temp.linux-x86_64-cpython-312/megatron/core/transformer/moe/ultraep_trace_ops.o.d -fno-strict-overflow -Wsign-compare -DNDEBUG -g -O2 -Wall -fPIC -I/usr/local/lib/python3.12/dist-packages/torch/include -I/usr/local/lib/python3.12/dist-packages/torch/include/torch/csrc/api/include -I/usr/local/cuda/include -I/usr/include/python3.12 -c -c /workspace/Megatron-UltraEP/megatron/core/transformer/moe/ultraep_trace_ops.cpp -o /workspace/Megatron-UltraEP/build/temp.linux-x86_64-cpython-312/megatron/core/transformer/moe/ultraep_trace_ops.o -O3 -std=c++17 -DTORCH_API_INCLUDE_EXTENSION_H '-DPYBIND11_COMPILER_TYPE="_gcc"' '-DPYBIND11_STDLIB="_libstdcpp"' '-DPYBIND11_BUILD_ABI="_cxxabi1016"' -DTORCH_EXTENSION_NAME=_ultraep_trace_C
54
+ [2/2] /usr/local/cuda/bin/nvcc --generate-dependencies-with-compile --dependency-output /workspace/Megatron-UltraEP/build/temp.linux-x86_64-cpython-312/megatron/core/transformer/moe/ultraep_trace_kernels.o.d -I/usr/local/lib/python3.12/dist-packages/torch/include -I/usr/local/lib/python3.12/dist-packages/torch/include/torch/csrc/api/include -I/usr/local/cuda/include -I/usr/include/python3.12 -c -c /workspace/Megatron-UltraEP/megatron/core/transformer/moe/ultraep_trace_kernels.cu -o /workspace/Megatron-UltraEP/build/temp.linux-x86_64-cpython-312/megatron/core/transformer/moe/ultraep_trace_kernels.o -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr --compiler-options ''"'"'-fPIC'"'"'' -O3 -std=c++17 -lineinfo -DTORCH_API_INCLUDE_EXTENSION_H '-DPYBIND11_COMPILER_TYPE="_gcc"' '-DPYBIND11_STDLIB="_libstdcpp"' '-DPYBIND11_BUILD_ABI="_cxxabi1016"' -DTORCH_EXTENSION_NAME=_ultraep_trace_C -gencode=arch=compute_100,code=sm_100 -gencode=arch=compute_120,code=compute_120 -gencode=arch=compute_120,code=sm_120 -gencode=arch=compute_75,code=sm_75 -gencode=arch=compute_80,code=sm_80 -gencode=arch=compute_86,code=sm_86 -gencode=arch=compute_90,code=sm_90
55
+ x86_64-linux-gnu-g++ -fno-strict-overflow -Wsign-compare -DNDEBUG -g -O2 -Wall -shared -Wl,-O1 -Wl,-Bsymbolic-functions /workspace/Megatron-UltraEP/build/temp.linux-x86_64-cpython-312/megatron/core/transformer/moe/ultraep_trace_kernels.o /workspace/Megatron-UltraEP/build/temp.linux-x86_64-cpython-312/megatron/core/transformer/moe/ultraep_trace_ops.o -L/usr/local/lib/python3.12/dist-packages/torch/lib -L/usr/local/cuda/lib64 -L/usr/lib/x86_64-linux-gnu -lc10 -ltorch -ltorch_cpu -ltorch_python -lcudart -lc10_cuda -ltorch_cuda -o build/lib.linux-x86_64-cpython-312/megatron/core/transformer/moe/_ultraep_trace_C.cpython-312-x86_64-linux-gnu.so
56
+ copying build/lib.linux-x86_64-cpython-312/megatron/core/datasets/helpers_cpp.cpython-312-x86_64-linux-gnu.so -> megatron/core/datasets
57
+ copying build/lib.linux-x86_64-cpython-312/megatron/core/transformer/moe/_ultraep_trace_C.cpython-312-x86_64-linux-gnu.so -> megatron/core/transformer/moe
58
+ load check passed: /workspace/Megatron-UltraEP/megatron/core/transformer/moe/_ultraep_trace_C.cpython-312-x86_64-linux-gnu.so -> torch.ops.megatron_ultraep_trace.fwd_ultraep_forward_begin
tiera-rerun-fixed-moonep/ultraep/tierbtrace/tierbtrace_gbs16/trace_results.tsv ADDED
@@ -0,0 +1,2 @@
 
 
 
1
+ model backend capacity status report async_overlap_json
2
+ G1 ultraep 2.40625 OK /workspace/tiera-rerun-fixed-moonep/ultraep/tierbtrace/tierbtrace_gbs16/G1__ultraep__matched__seq4096__gbs16__cap2p40625/nsys/G1__ultraep__matched__seq4096__gbs16__cap2p40625.nsys-rep