AbhishekGhosh1998 commited on
Commit
1da35b4
·
verified ·
1 Parent(s): d942369

granularity results tiera-rerun-fixed-moonep @ 2026-08-30T05:12:45Z from sp-aiinfra-interactive-aghosh-7jtjl-8xb200-moe-cg-0

Browse files
Files changed (25) hide show
  1. tiera-rerun-fixed-moonep/STATUS.txt +3 -3
  2. tiera-rerun-fixed-moonep/hf_watcher.log +3 -0
  3. tiera-rerun-fixed-moonep/steps.tsv +1 -0
  4. tiera-rerun-fixed-moonep/study.log +137 -0
  5. tiera-rerun-fixed-moonep/ultraep/tierbprobe/manifest.txt +19 -0
  6. tiera-rerun-fixed-moonep/ultraep/tierbprobe/pip_freeze.txt +317 -0
  7. tiera-rerun-fixed-moonep/ultraep/tierbprobe/provenance.txt +90 -0
  8. tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G1__ultraep__matched__seq4096__gbs16__cap1p25/rep1/mem_samples.csv +232 -0
  9. tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G1__ultraep__matched__seq4096__gbs16__cap1p25/rep1/run.env +70 -0
  10. tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G1__ultraep__matched__seq4096__gbs16__cap1p25/rep1/train.log +0 -0
  11. tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G1__ultraep__matched__seq4096__gbs16__cap1p8125/rep1/mem_samples.csv +232 -0
  12. tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G1__ultraep__matched__seq4096__gbs16__cap1p8125/rep1/run.env +70 -0
  13. tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G1__ultraep__matched__seq4096__gbs16__cap1p8125/rep1/train.log +0 -0
  14. tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G2__ultraep__matched__seq4096__gbs16__cap1p0625/rep1/mem_samples.csv +232 -0
  15. tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G2__ultraep__matched__seq4096__gbs16__cap1p0625/rep1/run.env +70 -0
  16. tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G2__ultraep__matched__seq4096__gbs16__cap1p0625/rep1/train.log +0 -0
  17. tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G2__ultraep__matched__seq4096__gbs16__cap1p25/rep1/mem_samples.csv +232 -0
  18. tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G2__ultraep__matched__seq4096__gbs16__cap1p25/rep1/run.env +70 -0
  19. tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G2__ultraep__matched__seq4096__gbs16__cap1p25/rep1/train.log +0 -0
  20. tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G2__ultraep__matched__seq4096__gbs16__cap1p25__forcebal/rep1/mem_samples.csv +152 -0
  21. tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G2__ultraep__matched__seq4096__gbs16__cap1p25__forcebal/rep1/run.env +70 -0
  22. tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G2__ultraep__matched__seq4096__gbs16__cap1p25__forcebal/rep1/train.log +1116 -0
  23. tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/capacity_attempts.tsv +5 -0
  24. tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/capacity_selection.tsv +2 -0
  25. tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/results.tsv +5 -0
tiera-rerun-fixed-moonep/STATUS.txt CHANGED
@@ -1,9 +1,9 @@
1
  study_root: /workspace/tiera-rerun-fixed-moonep
2
  gbs: 16
3
- updated: 2026-08-30T04:52:38+00:00
4
- current: echo__tierbprobe__gbs16
5
 
6
  completed steps:
7
 
8
  failed steps:
9
- (none)
 
1
  study_root: /workspace/tiera-rerun-fixed-moonep
2
  gbs: 16
3
+ updated: 2026-08-30T05:06:43+00:00
4
+ current: ultraep__tierbprobe__gbs16
5
 
6
  completed steps:
7
 
8
  failed steps:
9
+ FAIL echo tierbprobe 16 2026-08-30T04:52:38+00:00 2026-08-30T05:06:38+00:00 /workspace/tiera-rerun-fixed-moonep/echo/tierbprobe
tiera-rerun-fixed-moonep/hf_watcher.log CHANGED
@@ -4,3 +4,6 @@ Watching /workspace/tiera-rerun-fixed-moonep -> HF every 600s (kill to stop)
4
  uploaded: /workspace/tiera-rerun-fixed-moonep -> MoE-CUDA-Graph-Analysis/tierb-b200-gbs8/tiera-rerun-fixed-moonep
5
  https://huggingface.co/MoE-CUDA-Graph-Analysis/tierb-b200-gbs8/tree/main
6
  [2026-08-30T05:02:40+00:00] backup pass
 
 
 
 
4
  uploaded: /workspace/tiera-rerun-fixed-moonep -> MoE-CUDA-Graph-Analysis/tierb-b200-gbs8/tiera-rerun-fixed-moonep
5
  https://huggingface.co/MoE-CUDA-Graph-Analysis/tierb-b200-gbs8/tree/main
6
  [2026-08-30T05:02:40+00:00] backup pass
7
+ uploaded: /workspace/tiera-rerun-fixed-moonep -> MoE-CUDA-Graph-Analysis/tierb-b200-gbs8/tiera-rerun-fixed-moonep
8
+ https://huggingface.co/MoE-CUDA-Graph-Analysis/tierb-b200-gbs8/tree/main
9
+ [2026-08-30T05:12:45+00:00] backup pass
tiera-rerun-fixed-moonep/steps.tsv CHANGED
@@ -1 +1,2 @@
1
  status campaign phase gbs started ended result_dir
 
 
1
  status campaign phase gbs started ended result_dir
2
+ FAIL echo tierbprobe 16 2026-08-30T04:52:38+00:00 2026-08-30T05:06:38+00:00 /workspace/tiera-rerun-fixed-moonep/echo/tierbprobe
tiera-rerun-fixed-moonep/study.log CHANGED
@@ -52,3 +52,140 @@ driver_repo: /workspace/blackwell_run_scripts @ 774e9e2
52
  [guard] iter 12/12 (80s)
53
  --------------------------------------------------------------------------
54
  [05:02:24] CELL G4__echo__matched__seq4096__gbs16__cap2p00 (slots=16)
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
52
  [guard] iter 12/12 (80s)
53
  --------------------------------------------------------------------------
54
  [05:02:24] CELL G4__echo__matched__seq4096__gbs16__cap2p00 (slots=16)
55
+ [guard] iter 4/12 (70s)
56
+ [guard] iter 12/12 (80s)
57
+ [guard] tearing down process group (outcome=DONE)
58
+ [05:03:48] G4 echo: measured required=1.0166015625 at C=2.00 -> validating C=1.068359375 (safety=+5%)
59
+ --------------------------------------------------------------------------
60
+ [05:03:48] CELL G4__echo__matched__seq4096__gbs16__cap1p068359375 (slots=16)
61
+ [guard] iter 4/12 (70s)
62
+ [guard] iter 12/12 (80s)
63
+ [guard] tearing down process group (outcome=DONE)
64
+ --------------------------------------------------------------------------
65
+ [05:05:12] CELL G4__echo__matched__seq4096__gbs16__cap1p068359375__forcebal (slots=16)
66
+ [guard] iter 7/12 (70s)
67
+ [guard] iter 12/12 (80s)
68
+ [guard] tearing down process group (outcome=DONE)
69
+ --------------------------------------------------------------------------
70
+
71
+ ==============================================================
72
+ SWEEP COMPLETE ok=10 failed=0
73
+ ==============================================================
74
+ config backend policy slots seq gbs cap recompute rep status iter_ms_mean iter_ms_min peak_mib run_dir
75
+ G1 echo matched 1 4096 16 2.00 none 1 OK 671.53 666.90 125477 /workspace/tiera-rerun-fixed-moonep/echo/tierbprobe/tierbprobe_gbs16/G1__echo__matched__seq4096__gbs16__cap2p00/rep1
76
+ G2 echo matched 4 4096 16 2.00 none 1 OK 675.77 669.20 130289 /workspace/tiera-rerun-fixed-moonep/echo/tierbprobe/tierbprobe_gbs16/G2__echo__matched__seq4096__gbs16__cap2p00/rep1
77
+ G2 echo matched 4 4096 16 1.08203125 none 1 OK 660.43 642.00 119267 /workspace/tiera-rerun-fixed-moonep/echo/tierbprobe/tierbprobe_gbs16/G2__echo__matched__seq4096__gbs16__cap1p08203125/rep1
78
+ G2 echo matched 4 4096 16 1.08203125 none+forcebal 1 OK 628.07 616.10 119267 /workspace/tiera-rerun-fixed-moonep/echo/tierbprobe/tierbprobe_gbs16/G2__echo__matched__seq4096__gbs16__cap1p08203125__forcebal/rep1
79
+ G3 echo matched 8 4096 16 2.00 none 1 OK 721.76 713.60 136111 /workspace/tiera-rerun-fixed-moonep/echo/tierbprobe/tierbprobe_gbs16/G3__echo__matched__seq4096__gbs16__cap2p00/rep1
80
+ G3 echo matched 8 4096 16 1.072265625 none 1 OK 681.37 670.20 122853 /workspace/tiera-rerun-fixed-moonep/echo/tierbprobe/tierbprobe_gbs16/G3__echo__matched__seq4096__gbs16__cap1p072265625/rep1
81
+ G3 echo matched 8 4096 16 1.072265625 none+forcebal 1 OK 664.27 659.10 122899 /workspace/tiera-rerun-fixed-moonep/echo/tierbprobe/tierbprobe_gbs16/G3__echo__matched__seq4096__gbs16__cap1p072265625__forcebal/rep1
82
+ G4 echo matched 16 4096 16 2.00 none 1 OK 829.14 790.80 146037 /workspace/tiera-rerun-fixed-moonep/echo/tierbprobe/tierbprobe_gbs16/G4__echo__matched__seq4096__gbs16__cap2p00/rep1
83
+ G4 echo matched 16 4096 16 1.068359375 none 1 OK 775.73 738.90 129621 /workspace/tiera-rerun-fixed-moonep/echo/tierbprobe/tierbprobe_gbs16/G4__echo__matched__seq4096__gbs16__cap1p068359375/rep1
84
+ G4 echo matched 16 4096 16 1.068359375 none+forcebal 1 OK 768.04 733.40 129579 /workspace/tiera-rerun-fixed-moonep/echo/tierbprobe/tierbprobe_gbs16/G4__echo__matched__seq4096__gbs16__cap1p068359375__forcebal/rep1
85
+
86
+ Results: /workspace/tiera-rerun-fixed-moonep/echo/tierbprobe/tierbprobe_gbs16/results.tsv
87
+ Roll up: python3 /workspace/blackwell_run_scripts/granularity/tools/collect.py /workspace/tiera-rerun-fixed-moonep/echo/tierbprobe/tierbprobe_gbs16
88
+ ==============================================================
89
+ 10 probe runs -> /workspace/tiera-rerun-fixed-moonep/echo/tierbprobe/tierbprobe_gbs16/probe_summary.tsv
90
+ run arm cap rawExpI preRankI postRankI drop% reqCap
91
+ --------------------------------------------------------------------------------------------------------------------
92
+ G1__echo__matched__seq4096__gbs16__cap2p00/rep1 natural 2.00 3.6198* 3.6198 1.0827 1.9062
93
+ G2__echo__matched__seq4096__gbs16__cap1p08203125/rep natural 1.08203125 3.8720* 2.1703 1.0000 1.0273
94
+ G2__echo__matched__seq4096__gbs16__cap1p08203125__fo forced 1.08203125 1.0201 1.0069 1.0000 1.0234
95
+ G2__echo__matched__seq4096__gbs16__cap2p00/rep1 natural 2.00 3.8724* 2.1875 1.0000 1.0273
96
+ G3__echo__matched__seq4096__gbs16__cap1p072265625/re natural 1.072265625 3.9096* 1.7838 1.0000 1.0195
97
+ G3__echo__matched__seq4096__gbs16__cap1p072265625__f forced 1.072265625 1.0231 1.0048 1.0000 1.0176
98
+ G3__echo__matched__seq4096__gbs16__cap2p00/rep1 natural 2.00 3.9096* 1.7916 1.0000 1.0195
99
+ G4__echo__matched__seq4096__gbs16__cap1p068359375/re natural 1.068359375 3.9303* 1.5507 1.0000 1.0166
100
+ G4__echo__matched__seq4096__gbs16__cap1p068359375__f forced 1.068359375 1.0249 1.0034 1.0000 1.0137
101
+ G4__echo__matched__seq4096__gbs16__cap2p00/rep1 natural 2.00 3.9301* 1.5543 1.0000 1.0166
102
+
103
+ * At least one globally aggregated snapshot reached the E/K expert-Imax ceiling.
104
+ Inspect raw_expert_imax_mean and hottest_expert_token_fraction_mean before generalising.
105
+
106
+ ECHO note: echo_logged_pre_rank_metric_valid=0 and echo_logged_load_reduction_valid=0 are intentional.
107
+ The branch's built-in fields mix source-rank and destination-rank axes; this tool reconstructs both globally.
108
+
109
+ TIER B PROBE -- what these logs contain, per backend
110
+ ----------------------------------------------------
111
+ Capacity matrix:
112
+ ECHO: start C=2.00; on OOM descend 1.75, 1.50, 1.25. Stop at
113
+ the highest fitting zero-drop capacity. If a fitting capacity drops,
114
+ smaller values cannot help and the model is blocked.
115
+ UltraEP: measure required_capacity at C=1.25, round it UP to the next
116
+ 0.01, then validate that capacity once.
117
+ MoonEP: natural + forced, no capacity factor.
118
+
119
+ The forced arm runs once at the selected/high-headroom capacity. It validates
120
+ instrumentation and is not part of capacity selection.
121
+
122
+ Each backend instruments a different thing:
123
+
124
+ echo local raw-router histograms and local post-reroute placement.
125
+ reduce_probe.py aligns and sums all eight source ranks before
126
+ calculating raw_expert_*, pre_home_rank_*,
127
+ post_physical_rank_* and valid_max_rank_reduction_*.
128
+
129
+ ultraep global pre/post rank vectors plus explicit drop accounting:
130
+ drop_rate_pct and required_capacity -- the capacity factor the run
131
+ WOULD have needed. The vectors are duplicated identically on all
132
+ logging ranks and are deduplicated, not summed.
133
+
134
+ moonep local raw-router histograms, actual payload rows per physical
135
+ destination, and the bytes-copy side. The reducer globally sums
136
+ each (iteration, layer) snapshot first. ECHO reports no comparable
137
+ byte counters, so MoonEP bytes are a self-comparison across models,
138
+ never a cross-backend ranking.
139
+
140
+ Do NOT quote ECHO's logged pre_echo imbalance_ratio or
141
+ max_load_reduction_pct. The branch mixes source-rank and destination-rank
142
+ axes; the reducer marks both invalid and reconstructs the comparison globally.
143
+
144
+ Two boundary cases:
145
+ raw_expert_imax_max == E/K
146
+ -> at least one global snapshot had one expert in every
147
+ token's top-K set. Inspect the MEAN before generalising.
148
+ required_capacity or min-safe-capacity at the configured capacity
149
+ -> clipped by the buffer; it means "at or over", not
150
+ "fits comfortably".
151
+
152
+ tools/reduce_probe.py has already written probe_summary.tsv and .json here.
153
+
154
+ Campaign results: /workspace/tiera-rerun-fixed-moonep/echo/tierbprobe
155
+ Roll up: python3 /workspace/blackwell_run_scripts/granularity/tools/collect.py /workspace/tiera-rerun-fixed-moonep/echo/tierbprobe --compare
156
+ Already logged in to Hugging Face as: AbhishekGhosh1998
157
+ uploaded: /workspace/tiera-rerun-fixed-moonep/echo/tierbprobe -> MoE-CUDA-Graph-Analysis/tierb-b200-gbs8/tiera-rerun-fixed-moonep/echo/tierbprobe
158
+ https://huggingface.co/MoE-CUDA-Graph-Analysis/tierb-b200-gbs8/tree/main
159
+
160
+ === STEP ultraep__tierbprobe__gbs16
161
+ ==============================================================
162
+ CAMPAIGN: ultraep backend=ultraep phase=tierbprobe
163
+ Megatron: /workspace/Megatron-UltraEP @ ultraep-token-drop-analysis
164
+ GBS=16 seq=4096 repeats=3 nsys=1
165
+ results -> /workspace/tiera-rerun-fixed-moonep/ultraep/tierbprobe
166
+ ==============================================================
167
+ --------------------------------------------------------------------------
168
+ [05:06:59] CELL G1__ultraep__matched__seq4096__gbs16__cap1p25 (slots=1)
169
+ [guard] iter 4/12 (60s)
170
+ [guard] iter 12/12 (70s)
171
+ [guard] tearing down process group (outcome=DONE)
172
+ [05:08:13] G1 ultraep: validating C=1.8125 (safety=+5%)
173
+ --------------------------------------------------------------------------
174
+ [05:08:13] CELL G1__ultraep__matched__seq4096__gbs16__cap1p8125 (slots=1)
175
+ [guard] iter 4/12 (60s)
176
+ [guard] iter 12/12 (70s)
177
+ [guard] tearing down process group (outcome=DONE)
178
+ [05:09:28] G1 ultraep: C=1.8125 dropped; increasing headroom
179
+ --------------------------------------------------------------------------
180
+ [05:09:28] CELL G2__ultraep__matched__seq4096__gbs16__cap1p25 (slots=4)
181
+ [guard] iter 4/12 (60s)
182
+ [guard] iter 12/12 (70s)
183
+ [guard] tearing down process group (outcome=DONE)
184
+ [05:10:42] G2 ultraep: validating C=1.0625 (safety=+5%)
185
+ --------------------------------------------------------------------------
186
+ [05:10:42] CELL G2__ultraep__matched__seq4096__gbs16__cap1p0625 (slots=4)
187
+ [guard] iter 5/12 (60s)
188
+ [guard] iter 12/12 (70s)
189
+ [guard] tearing down process group (outcome=DONE)
190
+ --------------------------------------------------------------------------
191
+ [05:11:57] CELL G2__ultraep__matched__seq4096__gbs16__cap1p25__forcebal (slots=4)
tiera-rerun-fixed-moonep/ultraep/tierbprobe/manifest.txt ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ timestamp: 2026-08-30T05:06:58+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: 818ab62
7
+ campaign: ultraep
8
+ backend: ultraep
9
+ phase: tierbprobe
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-token-drop-analysis
18
+ megatron_branch_actual: ultraep-token-drop-analysis
19
+ megatron_commit: eaf829e7
tiera-rerun-fixed-moonep/ultraep/tierbprobe/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@b099fd8494b09e8c08ad34e26abf91fb6c83636d#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/tierbprobe/provenance.txt ADDED
@@ -0,0 +1,90 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ==============================================================
2
+ PROVENANCE: ultraep__tierbprobe__gbs16
3
+ ==============================================================
4
+ captured_at : 2026-08-30T05:06:44+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 tierbprobe --gbs 16 --out /workspace/tiera-rerun-fixed-moonep/ultraep/tierbprobe --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, 29
21
+ 1, 120 MHz, 1965 MHz, 1000.00 W, 34
22
+ 2, 120 MHz, 1965 MHz, 1000.00 W, 34
23
+ 3, 120 MHz, 1965 MHz, 1000.00 W, 29
24
+ 4, 120 MHz, 1965 MHz, 1000.00 W, 30
25
+ 5, 120 MHz, 1965 MHz, 1000.00 W, 34
26
+ 6, 120 MHz, 1965 MHz, 1000.00 W, 34
27
+ 7, 120 MHz, 1965 MHz, 1000.00 W, 29
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 token-drop-analysis b099fd8494b0
63
+ Megatron-UltraEP ultraep-token-drop-analysis eaf829e7ad21
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 818ab62ae4af
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/tierbprobe/tierbprobe_gbs16/G1__ultraep__matched__seq4096__gbs16__cap1p25/rep1/mem_samples.csv ADDED
@@ -0,0 +1,232 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 0, 0
2
+ 1, 0
3
+ 2, 0
4
+ 3, 0
5
+ 4, 0
6
+ 5, 0
7
+ 6, 0
8
+ 7, 0
9
+ 0, 0
10
+ 1, 0
11
+ 2, 0
12
+ 3, 0
13
+ 4, 0
14
+ 5, 0
15
+ 6, 0
16
+ 7, 0
17
+ 0, 4
18
+ 1, 1
19
+ 2, 1
20
+ 3, 1
21
+ 4, 1
22
+ 5, 1
23
+ 6, 1
24
+ 7, 1
25
+ 0, 4
26
+ 1, 4
27
+ 2, 4
28
+ 3, 4
29
+ 4, 2
30
+ 5, 1
31
+ 6, 1
32
+ 7, 1
33
+ 0, 4
34
+ 1, 4
35
+ 2, 4
36
+ 3, 4
37
+ 4, 4
38
+ 5, 4
39
+ 6, 4
40
+ 7, 4
41
+ 0, 819
42
+ 1, 819
43
+ 2, 819
44
+ 3, 819
45
+ 4, 819
46
+ 5, 819
47
+ 6, 819
48
+ 7, 819
49
+ 0, 1537
50
+ 1, 1537
51
+ 2, 1537
52
+ 3, 1537
53
+ 4, 1537
54
+ 5, 1537
55
+ 6, 1537
56
+ 7, 1537
57
+ 0, 2231
58
+ 1, 2231
59
+ 2, 2231
60
+ 3, 2231
61
+ 4, 2231
62
+ 5, 2231
63
+ 6, 2231
64
+ 7, 2231
65
+ 0, 2903
66
+ 1, 2903
67
+ 2, 2903
68
+ 3, 2903
69
+ 4, 2903
70
+ 5, 2903
71
+ 6, 2903
72
+ 7, 2903
73
+ 0, 42619
74
+ 1, 42635
75
+ 2, 42621
76
+ 3, 42607
77
+ 4, 42641
78
+ 5, 42607
79
+ 6, 42639
80
+ 7, 42645
81
+ 0, 43369
82
+ 1, 43357
83
+ 2, 43367
84
+ 3, 43355
85
+ 4, 43365
86
+ 5, 43367
87
+ 6, 43367
88
+ 7, 43349
89
+ 0, 44133
90
+ 1, 44131
91
+ 2, 44131
92
+ 3, 44131
93
+ 4, 44131
94
+ 5, 44131
95
+ 6, 44131
96
+ 7, 44131
97
+ 0, 44133
98
+ 1, 44131
99
+ 2, 44131
100
+ 3, 44131
101
+ 4, 44131
102
+ 5, 44131
103
+ 6, 44131
104
+ 7, 44131
105
+ 0, 44693
106
+ 1, 44691
107
+ 2, 44691
108
+ 3, 44691
109
+ 4, 44691
110
+ 5, 44691
111
+ 6, 44691
112
+ 7, 44691
113
+ 0, 44693
114
+ 1, 44691
115
+ 2, 44691
116
+ 3, 44691
117
+ 4, 44691
118
+ 5, 44691
119
+ 6, 44691
120
+ 7, 44691
121
+ 0, 70099
122
+ 1, 70097
123
+ 2, 70097
124
+ 3, 70097
125
+ 4, 70097
126
+ 5, 70097
127
+ 6, 70097
128
+ 7, 70097
129
+ 0, 70659
130
+ 1, 70657
131
+ 2, 70657
132
+ 3, 70657
133
+ 4, 70657
134
+ 5, 70657
135
+ 6, 70657
136
+ 7, 70657
137
+ 0, 70659
138
+ 1, 70657
139
+ 2, 70657
140
+ 3, 70657
141
+ 4, 70657
142
+ 5, 70657
143
+ 6, 70657
144
+ 7, 70657
145
+ 0, 82519
146
+ 1, 82519
147
+ 2, 82511
148
+ 3, 82523
149
+ 4, 82723
150
+ 5, 82529
151
+ 6, 82541
152
+ 7, 82551
153
+ 0, 83485
154
+ 1, 83483
155
+ 2, 83483
156
+ 3, 83483
157
+ 4, 83651
158
+ 5, 83483
159
+ 6, 83483
160
+ 7, 83483
161
+ 0, 89743
162
+ 1, 89921
163
+ 2, 90033
164
+ 3, 89937
165
+ 4, 90403
166
+ 5, 89971
167
+ 6, 89993
168
+ 7, 90015
169
+ 0, 107979
170
+ 1, 107945
171
+ 2, 107945
172
+ 3, 107945
173
+ 4, 107889
174
+ 5, 107945
175
+ 6, 107945
176
+ 7, 107945
177
+ 0, 107979
178
+ 1, 108731
179
+ 2, 108731
180
+ 3, 108731
181
+ 4, 108675
182
+ 5, 108731
183
+ 6, 108731
184
+ 7, 108731
185
+ 0, 107979
186
+ 1, 108731
187
+ 2, 108731
188
+ 3, 108731
189
+ 4, 108675
190
+ 5, 108731
191
+ 6, 108731
192
+ 7, 108731
193
+ 0, 107979
194
+ 1, 108731
195
+ 2, 108731
196
+ 3, 108731
197
+ 4, 108675
198
+ 5, 108731
199
+ 6, 108731
200
+ 7, 108731
201
+ 0, 107979
202
+ 1, 108731
203
+ 2, 108731
204
+ 3, 108731
205
+ 4, 108675
206
+ 5, 108731
207
+ 6, 108731
208
+ 7, 108731
209
+ 0, 107979
210
+ 1, 108731
211
+ 2, 108731
212
+ 3, 108731
213
+ 4, 108675
214
+ 5, 108731
215
+ 6, 108731
216
+ 7, 108731
217
+ 0, 103595
218
+ 1, 104609
219
+ 2, 104609
220
+ 3, 104609
221
+ 4, 104553
222
+ 5, 104673
223
+ 6, 104933
224
+ 7, 104929
225
+ 0, 193
226
+ 1, 104606
227
+ 2, 963
228
+ 3, 4
229
+ 4, 4
230
+ 5, 4
231
+ 6, 4
232
+ 7, 4
tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G1__ultraep__matched__seq4096__gbs16__cap1p25/rep1/run.env ADDED
@@ -0,0 +1,70 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ backend=ultraep
2
+ config=G1
3
+ family=tierA
4
+ repo=/workspace/Megatron-UltraEP
5
+ branch=ultraep-token-drop-analysis
6
+ commit=eaf829e7
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=1.25
45
+ cuda_graph=0
46
+ probe_mode=1
47
+ async_grad_reduce=0
48
+ branch_role=probe
49
+ memory_dump_mode=none
50
+ memory_snapshot_path=
51
+ cuda_graph_warmup_steps=3
52
+ recompute=none
53
+ force_balance=0
54
+ train_iters=12
55
+ ep_size=8
56
+ echo_dispatch_overlap=1
57
+ echo_dispatch_recompute=0
58
+ ultraep_quota_min_tokens=1024
59
+ ultraep_weight_sync_join=
60
+ profile_enable=0
61
+ kernel_trace=0
62
+ fp8_mode=mxfp8
63
+ fp8_recipe=mxfp8
64
+ training_seed=1234
65
+ learning_rate=1e-4
66
+ min_learning_rate=1.0e-5
67
+ lr_warmup_iters=5
68
+ state_dump_enable=0
69
+ state_dump_dir=
70
+ moonep_assert_no_pending_reduce=0
tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G1__ultraep__matched__seq4096__gbs16__cap1p25/rep1/train.log ADDED
The diff for this file is too large to render. See raw diff
 
tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G1__ultraep__matched__seq4096__gbs16__cap1p8125/rep1/mem_samples.csv ADDED
@@ -0,0 +1,232 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 0, 0
2
+ 1, 0
3
+ 2, 0
4
+ 3, 0
5
+ 4, 0
6
+ 5, 0
7
+ 6, 0
8
+ 7, 0
9
+ 0, 0
10
+ 1, 0
11
+ 2, 0
12
+ 3, 0
13
+ 4, 0
14
+ 5, 0
15
+ 6, 0
16
+ 7, 0
17
+ 0, 1
18
+ 1, 1
19
+ 2, 1
20
+ 3, 1
21
+ 4, 1
22
+ 5, 1
23
+ 6, 1
24
+ 7, 1
25
+ 0, 4
26
+ 1, 4
27
+ 2, 4
28
+ 3, 4
29
+ 4, 1
30
+ 5, 1
31
+ 6, 1
32
+ 7, 1
33
+ 0, 4
34
+ 1, 4
35
+ 2, 4
36
+ 3, 4
37
+ 4, 4
38
+ 5, 4
39
+ 6, 4
40
+ 7, 4
41
+ 0, 819
42
+ 1, 819
43
+ 2, 819
44
+ 3, 819
45
+ 4, 819
46
+ 5, 819
47
+ 6, 819
48
+ 7, 819
49
+ 0, 1537
50
+ 1, 1537
51
+ 2, 1537
52
+ 3, 1537
53
+ 4, 1537
54
+ 5, 1537
55
+ 6, 1537
56
+ 7, 1537
57
+ 0, 2231
58
+ 1, 2231
59
+ 2, 2231
60
+ 3, 2231
61
+ 4, 2231
62
+ 5, 2231
63
+ 6, 2231
64
+ 7, 2231
65
+ 0, 2903
66
+ 1, 2903
67
+ 2, 2901
68
+ 3, 2895
69
+ 4, 2903
70
+ 5, 2903
71
+ 6, 2903
72
+ 7, 2903
73
+ 0, 42699
74
+ 1, 42701
75
+ 2, 42651
76
+ 3, 42691
77
+ 4, 42683
78
+ 5, 42715
79
+ 6, 42691
80
+ 7, 42725
81
+ 0, 43433
82
+ 1, 43445
83
+ 2, 43421
84
+ 3, 43435
85
+ 4, 43437
86
+ 5, 43417
87
+ 6, 43473
88
+ 7, 43445
89
+ 0, 44133
90
+ 1, 44131
91
+ 2, 44131
92
+ 3, 44131
93
+ 4, 44131
94
+ 5, 44131
95
+ 6, 44131
96
+ 7, 44131
97
+ 0, 44133
98
+ 1, 44131
99
+ 2, 44131
100
+ 3, 44131
101
+ 4, 44131
102
+ 5, 44131
103
+ 6, 44131
104
+ 7, 44131
105
+ 0, 45351
106
+ 1, 45349
107
+ 2, 45349
108
+ 3, 45349
109
+ 4, 45349
110
+ 5, 45349
111
+ 6, 45349
112
+ 7, 45349
113
+ 0, 45351
114
+ 1, 45349
115
+ 2, 45349
116
+ 3, 45349
117
+ 4, 45349
118
+ 5, 45349
119
+ 6, 45349
120
+ 7, 45349
121
+ 0, 75333
122
+ 1, 75331
123
+ 2, 75331
124
+ 3, 75331
125
+ 4, 75331
126
+ 5, 75331
127
+ 6, 75331
128
+ 7, 75331
129
+ 0, 75739
130
+ 1, 75331
131
+ 2, 75331
132
+ 3, 75331
133
+ 4, 75331
134
+ 5, 75331
135
+ 6, 75331
136
+ 7, 75331
137
+ 0, 76551
138
+ 1, 76143
139
+ 2, 76143
140
+ 3, 76143
141
+ 4, 76143
142
+ 5, 76143
143
+ 6, 76143
144
+ 7, 76143
145
+ 0, 87535
146
+ 1, 87585
147
+ 2, 87597
148
+ 3, 87593
149
+ 4, 87565
150
+ 5, 87619
151
+ 6, 87639
152
+ 7, 87615
153
+ 0, 89213
154
+ 1, 89303
155
+ 2, 89303
156
+ 3, 89303
157
+ 4, 89223
158
+ 5, 89303
159
+ 6, 89303
160
+ 7, 89303
161
+ 0, 93655
162
+ 1, 93739
163
+ 2, 93761
164
+ 3, 93757
165
+ 4, 93723
166
+ 5, 93769
167
+ 6, 93815
168
+ 7, 93807
169
+ 0, 113545
170
+ 1, 113767
171
+ 2, 113767
172
+ 3, 113767
173
+ 4, 113667
174
+ 5, 113767
175
+ 6, 113767
176
+ 7, 113767
177
+ 0, 113545
178
+ 1, 113767
179
+ 2, 113767
180
+ 3, 113767
181
+ 4, 113667
182
+ 5, 113767
183
+ 6, 113767
184
+ 7, 113767
185
+ 0, 113545
186
+ 1, 113767
187
+ 2, 113767
188
+ 3, 113767
189
+ 4, 113667
190
+ 5, 113767
191
+ 6, 113767
192
+ 7, 113767
193
+ 0, 113545
194
+ 1, 113767
195
+ 2, 113767
196
+ 3, 113767
197
+ 4, 113667
198
+ 5, 113767
199
+ 6, 113767
200
+ 7, 113767
201
+ 0, 113545
202
+ 1, 113767
203
+ 2, 113767
204
+ 3, 113767
205
+ 4, 113667
206
+ 5, 113767
207
+ 6, 113767
208
+ 7, 113767
209
+ 0, 113545
210
+ 1, 113767
211
+ 2, 113767
212
+ 3, 113767
213
+ 4, 113667
214
+ 5, 113767
215
+ 6, 113767
216
+ 7, 113767
217
+ 0, 109775
218
+ 1, 109675
219
+ 2, 109661
220
+ 3, 109647
221
+ 4, 109545
222
+ 5, 109653
223
+ 6, 109755
224
+ 7, 110251
225
+ 0, 1
226
+ 1, 4
227
+ 2, 4
228
+ 3, 196
229
+ 4, 109545
230
+ 5, 964
231
+ 6, 4
232
+ 7, 4
tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G1__ultraep__matched__seq4096__gbs16__cap1p8125/rep1/run.env ADDED
@@ -0,0 +1,70 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ backend=ultraep
2
+ config=G1
3
+ family=tierA
4
+ repo=/workspace/Megatron-UltraEP
5
+ branch=ultraep-token-drop-analysis
6
+ commit=eaf829e7
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=1.8125
45
+ cuda_graph=0
46
+ probe_mode=1
47
+ async_grad_reduce=0
48
+ branch_role=probe
49
+ memory_dump_mode=none
50
+ memory_snapshot_path=
51
+ cuda_graph_warmup_steps=3
52
+ recompute=none
53
+ force_balance=0
54
+ train_iters=12
55
+ ep_size=8
56
+ echo_dispatch_overlap=1
57
+ echo_dispatch_recompute=0
58
+ ultraep_quota_min_tokens=1024
59
+ ultraep_weight_sync_join=
60
+ profile_enable=0
61
+ kernel_trace=0
62
+ fp8_mode=mxfp8
63
+ fp8_recipe=mxfp8
64
+ training_seed=1234
65
+ learning_rate=1e-4
66
+ min_learning_rate=1.0e-5
67
+ lr_warmup_iters=5
68
+ state_dump_enable=0
69
+ state_dump_dir=
70
+ moonep_assert_no_pending_reduce=0
tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G1__ultraep__matched__seq4096__gbs16__cap1p8125/rep1/train.log ADDED
The diff for this file is too large to render. See raw diff
 
tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G2__ultraep__matched__seq4096__gbs16__cap1p0625/rep1/mem_samples.csv ADDED
@@ -0,0 +1,232 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 0, 0
2
+ 1, 0
3
+ 2, 0
4
+ 3, 0
5
+ 4, 0
6
+ 5, 0
7
+ 6, 0
8
+ 7, 0
9
+ 0, 0
10
+ 1, 0
11
+ 2, 0
12
+ 3, 0
13
+ 4, 0
14
+ 5, 0
15
+ 6, 0
16
+ 7, 0
17
+ 0, 1
18
+ 1, 1
19
+ 2, 1
20
+ 3, 1
21
+ 4, 1
22
+ 5, 1
23
+ 6, 1
24
+ 7, 1
25
+ 0, 4
26
+ 1, 4
27
+ 2, 4
28
+ 3, 4
29
+ 4, 1
30
+ 5, 1
31
+ 6, 1
32
+ 7, 1
33
+ 0, 4
34
+ 1, 4
35
+ 2, 4
36
+ 3, 4
37
+ 4, 4
38
+ 5, 4
39
+ 6, 4
40
+ 7, 4
41
+ 0, 4
42
+ 1, 4
43
+ 2, 4
44
+ 3, 629
45
+ 4, 37
46
+ 5, 4
47
+ 6, 4
48
+ 7, 4
49
+ 0, 821
50
+ 1, 821
51
+ 2, 821
52
+ 3, 821
53
+ 4, 821
54
+ 5, 821
55
+ 6, 821
56
+ 7, 821
57
+ 0, 1895
58
+ 1, 1895
59
+ 2, 1895
60
+ 3, 1895
61
+ 4, 1895
62
+ 5, 1895
63
+ 6, 1895
64
+ 7, 1895
65
+ 0, 2399
66
+ 1, 2389
67
+ 2, 2401
68
+ 3, 2379
69
+ 4, 2393
70
+ 5, 2399
71
+ 6, 2399
72
+ 7, 2391
73
+ 0, 42337
74
+ 1, 42333
75
+ 2, 42333
76
+ 3, 42333
77
+ 4, 42333
78
+ 5, 42333
79
+ 6, 42333
80
+ 7, 42333
81
+ 0, 42743
82
+ 1, 42733
83
+ 2, 42733
84
+ 3, 42733
85
+ 4, 42733
86
+ 5, 42733
87
+ 6, 42733
88
+ 7, 42733
89
+ 0, 44049
90
+ 1, 44041
91
+ 2, 44041
92
+ 3, 44041
93
+ 4, 44041
94
+ 5, 44041
95
+ 6, 44041
96
+ 7, 44041
97
+ 0, 44321
98
+ 1, 44313
99
+ 2, 44313
100
+ 3, 44313
101
+ 4, 44313
102
+ 5, 44313
103
+ 6, 44313
104
+ 7, 44313
105
+ 0, 44321
106
+ 1, 44313
107
+ 2, 44313
108
+ 3, 44313
109
+ 4, 44313
110
+ 5, 44313
111
+ 6, 44313
112
+ 7, 44313
113
+ 0, 45721
114
+ 1, 45713
115
+ 2, 45713
116
+ 3, 45713
117
+ 4, 45713
118
+ 5, 45713
119
+ 6, 45713
120
+ 7, 45713
121
+ 0, 62317
122
+ 1, 62311
123
+ 2, 62311
124
+ 3, 62309
125
+ 4, 62313
126
+ 5, 62037
127
+ 6, 62313
128
+ 7, 62311
129
+ 0, 70515
130
+ 1, 70563
131
+ 2, 70563
132
+ 3, 70563
133
+ 4, 70521
134
+ 5, 70563
135
+ 6, 70563
136
+ 7, 70563
137
+ 0, 70991
138
+ 1, 71039
139
+ 2, 71039
140
+ 3, 71039
141
+ 4, 70997
142
+ 5, 71039
143
+ 6, 71039
144
+ 7, 71039
145
+ 0, 70991
146
+ 1, 71039
147
+ 2, 71039
148
+ 3, 71039
149
+ 4, 70997
150
+ 5, 71039
151
+ 6, 71039
152
+ 7, 71039
153
+ 0, 82919
154
+ 1, 82957
155
+ 2, 82959
156
+ 3, 82975
157
+ 4, 82967
158
+ 5, 82963
159
+ 6, 82985
160
+ 7, 82997
161
+ 0, 83857
162
+ 1, 83921
163
+ 2, 83939
164
+ 3, 83933
165
+ 4, 83991
166
+ 5, 83923
167
+ 6, 83957
168
+ 7, 83937
169
+ 0, 84961
170
+ 1, 85043
171
+ 2, 85027
172
+ 3, 85045
173
+ 4, 85047
174
+ 5, 85033
175
+ 6, 85059
176
+ 7, 85043
177
+ 0, 107987
178
+ 1, 107977
179
+ 2, 107977
180
+ 3, 107977
181
+ 4, 107915
182
+ 5, 107977
183
+ 6, 107977
184
+ 7, 107977
185
+ 0, 107987
186
+ 1, 107977
187
+ 2, 107977
188
+ 3, 107977
189
+ 4, 107915
190
+ 5, 107977
191
+ 6, 107977
192
+ 7, 107977
193
+ 0, 107987
194
+ 1, 107977
195
+ 2, 107977
196
+ 3, 107977
197
+ 4, 107915
198
+ 5, 107977
199
+ 6, 107977
200
+ 7, 107977
201
+ 0, 107071
202
+ 1, 107239
203
+ 2, 107123
204
+ 3, 107323
205
+ 4, 107309
206
+ 5, 107061
207
+ 6, 107061
208
+ 7, 107231
209
+ 0, 103143
210
+ 1, 103201
211
+ 2, 102813
212
+ 3, 102813
213
+ 4, 102815
214
+ 5, 103459
215
+ 6, 103197
216
+ 7, 103459
217
+ 0, 100755
218
+ 1, 101197
219
+ 2, 101779
220
+ 3, 101453
221
+ 4, 101071
222
+ 5, 100745
223
+ 6, 101133
224
+ 7, 101133
225
+ 0, 1
226
+ 1, 65
227
+ 2, 101746
228
+ 3, 321
229
+ 4, 4
230
+ 5, 4
231
+ 6, 4
232
+ 7, 4
tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G2__ultraep__matched__seq4096__gbs16__cap1p0625/rep1/run.env ADDED
@@ -0,0 +1,70 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ backend=ultraep
2
+ config=G2
3
+ family=tierA
4
+ repo=/workspace/Megatron-UltraEP
5
+ branch=ultraep-token-drop-analysis
6
+ commit=eaf829e7
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=1.0625
45
+ cuda_graph=0
46
+ probe_mode=1
47
+ async_grad_reduce=0
48
+ branch_role=probe
49
+ memory_dump_mode=none
50
+ memory_snapshot_path=
51
+ cuda_graph_warmup_steps=3
52
+ recompute=none
53
+ force_balance=0
54
+ train_iters=12
55
+ ep_size=8
56
+ echo_dispatch_overlap=1
57
+ echo_dispatch_recompute=0
58
+ ultraep_quota_min_tokens=1024
59
+ ultraep_weight_sync_join=
60
+ profile_enable=0
61
+ kernel_trace=0
62
+ fp8_mode=mxfp8
63
+ fp8_recipe=mxfp8
64
+ training_seed=1234
65
+ learning_rate=1e-4
66
+ min_learning_rate=1.0e-5
67
+ lr_warmup_iters=5
68
+ state_dump_enable=0
69
+ state_dump_dir=
70
+ moonep_assert_no_pending_reduce=0
tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G2__ultraep__matched__seq4096__gbs16__cap1p0625/rep1/train.log ADDED
The diff for this file is too large to render. See raw diff
 
tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G2__ultraep__matched__seq4096__gbs16__cap1p25/rep1/mem_samples.csv ADDED
@@ -0,0 +1,232 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 0, 0
2
+ 1, 0
3
+ 2, 0
4
+ 3, 0
5
+ 4, 0
6
+ 5, 0
7
+ 6, 0
8
+ 7, 0
9
+ 0, 0
10
+ 1, 0
11
+ 2, 0
12
+ 3, 0
13
+ 4, 0
14
+ 5, 0
15
+ 6, 0
16
+ 7, 0
17
+ 0, 2
18
+ 1, 1
19
+ 2, 1
20
+ 3, 1
21
+ 4, 1
22
+ 5, 1
23
+ 6, 1
24
+ 7, 1
25
+ 0, 4
26
+ 1, 4
27
+ 2, 4
28
+ 3, 4
29
+ 4, 2
30
+ 5, 1
31
+ 6, 1
32
+ 7, 1
33
+ 0, 4
34
+ 1, 4
35
+ 2, 4
36
+ 3, 4
37
+ 4, 4
38
+ 5, 4
39
+ 6, 4
40
+ 7, 4
41
+ 0, 4
42
+ 1, 4
43
+ 2, 4
44
+ 3, 4
45
+ 4, 4
46
+ 5, 4
47
+ 6, 4
48
+ 7, 4
49
+ 0, 821
50
+ 1, 821
51
+ 2, 821
52
+ 3, 821
53
+ 4, 821
54
+ 5, 821
55
+ 6, 821
56
+ 7, 821
57
+ 0, 1895
58
+ 1, 1895
59
+ 2, 1895
60
+ 3, 1895
61
+ 4, 1895
62
+ 5, 1895
63
+ 6, 1895
64
+ 7, 1895
65
+ 0, 2387
66
+ 1, 2389
67
+ 2, 2403
68
+ 3, 2423
69
+ 4, 2407
70
+ 5, 2397
71
+ 6, 2409
72
+ 7, 2425
73
+ 0, 42337
74
+ 1, 42333
75
+ 2, 42333
76
+ 3, 42333
77
+ 4, 42333
78
+ 5, 42333
79
+ 6, 42333
80
+ 7, 42333
81
+ 0, 42743
82
+ 1, 42733
83
+ 2, 42733
84
+ 3, 42733
85
+ 4, 42733
86
+ 5, 42733
87
+ 6, 42733
88
+ 7, 42733
89
+ 0, 44049
90
+ 1, 44041
91
+ 2, 44041
92
+ 3, 44041
93
+ 4, 44041
94
+ 5, 44041
95
+ 6, 44041
96
+ 7, 44041
97
+ 0, 44049
98
+ 1, 44041
99
+ 2, 44041
100
+ 3, 44041
101
+ 4, 44041
102
+ 5, 44041
103
+ 6, 44041
104
+ 7, 44041
105
+ 0, 44369
106
+ 1, 44361
107
+ 2, 44361
108
+ 3, 44361
109
+ 4, 44361
110
+ 5, 44361
111
+ 6, 44361
112
+ 7, 44361
113
+ 0, 46011
114
+ 1, 46003
115
+ 2, 46003
116
+ 3, 46003
117
+ 4, 46003
118
+ 5, 46003
119
+ 6, 46003
120
+ 7, 46003
121
+ 0, 56661
122
+ 1, 56653
123
+ 2, 56653
124
+ 3, 56653
125
+ 4, 56653
126
+ 5, 56653
127
+ 6, 56653
128
+ 7, 56653
129
+ 0, 72849
130
+ 1, 72841
131
+ 2, 72841
132
+ 3, 72841
133
+ 4, 72841
134
+ 5, 72841
135
+ 6, 72841
136
+ 7, 72841
137
+ 0, 73409
138
+ 1, 73401
139
+ 2, 73401
140
+ 3, 73401
141
+ 4, 73401
142
+ 5, 73401
143
+ 6, 73401
144
+ 7, 73401
145
+ 0, 73409
146
+ 1, 73401
147
+ 2, 73401
148
+ 3, 73401
149
+ 4, 73401
150
+ 5, 73401
151
+ 6, 73401
152
+ 7, 73401
153
+ 0, 85139
154
+ 1, 85189
155
+ 2, 85205
156
+ 3, 85219
157
+ 4, 85131
158
+ 5, 85171
159
+ 6, 85233
160
+ 7, 85199
161
+ 0, 86239
162
+ 1, 86301
163
+ 2, 86301
164
+ 3, 86301
165
+ 4, 86231
166
+ 5, 86301
167
+ 6, 86301
168
+ 7, 86301
169
+ 0, 87255
170
+ 1, 87315
171
+ 2, 87331
172
+ 3, 87345
173
+ 4, 87263
174
+ 5, 87333
175
+ 6, 87317
176
+ 7, 87297
177
+ 0, 110343
178
+ 1, 110333
179
+ 2, 110333
180
+ 3, 110333
181
+ 4, 110319
182
+ 5, 110333
183
+ 6, 110333
184
+ 7, 110333
185
+ 0, 110343
186
+ 1, 110333
187
+ 2, 110333
188
+ 3, 110333
189
+ 4, 110319
190
+ 5, 110333
191
+ 6, 110333
192
+ 7, 110333
193
+ 0, 110343
194
+ 1, 110333
195
+ 2, 110333
196
+ 3, 110333
197
+ 4, 110319
198
+ 5, 110333
199
+ 6, 110333
200
+ 7, 110333
201
+ 0, 109427
202
+ 1, 109417
203
+ 2, 109481
204
+ 3, 110063
205
+ 4, 109723
206
+ 5, 109417
207
+ 6, 109417
208
+ 7, 109417
209
+ 0, 105631
210
+ 1, 105815
211
+ 2, 105941
212
+ 3, 105815
213
+ 4, 105475
214
+ 5, 105621
215
+ 6, 105815
216
+ 7, 105489
217
+ 0, 103563
218
+ 1, 104135
219
+ 2, 103873
220
+ 3, 104135
221
+ 4, 103795
222
+ 5, 103553
223
+ 6, 104135
224
+ 7, 103809
225
+ 0, 65
226
+ 1, 104134
227
+ 2, 324
228
+ 3, 4
229
+ 4, 4
230
+ 5, 4
231
+ 6, 4
232
+ 7, 4
tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G2__ultraep__matched__seq4096__gbs16__cap1p25/rep1/run.env ADDED
@@ -0,0 +1,70 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ backend=ultraep
2
+ config=G2
3
+ family=tierA
4
+ repo=/workspace/Megatron-UltraEP
5
+ branch=ultraep-token-drop-analysis
6
+ commit=eaf829e7
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=1.25
45
+ cuda_graph=0
46
+ probe_mode=1
47
+ async_grad_reduce=0
48
+ branch_role=probe
49
+ memory_dump_mode=none
50
+ memory_snapshot_path=
51
+ cuda_graph_warmup_steps=3
52
+ recompute=none
53
+ force_balance=0
54
+ train_iters=12
55
+ ep_size=8
56
+ echo_dispatch_overlap=1
57
+ echo_dispatch_recompute=0
58
+ ultraep_quota_min_tokens=1024
59
+ ultraep_weight_sync_join=
60
+ profile_enable=0
61
+ kernel_trace=0
62
+ fp8_mode=mxfp8
63
+ fp8_recipe=mxfp8
64
+ training_seed=1234
65
+ learning_rate=1e-4
66
+ min_learning_rate=1.0e-5
67
+ lr_warmup_iters=5
68
+ state_dump_enable=0
69
+ state_dump_dir=
70
+ moonep_assert_no_pending_reduce=0
tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G2__ultraep__matched__seq4096__gbs16__cap1p25/rep1/train.log ADDED
The diff for this file is too large to render. See raw diff
 
tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G2__ultraep__matched__seq4096__gbs16__cap1p25__forcebal/rep1/mem_samples.csv ADDED
@@ -0,0 +1,152 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 0, 0
2
+ 1, 0
3
+ 2, 0
4
+ 3, 0
5
+ 4, 0
6
+ 5, 0
7
+ 6, 0
8
+ 7, 0
9
+ 0, 0
10
+ 1, 0
11
+ 2, 0
12
+ 3, 0
13
+ 4, 0
14
+ 5, 0
15
+ 6, 0
16
+ 7, 0
17
+ 0, 1
18
+ 1, 1
19
+ 2, 1
20
+ 3, 1
21
+ 4, 1
22
+ 5, 1
23
+ 6, 1
24
+ 7, 1
25
+ 0, 4
26
+ 1, 4
27
+ 2, 4
28
+ 3, 4
29
+ 4, 1
30
+ 5, 1
31
+ 6, 1
32
+ 7, 1
33
+ 0, 4
34
+ 1, 4
35
+ 2, 4
36
+ 3, 4
37
+ 4, 4
38
+ 5, 4
39
+ 6, 4
40
+ 7, 4
41
+ 0, 4
42
+ 1, 629
43
+ 2, 629
44
+ 3, 629
45
+ 4, 629
46
+ 5, 629
47
+ 6, 629
48
+ 7, 629
49
+ 0, 913
50
+ 1, 903
51
+ 2, 917
52
+ 3, 933
53
+ 4, 929
54
+ 5, 933
55
+ 6, 941
56
+ 7, 925
57
+ 0, 1895
58
+ 1, 1895
59
+ 2, 1895
60
+ 3, 1895
61
+ 4, 1895
62
+ 5, 1895
63
+ 6, 1895
64
+ 7, 1895
65
+ 0, 2349
66
+ 1, 2351
67
+ 2, 2343
68
+ 3, 2329
69
+ 4, 2339
70
+ 5, 2353
71
+ 6, 2359
72
+ 7, 2359
73
+ 0, 42337
74
+ 1, 42333
75
+ 2, 42333
76
+ 3, 42333
77
+ 4, 42333
78
+ 5, 42333
79
+ 6, 42333
80
+ 7, 42333
81
+ 0, 42743
82
+ 1, 42733
83
+ 2, 42733
84
+ 3, 42733
85
+ 4, 42733
86
+ 5, 42733
87
+ 6, 42733
88
+ 7, 42733
89
+ 0, 44049
90
+ 1, 44041
91
+ 2, 44041
92
+ 3, 44041
93
+ 4, 44041
94
+ 5, 44041
95
+ 6, 44041
96
+ 7, 44041
97
+ 0, 44369
98
+ 1, 44361
99
+ 2, 44361
100
+ 3, 44361
101
+ 4, 44361
102
+ 5, 44361
103
+ 6, 44361
104
+ 7, 44361
105
+ 0, 44369
106
+ 1, 44361
107
+ 2, 44361
108
+ 3, 44361
109
+ 4, 44361
110
+ 5, 44361
111
+ 6, 44361
112
+ 7, 44361
113
+ 0, 46011
114
+ 1, 46003
115
+ 2, 46003
116
+ 3, 46003
117
+ 4, 46003
118
+ 5, 46003
119
+ 6, 46003
120
+ 7, 46003
121
+ 0, 72849
122
+ 1, 72841
123
+ 2, 72841
124
+ 3, 72841
125
+ 4, 72841
126
+ 5, 72841
127
+ 6, 72841
128
+ 7, 72841
129
+ 0, 72849
130
+ 1, 72841
131
+ 2, 72841
132
+ 3, 72841
133
+ 4, 72841
134
+ 5, 72841
135
+ 6, 72841
136
+ 7, 72841
137
+ 0, 73409
138
+ 1, 73401
139
+ 2, 73401
140
+ 3, 73401
141
+ 4, 73401
142
+ 5, 73401
143
+ 6, 73401
144
+ 7, 73401
145
+ 0, 84977
146
+ 1, 85037
147
+ 2, 85031
148
+ 3, 85039
149
+ 4, 84971
150
+ 5, 85043
151
+ 6, 85053
152
+ 7, 85041
tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G2__ultraep__matched__seq4096__gbs16__cap1p25__forcebal/rep1/run.env ADDED
@@ -0,0 +1,70 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ backend=ultraep
2
+ config=G2
3
+ family=tierA
4
+ repo=/workspace/Megatron-UltraEP
5
+ branch=ultraep-token-drop-analysis
6
+ commit=eaf829e7
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=1.25
45
+ cuda_graph=0
46
+ probe_mode=1
47
+ async_grad_reduce=0
48
+ branch_role=probe
49
+ memory_dump_mode=none
50
+ memory_snapshot_path=
51
+ cuda_graph_warmup_steps=3
52
+ recompute=none
53
+ force_balance=1
54
+ train_iters=12
55
+ ep_size=8
56
+ echo_dispatch_overlap=1
57
+ echo_dispatch_recompute=0
58
+ ultraep_quota_min_tokens=1024
59
+ ultraep_weight_sync_join=
60
+ profile_enable=0
61
+ kernel_trace=0
62
+ fp8_mode=mxfp8
63
+ fp8_recipe=mxfp8
64
+ training_seed=1234
65
+ learning_rate=1e-4
66
+ min_learning_rate=1.0e-5
67
+ lr_warmup_iters=5
68
+ state_dump_enable=0
69
+ state_dump_dir=
70
+ moonep_assert_no_pending_reduce=0
tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G2__ultraep__matched__seq4096__gbs16__cap1p25__forcebal/rep1/train.log ADDED
@@ -0,0 +1,1116 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [05:11:57] WARN: PROBE_MODE forces CUDA_GRAPH=0 (routing instrumentation is not capture-safe)
2
+ ==============================================================
3
+ G2 / ultraep (slots/rank=4 policy=matched)
4
+ E=32 K=8 H'=3584 m=4
5
+ L=16 seq=4096 gbs=16 (2 microbatches)
6
+ cap_mode=static cap=1.25 cg=0 recompute=none
7
+ memory_dump=none snapshots=/workspace/tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G2__ultraep__matched__seq4096__gbs16__cap1p25__forcebal/rep1/memory_snapshots/snapshot.pickle
8
+ repo=/workspace/Megatron-UltraEP @ eaf829e7
9
+ ==============================================================
10
+ W0830 05:11:58.814000 492296 torch/distributed/run.py:766]
11
+ W0830 05:11:58.814000 492296 torch/distributed/run.py:766] *****************************************
12
+ W0830 05:11:58.814000 492296 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.
13
+ W0830 05:11:58.814000 492296 torch/distributed/run.py:766] *****************************************
14
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
15
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
16
+ 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
17
+ using world size: 8, data-parallel size: 8, context-parallel size: 1, hierarchical context-parallel sizes: None, tensor-model-parallel size: 1, pipeline-model-parallel size: 1
18
+ WARNING: overriding default arguments for tokenizer_type:GPT2BPETokenizer with tokenizer_type:GPT2BPETokenizer
19
+ Number of virtual stages per pipeline stage: None
20
+ accumulate and all-reduce gradients in fp32 for bfloat16 data type.
21
+ using torch.bfloat16 for parameters ...
22
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
23
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
24
+ ------------------------ arguments ------------------------
25
+ account_for_embedding_in_pipeline_split ......... False
26
+ account_for_loss_in_pipeline_split .............. False
27
+ accumulate_allreduce_grads_in_fp32 .............. True
28
+ adam_beta1 ...................................... 0.9
29
+ adam_beta2 ...................................... 0.999
30
+ adam_eps ........................................ 1e-08
31
+ add_bias_linear ................................. False
32
+ add_position_embedding .......................... False
33
+ add_qkv_bias .................................... False
34
+ adlr_autoresume ................................. False
35
+ adlr_autoresume_interval ........................ 1000
36
+ align_grad_reduce ............................... True
37
+ align_param_gather .............................. False
38
+ app_tag_run_name ................................ None
39
+ app_tag_run_version ............................. 0.0.0
40
+ apply_layernorm_1p .............................. False
41
+ apply_query_key_layer_scaling ................... False
42
+ apply_residual_connection_post_layernorm ........ False
43
+ apply_rope_fusion ............................... True
44
+ async_save ...................................... None
45
+ async_tensor_model_parallel_allreduce ........... True
46
+ attention_backend ............................... AttnBackend.auto
47
+ attention_dropout ............................... 0.0
48
+ attention_softmax_in_fp32 ....................... False
49
+ auto_detect_ckpt_format ......................... False
50
+ barrier_with_L1_time ............................ True
51
+ bert_binary_head ................................ True
52
+ bert_embedder_type .............................. megatron
53
+ bert_load ....................................... None
54
+ bf16 ............................................ True
55
+ bias_dropout_fusion ............................. True
56
+ bias_gelu_fusion ................................ False
57
+ bias_swiglu_fusion .............................. True
58
+ biencoder_projection_dim ........................ 0
59
+ biencoder_shared_query_context_model ............ False
60
+ block_data_path ................................. None
61
+ calc_ft_timeouts ................................ False
62
+ calculate_per_token_loss ........................ False
63
+ check_for_large_grads ........................... False
64
+ check_for_nan_in_loss_and_grad .................. True
65
+ check_for_spiky_loss ............................ False
66
+ check_weight_hash_across_dp_replicas_interval ... None
67
+ ckpt_assume_constant_structure .................. False
68
+ ckpt_convert_format ............................. None
69
+ ckpt_convert_save ............................... None
70
+ ckpt_convert_update_legacy_dist_opt_format ...... False
71
+ ckpt_format ..................................... torch_dist
72
+ ckpt_fully_parallel_load ........................ False
73
+ ckpt_fully_parallel_save ........................ True
74
+ ckpt_fully_parallel_save_deprecated ............. False
75
+ ckpt_step ....................................... None
76
+ classes_fraction ................................ 1.0
77
+ clip_grad ....................................... 1.0
78
+ clone_scatter_output_in_embedding ............... True
79
+ config_logger_dir ...............................
80
+ consumed_train_samples .......................... 0
81
+ consumed_valid_samples .......................... 0
82
+ context_parallel_size ........................... 1
83
+ cp_comm_type .................................... ['p2p']
84
+ create_attention_mask_in_dataloader ............. True
85
+ cross_entropy_fusion_impl ....................... native
86
+ cross_entropy_loss_fusion ....................... False
87
+ cuda_graph_scope ................................ full
88
+ cuda_graph_warmup_steps ......................... 3
89
+ data_args_path .................................. None
90
+ data_cache_path ................................. None
91
+ data_parallel_random_init ....................... False
92
+ data_parallel_sharding_strategy ................. no_shard
93
+ data_parallel_size .............................. 8
94
+ data_path ....................................... ['/workspace/data/wikitext/wikitext103_gpt2_text_document']
95
+ data_per_class_fraction ......................... 1.0
96
+ data_sharding ................................... True
97
+ dataloader_type ................................. single
98
+ ddp_average_in_collective ....................... False
99
+ ddp_bucket_size ................................. None
100
+ ddp_num_buckets ................................. None
101
+ ddp_pad_buckets_for_high_nccl_busbw ............. False
102
+ decoder_first_pipeline_num_layers ............... None
103
+ decoder_last_pipeline_num_layers ................ None
104
+ decoder_num_layers .............................. None
105
+ decoder_seq_length .............................. None
106
+ decoupled_lr .................................... None
107
+ decoupled_min_lr ................................ None
108
+ decrease_batch_size_if_needed ................... False
109
+ defer_embedding_wgrad_compute ................... False
110
+ delay_wgrad_compute ............................. False
111
+ deprecated_use_mcore_models ..................... True
112
+ deterministic_mode .............................. False
113
+ dino_bottleneck_size ............................ 256
114
+ dino_freeze_last_layer .......................... 1
115
+ dino_head_hidden_size ........................... 2048
116
+ dino_local_crops_number ......................... 10
117
+ dino_local_img_size ............................. 96
118
+ dino_norm_last_layer ............................ False
119
+ dino_teacher_temp ............................... 0.07
120
+ dino_warmup_teacher_temp ........................ 0.04
121
+ dino_warmup_teacher_temp_epochs ................. 30
122
+ disable_bf16_reduced_precision_matmul ........... False
123
+ disable_mamba_mem_eff_path ...................... False
124
+ disable_straggler_on_startup .................... False
125
+ dist_ckpt_format_deprecated ..................... None
126
+ dist_ckpt_strictness ............................ assume_ok_unexpected
127
+ distribute_saved_activations .................... False
128
+ distributed_backend ............................. nccl
129
+ distributed_timeout_minutes ..................... 10
130
+ embedding_init_method_std ....................... None
131
+ embedding_path .................................. None
132
+ empty_unused_memory_level ....................... 0
133
+ enable_cuda_graph ............................... False
134
+ enable_experimental ............................. False
135
+ enable_ft_package ............................... False
136
+ enable_gloo_process_groups ...................... True
137
+ enable_msc ...................................... True
138
+ enable_one_logger ............................... True
139
+ encoder_num_layers .............................. 16
140
+ encoder_seq_length .............................. 4096
141
+ end_weight_decay ................................ 0.1
142
+ eod_mask_loss ................................... False
143
+ error_injection_rate ............................ 0
144
+ error_injection_type ............................ transient_error
145
+ eval_interval ................................... 1000000
146
+ eval_iters ...................................... 0
147
+ evidence_data_path .............................. None
148
+ exit_duration_in_mins ........................... None
149
+ exit_interval ................................... None
150
+ exit_on_missing_checkpoint ...................... False
151
+ exit_signal_handler ............................. False
152
+ exp_avg_dtype ................................... torch.float32
153
+ exp_avg_sq_dtype ................................ torch.float32
154
+ expert_model_parallel_size ...................... 8
155
+ expert_tensor_parallel_size ..................... 1
156
+ export_force_local_attention .................... False
157
+ export_kd_cfg ................................... None
158
+ export_kd_teacher_ckpt_format ................... None
159
+ export_kd_teacher_load .......................... None
160
+ export_kv_cache_quant ........................... False
161
+ export_legacy_megatron .......................... False
162
+ export_model_type ............................... GPTModel
163
+ export_moe_apply_probs_on_input ................. False
164
+ export_qk_l2_norm ............................... False
165
+ export_quant_cfg ................................ None
166
+ export_real_quant_cfg ........................... None
167
+ export_te_mcore_model ........................... False
168
+ external_cuda_graph ............................. False
169
+ ffn_hidden_size ................................. 14336
170
+ finetune ........................................ False
171
+ finetune_data_split ............................. train
172
+ finetune_hf_dataset ............................. None
173
+ first_last_layers_bf16 .......................... False
174
+ flash_decode .................................... False
175
+ fp16 ............................................ False
176
+ fp16_lm_cross_entropy ........................... False
177
+ fp32_residual_connection ........................ False
178
+ fp8 ............................................. e4m3
179
+ fp8_amax_compute_algo ........................... most_recent
180
+ fp8_amax_history_len ............................ 1
181
+ fp8_interval .................................... 1
182
+ fp8_margin ...................................... 0
183
+ fp8_param_gather ................................ False
184
+ fp8_recipe ...................................... mxfp8
185
+ fp8_wgrad ....................................... True
186
+ fsdp_double_buffer .............................. False
187
+ global_batch_size ............................... 16
188
+ grad_reduce_in_bf16 ............................. False
189
+ gradient_accumulation_fusion .................... True
190
+ gradient_reduce_div_fusion ...................... True
191
+ group_query_attention ........................... True
192
+ head_lr_mult .................................... 1.0
193
+ heterogeneous_layers_config_encoded_json ........ None
194
+ heterogeneous_layers_config_path ................ None
195
+ hidden_dropout .................................. 0.0
196
+ hidden_size ..................................... 4096
197
+ hierarchical_context_parallel_sizes ............. None
198
+ high_priority_stream_groups ..................... []
199
+ hybrid_attention_ratio .......................... 0.0
200
+ hybrid_mlp_ratio ................................ 0.0
201
+ hybrid_override_pattern ......................... None
202
+ hysteresis ...................................... 2
203
+ ict_head_size ................................... None
204
+ ict_load ........................................ None
205
+ img_h ........................................... 224
206
+ img_w ........................................... 224
207
+ indexer_batch_size .............................. 128
208
+ indexer_log_interval ............................ 1000
209
+ inference_batch_times_seqlen_threshold .......... -1
210
+ inference_dynamic_batching ...................... False
211
+ inference_dynamic_batching_buffer_guaranteed_fraction 0.2
212
+ inference_dynamic_batching_buffer_overflow_factor None
213
+ inference_dynamic_batching_buffer_size_gb ....... 40.0
214
+ inference_dynamic_batching_chunk_size ........... 256
215
+ inference_dynamic_batching_max_requests_override None
216
+ inference_dynamic_batching_max_tokens_override .. None
217
+ inference_dynamic_batching_num_cuda_graphs ...... 16
218
+ inference_max_batch_size ........................ 8
219
+ inference_max_seq_length ........................ 2560
220
+ inference_rng_tracker ........................... False
221
+ init_method_std ................................. 0.01
222
+ init_method_xavier_uniform ...................... False
223
+ init_model_with_meta_device ..................... False
224
+ initial_loss_scale .............................. 4294967296
225
+ inprocess_active_world_size ..................... 8
226
+ inprocess_barrier_timeout ....................... 120
227
+ inprocess_completion_timeout .................... 120
228
+ inprocess_empty_cuda_cache ...................... False
229
+ inprocess_granularity ........................... node
230
+ inprocess_hard_timeout .......................... 90
231
+ inprocess_heartbeat_interval .................... 30
232
+ inprocess_heartbeat_timeout ..................... 60
233
+ inprocess_last_call_wait ........................ 1
234
+ inprocess_max_iterations ........................ None
235
+ inprocess_monitor_process_interval .............. 1.0
236
+ inprocess_monitor_thread_interval ............... 1.0
237
+ inprocess_progress_watchdog_interval ............ 1.0
238
+ inprocess_restart ............................... False
239
+ inprocess_soft_timeout .......................... 60
240
+ inprocess_termination_grace_time ................ 1
241
+ is_hybrid_model ................................. False
242
+ iter_per_epoch .................................. 1250
243
+ iterations_to_skip .............................. []
244
+ keep_fp8_transpose_cache_when_using_custom_fsdp . False
245
+ kitchen_config_file ............................. None
246
+ kitchen_recipe_number ........................... None
247
+ kv_channels ..................................... 128
248
+ kv_lora_rank .................................... 32
249
+ lazy_mpu_init ................................... None
250
+ load ............................................ None
251
+ load_main_params_from_ckpt ...................... None
252
+ load_model_opt_format ........................... False
253
+ local_rank ...................................... 0
254
+ log_energy ...................................... False
255
+ log_interval .................................... 1
256
+ log_loss_scale_to_tensorboard ................... True
257
+ log_memory_to_tensorboard ....................... False
258
+ log_num_zeros_in_grad ........................... False
259
+ log_params_norm ................................. False
260
+ log_progress .................................... False
261
+ log_straggler ................................... False
262
+ log_throughput .................................. True
263
+ log_timers_to_tensorboard ....................... False
264
+ log_validation_ppl_to_tensorboard ............... False
265
+ log_world_size_to_tensorboard ................... False
266
+ logging_level ................................... None
267
+ loss_scale ...................................... None
268
+ loss_scale_window ............................... 1000
269
+ lr .............................................. 0.0001
270
+ lr_decay_iters .................................. 8
271
+ lr_decay_samples ................................ None
272
+ lr_decay_style .................................. cosine
273
+ lr_warmup_fraction .............................. None
274
+ lr_warmup_init .................................. 0.0
275
+ lr_warmup_iters ................................. 5
276
+ lr_warmup_samples ............................... 0
277
+ lr_wsd_decay_iters .............................. None
278
+ lr_wsd_decay_samples ............................ None
279
+ lr_wsd_decay_style .............................. exponential
280
+ main_grads_dtype ................................ torch.float32
281
+ main_params_dtype ............................... torch.float32
282
+ make_vocab_size_divisible_by .................... 128
283
+ mamba_head_dim .................................. 64
284
+ mamba_num_groups ................................ 8
285
+ mamba_num_heads ................................. None
286
+ mamba_state_dim ................................. 128
287
+ manual_gc ....................................... False
288
+ manual_gc_eval .................................. True
289
+ manual_gc_interval .............................. 0
290
+ mask_factor ..................................... 1.0
291
+ mask_prob ....................................... 0.15
292
+ mask_type ....................................... random
293
+ masked_softmax_fusion ........................... False
294
+ max_position_embeddings ......................... 32768
295
+ max_tokens_to_oom ............................... 12000
296
+ memory_snapshot_path ............................ snapshot.pickle
297
+ merge_file ...................................... /workspace/data/wikitext/gpt2-merges.txt
298
+ micro_batch_size ................................ 1
299
+ microbatch_group_size_per_vp_stage .............. None
300
+ mid_level_dataset_surplus ....................... 0.005
301
+ min_loss_scale .................................. 1.0
302
+ min_lr .......................................... 1e-05
303
+ mlp_chunks_for_prefill .......................... 1
304
+ mmap_bin_files .................................. True
305
+ mock_data ....................................... False
306
+ moe_apply_probs_on_input ........................ False
307
+ moe_aux_loss_coeff .............................. 0.01
308
+ moe_capacity_monitor ............................ True
309
+ moe_debug_drop_json ............................. None
310
+ moe_debug_drop_logging .......................... True
311
+ moe_debug_drop_logging_interval ................. 1
312
+ moe_deepep_num_sms .............................. 20
313
+ moe_echo_enable_random_offloading ............... False
314
+ moe_echo_expert_dispatch_overlap ................ False
315
+ moe_echo_expert_dispatcher_type ................. hybridep
316
+ moe_echo_recompute_expert_dispatch .............. False
317
+ moe_enable_deepep ............................... False
318
+ moe_enable_echo ................................. False
319
+ moe_enable_hybridep ............................. True
320
+ moe_enable_ultraep .............................. True
321
+ moe_expert_capacity_factor ...................... None
322
+ moe_extended_tp ................................. False
323
+ moe_ffn_hidden_size ............................. 3584
324
+ moe_grouped_gemm ................................ True
325
+ moe_hybridep_num_sms ............................ 16
326
+ moe_input_jitter_eps ............................ None
327
+ moe_layer_freq .................................. 1
328
+ moe_layer_recompute ............................. False
329
+ moe_num_echo_experts ............................ None
330
+ moe_num_redundant_experts_per_rank .............. 4
331
+ moe_pad_expert_input_to_capacity ................ False
332
+ moe_per_layer_logging ........................... False
333
+ moe_permute_fusion .............................. False
334
+ moe_received_token_capacity ..................... 1.25
335
+ moe_router_bias_update_rate ..................... 0.001
336
+ moe_router_dtype ................................ fp32
337
+ moe_router_enable_expert_bias ................... False
338
+ moe_router_force_load_balancing ................. True
339
+ moe_router_group_topk ........................... None
340
+ moe_router_load_balancing_type .................. aux_loss
341
+ moe_router_num_groups ........................... None
342
+ moe_router_padding_for_fp8 ...................... False
343
+ moe_router_pre_softmax .......................... False
344
+ moe_router_score_function ....................... softmax
345
+ moe_router_topk ................................. 8
346
+ moe_router_topk_scaling_factor .................. None
347
+ moe_shared_expert_intermediate_size ............. None
348
+ moe_shared_expert_overlap ....................... False
349
+ moe_token_dispatcher_type ....................... flex
350
+ moe_token_drop_policy ........................... probs
351
+ moe_ultraep_max_microbatches .................... None
352
+ moe_ultraep_overlap_weight_sync ................. True
353
+ moe_ultraep_weight_sync_join .................... before_dispatch
354
+ moe_upcycling_granularity ....................... 1
355
+ moe_use_device_initiated_grouped_gemm ........... True
356
+ moe_use_legacy_grouped_gemm ..................... False
357
+ moe_use_upcycling ............................... False
358
+ moe_z_loss_coeff ................................ None
359
+ mrope_section ................................... None
360
+ mscale .......................................... 1.0
361
+ mscale_all_dim .................................. 0.0
362
+ mtp_loss_scaling_factor ......................... 0.1
363
+ mtp_num_layers .................................. None
364
+ multi_latent_attention .......................... False
365
+ nccl_all_reduce_for_prefill ..................... False
366
+ nccl_communicator_config_path ................... None
367
+ nccl_ub ......................................... False
368
+ no_load_optim ................................... None
369
+ no_load_rng ..................................... None
370
+ no_persist_layer_norm ........................... False
371
+ no_rope_freq .................................... None
372
+ no_save_optim ................................... None
373
+ no_save_rng ..................................... None
374
+ non_persistent_ckpt_type ........................ None
375
+ non_persistent_global_ckpt_dir .................. None
376
+ non_persistent_local_ckpt_algo .................. fully_parallel
377
+ non_persistent_local_ckpt_dir ................... None
378
+ non_persistent_save_interval .................... None
379
+ norm_epsilon .................................... 1e-05
380
+ normalization ................................... RMSNorm
381
+ num_attention_heads ............................. 32
382
+ num_channels .................................... 3
383
+ num_classes ..................................... 1000
384
+ num_dataset_builder_threads ..................... 1
385
+ num_distributed_optimizer_instances ............. 1
386
+ num_experts ..................................... 32
387
+ num_layers ...................................... 16
388
+ num_layers_at_end_in_bf16 ....................... 1
389
+ num_layers_at_start_in_bf16 ..................... 1
390
+ num_layers_per_virtual_pipeline_stage ........... None
391
+ num_query_groups ................................ 8
392
+ num_virtual_stages_per_pipeline_rank ............ None
393
+ num_workers ..................................... 2
394
+ object_storage_cache_path ....................... None
395
+ one_logger_async ................................ False
396
+ one_logger_project .............................. megatron-lm
397
+ one_logger_run_name ............................. None
398
+ onnx_safe ....................................... None
399
+ openai_gelu ..................................... False
400
+ optimizer ....................................... adam
401
+ optimizer_cpu_offload ........................... False
402
+ optimizer_offload_fraction ...................... 1.0
403
+ output_bert_embeddings .......................... False
404
+ overlap_cpu_optimizer_d2h_h2d ................... False
405
+ overlap_grad_reduce ............................. True
406
+ overlap_moe_expert_parallel_comm ................ False
407
+ overlap_p2p_comm ................................ False
408
+ overlap_p2p_comm_warmup_flush ................... False
409
+ overlap_param_gather ............................ True
410
+ overlap_param_gather_with_optimizer_step ........ False
411
+ override_opt_param_scheduler .................... False
412
+ params_dtype .................................... torch.bfloat16
413
+ patch_dim ....................................... 16
414
+ per_split_data_args_path ........................ None
415
+ perform_initialization .......................... True
416
+ pin_cpu_grads ................................... True
417
+ pin_cpu_params .................................. True
418
+ pipeline_model_parallel_comm_backend ............ None
419
+ pipeline_model_parallel_layout .................. None
420
+ pipeline_model_parallel_size .................... 1
421
+ position_embedding_type ......................... rope
422
+ pretrained_checkpoint ........................... None
423
+ profile ......................................... False
424
+ profile_ranks ................................... [0]
425
+ profile_step_end ................................ 12
426
+ profile_step_start .............................. 10
427
+ q_lora_rank ..................................... None
428
+ qk_head_dim ..................................... 128
429
+ qk_l2_norm ...................................... False
430
+ qk_layernorm .................................... False
431
+ qk_pos_emb_head_dim ............................. 64
432
+ query_in_block_prob ............................. 0.1
433
+ rampup_batch_size ............................... None
434
+ rank ............................................ 0
435
+ recompute_granularity ........................... None
436
+ recompute_method ................................ None
437
+ recompute_modules ............................... None
438
+ recompute_num_layers ............................ None
439
+ record_memory_history ........................... False
440
+ relative_attention_max_distance ................. 128
441
+ relative_attention_num_buckets .................. 32
442
+ replication ..................................... False
443
+ replication_factor .............................. 2
444
+ replication_jump ................................ None
445
+ rerun_mode ...................................... validate_results
446
+ reset_attention_mask ............................ False
447
+ reset_position_ids .............................. False
448
+ result_rejected_tracker_filename ................ None
449
+ retriever_report_topk_accuracies ................ []
450
+ retriever_score_scaling ......................... False
451
+ retriever_seq_length ............................ 256
452
+ retro_add_retriever ............................. False
453
+ retro_attention_gate ............................ 1
454
+ retro_cyclic_train_iters ........................ None
455
+ retro_encoder_attention_dropout ................. 0.1
456
+ retro_encoder_hidden_dropout .................... 0.1
457
+ retro_encoder_layers ............................ 2
458
+ retro_num_neighbors ............................. 2
459
+ retro_num_retrieved_chunks ...................... 2
460
+ retro_project_dir ............................... None
461
+ retro_verify_neighbor_count ..................... True
462
+ reuse_grad_buf_for_mxfp8_param_ag ............... False
463
+ rope_scaling_factor ............................. 8.0
464
+ rope_type ....................................... None
465
+ rotary_base ..................................... 1000000
466
+ rotary_interleaved .............................. False
467
+ rotary_percent .................................. 1.0
468
+ rotary_scaling_factor ........................... 1.0
469
+ rotary_seq_len_interpolation_factor ............. None
470
+ run_workload_inspector_server ................... False
471
+ sample_rate ..................................... 1.0
472
+ save ............................................ None
473
+ save_interval ................................... None
474
+ save_retain_interval ............................ None
475
+ scatter_gather_tensors_in_pipeline .............. True
476
+ seed ............................................ 1234
477
+ seq_length ...................................... 4096
478
+ sequence_parallel ............................... False
479
+ sft ............................................. False
480
+ sft_tokenizer_prompt_format ..................... nemotron-h-aligned
481
+ sgd_momentum .................................... 0.9
482
+ short_seq_prob .................................. 0.1
483
+ skip_train ...................................... False
484
+ skipped_train_samples ........................... 0
485
+ spec ............................................ None
486
+ split ........................................... 100,0,0
487
+ squared_relu .................................... False
488
+ start_weight_decay .............................. 0.1
489
+ straggler_ctrlr_port ............................ 65535
490
+ straggler_minmax_count .......................... 1
491
+ suggested_communication_unit_size ............... None
492
+ swiglu .......................................... True
493
+ swin_backbone_type .............................. tiny
494
+ symmetric_ar_type ............................... None
495
+ te_rng_tracker .................................. False
496
+ tensor_model_parallel_size ...................... 1
497
+ tensorboard_dir ................................. None
498
+ tensorboard_log_interval ........................ 1
499
+ tensorboard_queue_size .......................... 1000
500
+ test_data_path .................................. None
501
+ test_mode ....................................... False
502
+ tiktoken_num_special_tokens ..................... 1000
503
+ tiktoken_pattern ................................ None
504
+ tiktoken_special_tokens ......................... None
505
+ timing_log_level ................................ 0
506
+ timing_log_option ............................... minmax
507
+ titles_data_path ................................ None
508
+ tokenizer_model ................................. None
509
+ tokenizer_type .................................. GPT2BPETokenizer
510
+ torch_fsdp2_reshard_after_forward ............... True
511
+ tp_comm_bootstrap_backend ....................... nccl
512
+ tp_comm_bulk_dgrad .............................. True
513
+ tp_comm_bulk_wgrad .............................. True
514
+ tp_comm_overlap ................................. False
515
+ tp_comm_overlap_ag .............................. True
516
+ tp_comm_overlap_cfg ............................. None
517
+ tp_comm_overlap_rs .............................. True
518
+ tp_comm_overlap_rs_dgrad ........................ False
519
+ tp_comm_split_ag ................................ True
520
+ tp_comm_split_rs ................................ True
521
+ train_data_path ................................. None
522
+ train_iters ..................................... 12
523
+ train_samples ................................... None
524
+ train_sync_interval ............................. None
525
+ transformer_impl ................................ transformer_engine
526
+ transformer_pipeline_model_parallel_size ........ 1
527
+ untie_embeddings_and_output_weights ............. True
528
+ use_checkpoint_args ............................. False
529
+ use_checkpoint_opt_param_scheduler .............. False
530
+ use_cpu_initialization .......................... None
531
+ use_custom_fsdp ................................. False
532
+ use_dist_ckpt ................................... True
533
+ use_dist_ckpt_deprecated ........................ False
534
+ use_distributed_optimizer ....................... True
535
+ use_flash_attn .................................. False
536
+ use_legacy_models ............................... False
537
+ use_mp_args_from_checkpoint_args ................ False
538
+ use_one_sent_docs ............................... False
539
+ use_persistent_ckpt_worker ...................... False
540
+ use_precision_aware_optimizer ................... False
541
+ use_pytorch_profiler ............................ False
542
+ use_ring_exchange_p2p ........................... False
543
+ use_rope_scaling ................................ False
544
+ use_rotary_position_embeddings .................. False
545
+ use_sharp ....................................... False
546
+ use_tokenizer_model_from_checkpoint_args ........ True
547
+ use_torch_fsdp2 ................................. False
548
+ use_torch_optimizer_for_cpu_offload ............. False
549
+ use_tp_pp_dp_mapping ............................ False
550
+ v_head_dim ...................................... 128
551
+ valid_data_path ................................. None
552
+ variable_seq_lengths ............................ False
553
+ virtual_pipeline_model_parallel_size ............ None
554
+ vision_backbone_type ............................ vit
555
+ vision_pretraining .............................. False
556
+ vision_pretraining_type ......................... classify
557
+ vocab_extra_ids ................................. 0
558
+ vocab_file ...................................... /workspace/data/wikitext/gpt2-vocab.json
559
+ vocab_size ...................................... None
560
+ wandb_exp_name ..................................
561
+ wandb_project ...................................
562
+ wandb_save_dir ..................................
563
+ weight_decay .................................... 0.1
564
+ weight_decay_incr_style ......................... constant
565
+ wgrad_deferral_limit ............................ 0
566
+ world_size ...................................... 8
567
+ yaml_cfg ........................................ None
568
+ -------------------- end of arguments ---------------------
569
+ INFO:megatron.core.num_microbatches_calculator:setting number of microbatches to constant 2
570
+ > building GPT2BPETokenizer tokenizer ...
571
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
572
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
573
+ > padded vocab (size: 50257) with 47 dummy tokens (new size: 50304)
574
+ WARNING:megatron.core.rerun_state_machine:RerunStateMachine initialized in mode RerunMode.VALIDATE_RESULTS
575
+ > initializing torch distributed ...
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
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
583
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
584
+ /workspace/Megatron-UltraEP/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
585
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
586
+ [Gloo] Rank [Gloo] Rank 0 is connected to 77 is connected to peer ranks. 7Expected number of connected peer ranks is : peer ranks. 7Expected number of connected peer ranks is : [Gloo] Rank
587
+ 72
588
+ is connected to [Gloo] Rank 7 peer ranks. Expected number of connected peer ranks is : 17 is connected to
589
+ 7 peer ranks. Expected number of connected peer ranks is : 7
590
+ [Gloo] Rank 4 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
591
+ [Gloo] Rank 3 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
592
+ [Gloo] Rank [Gloo] Rank 6 is connected to 75 peer ranks. is connected to Expected number of connected peer ranks is : 77 peer ranks.
593
+ Expected number of connected peer ranks is : 7
594
+ [Gloo] Rank 4 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
595
+ [Gloo] Rank 7 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
596
+ [Gloo] Rank 0 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
597
+ [Gloo] Rank 1 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
598
+ [Gloo] Rank 2 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
599
+ [Gloo] Rank 3 is connected to 7[Gloo] Rank peer ranks. 6[Gloo] Rank Expected number of connected peer ranks is : is connected to 577 is connected to peer ranks.
600
+ 7Expected number of connected peer ranks is : peer ranks. 7Expected number of connected peer ranks is :
601
+ 7
602
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
603
+ > initialized tensor model parallel with size 1
604
+ > initialized pipeline model parallel with size 1
605
+ > setting random seeds to 1234 ...
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
+ [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
610
+
611
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
612
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
613
+ [rank7]:[W830 05:12:17.285927019 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 7] using GPU 7 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
614
+ [rank6]:[W830 05:12:17.286012001 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
+ [rank3]:[W830 05:12:17.286134516 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 3] using GPU 3 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
616
+ [rank4]:[W830 05:12:17.286198702 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.
617
+ [rank2]:[W830 05:12:17.286272222 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.
618
+ [rank5]:[W830 05:12:17.286962301 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.
619
+ [rank1]:[W830 05:12:17.287092983 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.
620
+ > compiling dataset index builder ...
621
+ make: Entering directory '/workspace/Megatron-UltraEP/megatron/core/datasets'
622
+ make: Nothing to be done for 'default'.
623
+ make: Leaving directory '/workspace/Megatron-UltraEP/megatron/core/datasets'
624
+ >>> done with dataset index builder. Compilation time: 0.034 seconds
625
+ WARNING: constraints for invoking optimized fused softmax kernel are not met. We default back to unfused kernel invocations.
626
+ > compiling and loading fused kernels ...
627
+ [rank0]:[W830 05:12:17.455273505 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.
628
+ >>> done with compiling and loading fused kernels. Compilation time: 3.741 seconds
629
+ time to initialize megatron (seconds): 7.181
630
+ [after megatron is initialized] datetime: 2026-08-30 05:12:22
631
+ building GPT model ...
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
+ /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.
645
+ warnings.warn(
646
+ /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.
647
+ warnings.warn(
648
+ [INFO] Use MNNVL fabric: False
649
+ [WARN] MAX_NUM_NVL_PEERS is not set. Using detected value 8.
650
+ NVSHMEM v3.7.2
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
+ /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.
664
+
665
+ /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.
666
+
667
+ UltraEP: NVL domain size <= 8; forcing weight_sync plan mode to direct (was adaptive).
668
+ > number of parameters on (tensor, pipeline) model parallel rank (0, 0): 6722555904
669
+ 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=True, 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)
670
+ INFO:megatron.core.distributed.param_and_grad_buffer:Number of buckets for gradient all-reduce / reduce-scatter: 18
671
+ Params for bucket 1 (206045184 elements, 206045184 padded size):
672
+ module.output_layer.weight
673
+ Params for bucket 2 (42086400 elements, 42086400 padded size):
674
+ module.decoder.layers.15.mlp.router.weight
675
+ module.decoder.layers.15.self_attention.linear_qkv.weight
676
+ module.decoder.layers.15.self_attention.linear_qkv.layer_norm_weight
677
+ module.decoder.layers.15.self_attention.linear_proj.weight
678
+ module.decoder.final_layernorm.weight
679
+ module.decoder.layers.15.pre_mlp_layernorm.weight
680
+ Params for bucket 3 (42082304 elements, 42082304 padded size):
681
+ module.decoder.layers.14.pre_mlp_layernorm.weight
682
+ module.decoder.layers.14.mlp.router.weight
683
+ module.decoder.layers.14.self_attention.linear_proj.weight
684
+ module.decoder.layers.14.self_attention.linear_qkv.layer_norm_weight
685
+ module.decoder.layers.14.self_attention.linear_qkv.weight
686
+ Params for bucket 4 (42082304 elements, 42082304 padded size):
687
+ module.decoder.layers.13.pre_mlp_layernorm.weight
688
+ module.decoder.layers.13.mlp.router.weight
689
+ module.decoder.layers.13.self_attention.linear_proj.weight
690
+ module.decoder.layers.13.self_attention.linear_qkv.layer_norm_weight
691
+ module.decoder.layers.13.self_attention.linear_qkv.weight
692
+ Params for bucket 5 (42082304 elements, 42082304 padded size):
693
+ module.decoder.layers.12.pre_mlp_layernorm.weight
694
+ module.decoder.layers.12.mlp.router.weight
695
+ module.decoder.layers.12.self_attention.linear_proj.weight
696
+ module.decoder.layers.12.self_attention.linear_qkv.layer_norm_weight
697
+ module.decoder.layers.12.self_attention.linear_qkv.weight
698
+ Params for bucket 6 (42082304 elements, 42082304 padded size):
699
+ module.decoder.layers.11.mlp.router.weight
700
+ module.decoder.layers.11.self_attention.linear_qkv.weight
701
+ module.decoder.layers.11.self_attention.linear_proj.weight
702
+ module.decoder.layers.11.self_attention.linear_qkv.layer_norm_weight
703
+ module.decoder.layers.11.pre_mlp_layernorm.weight
704
+ Params for bucket 7 (42082304 elements, 42082304 padded size):
705
+ module.decoder.layers.10.mlp.router.weight
706
+ module.decoder.layers.10.self_attention.linear_qkv.weight
707
+ module.decoder.layers.10.self_attention.linear_proj.weight
708
+ module.decoder.layers.10.self_attention.linear_qkv.layer_norm_weight
709
+ module.decoder.layers.10.pre_mlp_layernorm.weight
710
+ Params for bucket 8 (42082304 elements, 42082304 padded size):
711
+ module.decoder.layers.9.pre_mlp_layernorm.weight
712
+ module.decoder.layers.9.mlp.router.weight
713
+ module.decoder.layers.9.self_attention.linear_qkv.weight
714
+ module.decoder.layers.9.self_attention.linear_proj.weight
715
+ module.decoder.layers.9.self_attention.linear_qkv.layer_norm_weight
716
+ Params for bucket 9 (42082304 elements, 42082304 padded size):
717
+ module.decoder.layers.8.mlp.router.weight
718
+ module.decoder.layers.8.self_attention.linear_qkv.layer_norm_weight
719
+ module.decoder.layers.8.self_attention.linear_qkv.weight
720
+ module.decoder.layers.8.self_attention.linear_proj.weight
721
+ module.decoder.layers.8.pre_mlp_layernorm.weight
722
+ Params for bucket 10 (42082304 elements, 42082304 padded size):
723
+ module.decoder.layers.7.pre_mlp_layernorm.weight
724
+ module.decoder.layers.7.mlp.router.weight
725
+ module.decoder.layers.7.self_attention.linear_qkv.layer_norm_weight
726
+ module.decoder.layers.7.self_attention.linear_qkv.weight
727
+ module.decoder.layers.7.self_attention.linear_proj.weight
728
+ Params for bucket 11 (42082304 elements, 42082304 padded size):
729
+ module.decoder.layers.6.pre_mlp_layernorm.weight
730
+ module.decoder.layers.6.mlp.router.weight
731
+ module.decoder.layers.6.self_attention.linear_qkv.layer_norm_weight
732
+ module.decoder.layers.6.self_attention.linear_proj.weight
733
+ module.decoder.layers.6.self_attention.linear_qkv.weight
734
+ Params for bucket 12 (42082304 elements, 42082304 padded size):
735
+ module.decoder.layers.5.mlp.router.weight
736
+ module.decoder.layers.5.self_attention.linear_qkv.weight
737
+ module.decoder.layers.5.self_attention.linear_qkv.layer_norm_weight
738
+ module.decoder.layers.5.pre_mlp_layernorm.weight
739
+ module.decoder.layers.5.self_attention.linear_proj.weight
740
+ Params for bucket 13 (42082304 elements, 42082304 padded size):
741
+ module.decoder.layers.4.pre_mlp_layernorm.weight
742
+ module.decoder.layers.4.mlp.router.weight
743
+ module.decoder.layers.4.self_attention.linear_qkv.weight
744
+ module.decoder.layers.4.self_attention.linear_proj.weight
745
+ module.decoder.layers.4.self_attention.linear_qkv.layer_norm_weight
746
+ Params for bucket 14 (42082304 elements, 42082304 padded size):
747
+ module.decoder.layers.3.pre_mlp_layernorm.weight
748
+ module.decoder.layers.3.mlp.router.weight
749
+ module.decoder.layers.3.self_attention.linear_qkv.weight
750
+ module.decoder.layers.3.self_attention.linear_proj.weight
751
+ module.decoder.layers.3.self_attention.linear_qkv.layer_norm_weight
752
+ Params for bucket 15 (42082304 elements, 42082304 padded size):
753
+ module.decoder.layers.2.pre_mlp_layernorm.weight
754
+ module.decoder.layers.2.mlp.router.weight
755
+ module.decoder.layers.2.self_attention.linear_qkv.weight
756
+ module.decoder.layers.2.self_attention.linear_proj.weight
757
+ module.decoder.layers.2.self_attention.linear_qkv.layer_norm_weight
758
+ Params for bucket 16 (42082304 elements, 42082304 padded size):
759
+ module.decoder.layers.1.pre_mlp_layernorm.weight
760
+ module.decoder.layers.1.mlp.router.weight
761
+ module.decoder.layers.1.self_attention.linear_proj.weight
762
+ module.decoder.layers.1.self_attention.linear_qkv.weight
763
+ module.decoder.layers.1.self_attention.linear_qkv.layer_norm_weight
764
+ Params for bucket 17 (42082304 elements, 42082304 padded size):
765
+ module.decoder.layers.0.self_attention.linear_qkv.layer_norm_weight
766
+ module.decoder.layers.0.self_attention.linear_proj.weight
767
+ module.decoder.layers.0.mlp.router.weight
768
+ module.decoder.layers.0.pre_mlp_layernorm.weight
769
+ module.decoder.layers.0.self_attention.linear_qkv.weight
770
+ Params for bucket 18 (206045184 elements, 206045184 padded size):
771
+ module.embedding.word_embeddings.weight
772
+ INFO:megatron.core.distributed.param_and_grad_buffer:Number of buckets for gradient all-reduce / reduce-scatter: 56
773
+ Params for bucket 1 (44040192 elements, 44040192 padded size):
774
+ module.decoder.layers.15.mlp.experts.linear_fc2.weight3
775
+ module.decoder.layers.15.mlp.experts.linear_fc2.weight2
776
+ module.decoder.layers.15.mlp.experts.linear_fc2.weight1
777
+ Params for bucket 2 (44040192 elements, 44040192 padded size):
778
+ module.decoder.layers.15.mlp.experts.linear_fc2.weight0
779
+ module.decoder.layers.15.mlp.experts.linear_fc1.weight3
780
+ Params for bucket 3 (58720256 elements, 58720256 padded size):
781
+ module.decoder.layers.15.mlp.experts.linear_fc1.weight2
782
+ module.decoder.layers.15.mlp.experts.linear_fc1.weight1
783
+ Params for bucket 4 (44040192 elements, 44040192 padded size):
784
+ module.decoder.layers.15.mlp.experts.linear_fc1.weight0
785
+ module.decoder.layers.14.mlp.experts.linear_fc2.weight3
786
+ Params for bucket 5 (44040192 elements, 44040192 padded size):
787
+ module.decoder.layers.14.mlp.experts.linear_fc2.weight2
788
+ module.decoder.layers.14.mlp.experts.linear_fc2.weight0
789
+ module.decoder.layers.14.mlp.experts.linear_fc2.weight1
790
+ Params for bucket 6 (58720256 elements, 58720256 padded size):
791
+ module.decoder.layers.14.mlp.experts.linear_fc1.weight3
792
+ module.decoder.layers.14.mlp.experts.linear_fc1.weight2
793
+ Params for bucket 7 (58720256 elements, 58720256 padded size):
794
+ module.decoder.layers.14.mlp.experts.linear_fc1.weight1
795
+ module.decoder.layers.14.mlp.experts.linear_fc1.weight0
796
+ Params for bucket 8 (44040192 elements, 44040192 padded size):
797
+ module.decoder.layers.13.mlp.experts.linear_fc2.weight3
798
+ module.decoder.layers.13.mlp.experts.linear_fc2.weight2
799
+ module.decoder.layers.13.mlp.experts.linear_fc2.weight1
800
+ Params for bucket 9 (44040192 elements, 44040192 padded size):
801
+ module.decoder.layers.13.mlp.experts.linear_fc2.weight0
802
+ module.decoder.layers.13.mlp.experts.linear_fc1.weight3
803
+ Params for bucket 10 (58720256 elements, 58720256 padded size):
804
+ module.decoder.layers.13.mlp.experts.linear_fc1.weight2
805
+ module.decoder.layers.13.mlp.experts.linear_fc1.weight1
806
+ Params for bucket 11 (44040192 elements, 44040192 padded size):
807
+ module.decoder.layers.13.mlp.experts.linear_fc1.weight0
808
+ module.decoder.layers.12.mlp.experts.linear_fc2.weight3
809
+ Params for bucket 12 (44040192 elements, 44040192 padded size):
810
+ module.decoder.layers.12.mlp.experts.linear_fc2.weight2
811
+ module.decoder.layers.12.mlp.experts.linear_fc2.weight0
812
+ module.decoder.layers.12.mlp.experts.linear_fc2.weight1
813
+ Params for bucket 13 (58720256 elements, 58720256 padded size):
814
+ module.decoder.layers.12.mlp.experts.linear_fc1.weight3
815
+ module.decoder.layers.12.mlp.experts.linear_fc1.weight2
816
+ Params for bucket 14 (58720256 elements, 58720256 padded size):
817
+ module.decoder.layers.12.mlp.experts.linear_fc1.weight1
818
+ module.decoder.layers.12.mlp.experts.linear_fc1.weight0
819
+ Params for bucket 15 (44040192 elements, 44040192 padded size):
820
+ module.decoder.layers.11.mlp.experts.linear_fc2.weight3
821
+ module.decoder.layers.11.mlp.experts.linear_fc2.weight2
822
+ module.decoder.layers.11.mlp.experts.linear_fc2.weight1
823
+ Params for bucket 16 (44040192 elements, 44040192 padded size):
824
+ module.decoder.layers.11.mlp.experts.linear_fc2.weight0
825
+ module.decoder.layers.11.mlp.experts.linear_fc1.weight3
826
+ Params for bucket 17 (58720256 elements, 58720256 padded size):
827
+ module.decoder.layers.11.mlp.experts.linear_fc1.weight2
828
+ module.decoder.layers.11.mlp.experts.linear_fc1.weight1
829
+ Params for bucket 18 (44040192 elements, 44040192 padded size):
830
+ module.decoder.layers.11.mlp.experts.linear_fc1.weight0
831
+ module.decoder.layers.10.mlp.experts.linear_fc2.weight3
832
+ Params for bucket 19 (44040192 elements, 44040192 padded size):
833
+ module.decoder.layers.10.mlp.experts.linear_fc2.weight2
834
+ module.decoder.layers.10.mlp.experts.linear_fc2.weight0
835
+ module.decoder.layers.10.mlp.experts.linear_fc2.weight1
836
+ Params for bucket 20 (58720256 elements, 58720256 padded size):
837
+ module.decoder.layers.10.mlp.experts.linear_fc1.weight3
838
+ module.decoder.layers.10.mlp.experts.linear_fc1.weight2
839
+ Params for bucket 21 (58720256 elements, 58720256 padded size):
840
+ module.decoder.layers.10.mlp.experts.linear_fc1.weight1
841
+ module.decoder.layers.10.mlp.experts.linear_fc1.weight0
842
+ Params for bucket 22 (44040192 elements, 44040192 padded size):
843
+ module.decoder.layers.9.mlp.experts.linear_fc2.weight3
844
+ module.decoder.layers.9.mlp.experts.linear_fc2.weight1
845
+ module.decoder.layers.9.mlp.experts.linear_fc2.weight2
846
+ Params for bucket 23 (44040192 elements, 44040192 padded size):
847
+ module.decoder.layers.9.mlp.experts.linear_fc2.weight0
848
+ module.decoder.layers.9.mlp.experts.linear_fc1.weight3
849
+ Params for bucket 24 (58720256 elements, 58720256 padded size):
850
+ module.decoder.layers.9.mlp.experts.linear_fc1.weight2
851
+ module.decoder.layers.9.mlp.experts.linear_fc1.weight1
852
+ Params for bucket 25 (44040192 elements, 44040192 padded size):
853
+ module.decoder.layers.9.mlp.experts.linear_fc1.weight0
854
+ module.decoder.layers.8.mlp.experts.linear_fc2.weight3
855
+ Params for bucket 26 (44040192 elements, 44040192 padded size):
856
+ module.decoder.layers.8.mlp.experts.linear_fc2.weight2
857
+ module.decoder.layers.8.mlp.experts.linear_fc2.weight0
858
+ module.decoder.layers.8.mlp.experts.linear_fc2.weight1
859
+ Params for bucket 27 (58720256 elements, 58720256 padded size):
860
+ module.decoder.layers.8.mlp.experts.linear_fc1.weight3
861
+ module.decoder.layers.8.mlp.experts.linear_fc1.weight2
862
+ Params for bucket 28 (58720256 elements, 58720256 padded size):
863
+ module.decoder.layers.8.mlp.experts.linear_fc1.weight1
864
+ module.decoder.layers.8.mlp.experts.linear_fc1.weight0
865
+ Params for bucket 29 (44040192 elements, 44040192 padded size):
866
+ module.decoder.layers.7.mlp.experts.linear_fc2.weight3
867
+ module.decoder.layers.7.mlp.experts.linear_fc2.weight2
868
+ module.decoder.layers.7.mlp.experts.linear_fc2.weight1
869
+ Params for bucket 30 (44040192 elements, 44040192 padded size):
870
+ module.decoder.layers.7.mlp.experts.linear_fc2.weight0
871
+ module.decoder.layers.7.mlp.experts.linear_fc1.weight3
872
+ Params for bucket 31 (58720256 elements, 58720256 padded size):
873
+ module.decoder.layers.7.mlp.experts.linear_fc1.weight2
874
+ module.decoder.layers.7.mlp.experts.linear_fc1.weight1
875
+ Params for bucket 32 (44040192 elements, 44040192 padded size):
876
+ module.decoder.layers.7.mlp.experts.linear_fc1.weight0
877
+ module.decoder.layers.6.mlp.experts.linear_fc2.weight3
878
+ Params for bucket 33 (44040192 elements, 44040192 padded size):
879
+ module.decoder.layers.6.mlp.experts.linear_fc2.weight2
880
+ module.decoder.layers.6.mlp.experts.linear_fc2.weight0
881
+ module.decoder.layers.6.mlp.experts.linear_fc2.weight1
882
+ Params for bucket 34 (58720256 elements, 58720256 padded size):
883
+ module.decoder.layers.6.mlp.experts.linear_fc1.weight3
884
+ module.decoder.layers.6.mlp.experts.linear_fc1.weight2
885
+ Params for bucket 35 (58720256 elements, 58720256 padded size):
886
+ module.decoder.layers.6.mlp.experts.linear_fc1.weight1
887
+ module.decoder.layers.6.mlp.experts.linear_fc1.weight0
888
+ Params for bucket 36 (44040192 elements, 44040192 padded size):
889
+ module.decoder.layers.5.mlp.experts.linear_fc2.weight3
890
+ module.decoder.layers.5.mlp.experts.linear_fc2.weight1
891
+ module.decoder.layers.5.mlp.experts.linear_fc2.weight2
892
+ Params for bucket 37 (44040192 elements, 44040192 padded size):
893
+ module.decoder.layers.5.mlp.experts.linear_fc2.weight0
894
+ module.decoder.layers.5.mlp.experts.linear_fc1.weight3
895
+ Params for bucket 38 (58720256 elements, 58720256 padded size):
896
+ module.decoder.layers.5.mlp.experts.linear_fc1.weight2
897
+ module.decoder.layers.5.mlp.experts.linear_fc1.weight1
898
+ Params for bucket 39 (44040192 elements, 44040192 padded size):
899
+ module.decoder.layers.5.mlp.experts.linear_fc1.weight0
900
+ module.decoder.layers.4.mlp.experts.linear_fc2.weight3
901
+ Params for bucket 40 (44040192 elements, 44040192 padded size):
902
+ module.decoder.layers.4.mlp.experts.linear_fc2.weight2
903
+ module.decoder.layers.4.mlp.experts.linear_fc2.weight0
904
+ module.decoder.layers.4.mlp.experts.linear_fc2.weight1
905
+ Params for bucket 41 (58720256 elements, 58720256 padded size):
906
+ module.decoder.layers.4.mlp.experts.linear_fc1.weight3
907
+ module.decoder.layers.4.mlp.experts.linear_fc1.weight2
908
+ Params for bucket 42 (58720256 elements, 58720256 padded size):
909
+ module.decoder.layers.4.mlp.experts.linear_fc1.weight1
910
+ module.decoder.layers.4.mlp.experts.linear_fc1.weight0
911
+ Params for bucket 43 (44040192 elements, 44040192 padded size):
912
+ module.decoder.layers.3.mlp.experts.linear_fc2.weight3
913
+ module.decoder.layers.3.mlp.experts.linear_fc2.weight1
914
+ module.decoder.layers.3.mlp.experts.linear_fc2.weight2
915
+ Params for bucket 44 (44040192 elements, 44040192 padded size):
916
+ module.decoder.layers.3.mlp.experts.linear_fc2.weight0
917
+ module.decoder.layers.3.mlp.experts.linear_fc1.weight3
918
+ Params for bucket 45 (58720256 elements, 58720256 padded size):
919
+ module.decoder.layers.3.mlp.experts.linear_fc1.weight2
920
+ module.decoder.layers.3.mlp.experts.linear_fc1.weight1
921
+ Params for bucket 46 (44040192 elements, 44040192 padded size):
922
+ module.decoder.layers.3.mlp.experts.linear_fc1.weight0
923
+ module.decoder.layers.2.mlp.experts.linear_fc2.weight3
924
+ Params for bucket 47 (44040192 elements, 44040192 padded size):
925
+ module.decoder.layers.2.mlp.experts.linear_fc2.weight2
926
+ module.decoder.layers.2.mlp.experts.linear_fc2.weight1
927
+ module.decoder.layers.2.mlp.experts.linear_fc2.weight0
928
+ Params for bucket 48 (58720256 elements, 58720256 padded size):
929
+ module.decoder.layers.2.mlp.experts.linear_fc1.weight3
930
+ module.decoder.layers.2.mlp.experts.linear_fc1.weight2
931
+ Params for bucket 49 (58720256 elements, 58720256 padded size):
932
+ module.decoder.layers.2.mlp.experts.linear_fc1.weight1
933
+ module.decoder.layers.2.mlp.experts.linear_fc1.weight0
934
+ Params for bucket 50 (44040192 elements, 44040192 padded size):
935
+ module.decoder.layers.1.mlp.experts.linear_fc2.weight3
936
+ module.decoder.layers.1.mlp.experts.linear_fc2.weight2
937
+ module.decoder.layers.1.mlp.experts.linear_fc2.weight1
938
+ Params for bucket 51 (44040192 elements, 44040192 padded size):
939
+ module.decoder.layers.1.mlp.experts.linear_fc2.weight0
940
+ module.decoder.layers.1.mlp.experts.linear_fc1.weight3
941
+ Params for bucket 52 (58720256 elements, 58720256 padded size):
942
+ module.decoder.layers.1.mlp.experts.linear_fc1.weight2
943
+ module.decoder.layers.1.mlp.experts.linear_fc1.weight1
944
+ Params for bucket 53 (44040192 elements, 44040192 padded size):
945
+ module.decoder.layers.1.mlp.experts.linear_fc1.weight0
946
+ module.decoder.layers.0.mlp.experts.linear_fc2.weight3
947
+ Params for bucket 54 (44040192 elements, 44040192 padded size):
948
+ module.decoder.layers.0.mlp.experts.linear_fc2.weight2
949
+ module.decoder.layers.0.mlp.experts.linear_fc2.weight1
950
+ module.decoder.layers.0.mlp.experts.linear_fc2.weight0
951
+ Params for bucket 55 (58720256 elements, 58720256 padded size):
952
+ module.decoder.layers.0.mlp.experts.linear_fc1.weight3
953
+ module.decoder.layers.0.mlp.experts.linear_fc1.weight2
954
+ Params for bucket 56 (58720256 elements, 58720256 padded size):
955
+ module.decoder.layers.0.mlp.experts.linear_fc1.weight1
956
+ module.decoder.layers.0.mlp.experts.linear_fc1.weight0
957
+ 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 0x74a909648ce0>, config_logger_dir='')
958
+ INFO:megatron.core.optimizer_param_scheduler:> learning rate decay style: cosine
959
+ [after model, optimizer, and learning rate scheduler are built] datetime: 2026-08-30 05:12:26
960
+ > building train, validation, and test datasets ...
961
+ > datasets target sizes (minimum size):
962
+ train: 192
963
+ validation: 0
964
+ test: 0
965
+ INFO:megatron.core.datasets.blended_megatron_dataset_config:Let split_matrix = [(0, 1.0), None, None]
966
+ > building train, validation, and test datasets for GPT ...
967
+ INFO:megatron.core.datasets.blended_megatron_dataset_builder:Building GPTDataset splits with sizes=(192, 0, 0) and config=GPTDatasetConfig(random_seed=1234, sequence_length=4096, blend=(['/workspace/data/wikitext/wikitext103_gpt2_text_document'], None), blend_per_split=None, split='100,0,0', split_matrix=[(0, 1.0), None, None], num_dataset_builder_threads=1, path_to_cache=None, mmap_bin_files=True, mock=False, tokenizer=<megatron.training.tokenizer.tokenizer._GPT2BPETokenizer object at 0x74a9098f4260>, 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)
968
+ INFO:megatron.core.datasets.indexed_dataset:Load the _IndexReader from /workspace/data/wikitext/wikitext103_gpt2_text_document.idx
969
+ INFO:megatron.core.datasets.indexed_dataset: Extract the sequence lengths
970
+ INFO:megatron.core.datasets.indexed_dataset: Extract the sequence pointers
971
+ INFO:megatron.core.datasets.indexed_dataset: Extract the document indices
972
+ INFO:megatron.core.datasets.indexed_dataset:> total number of sequences: 1165029
973
+ INFO:megatron.core.datasets.indexed_dataset:> total number of documents: 1801350
974
+ INFO:megatron.core.datasets.gpt_dataset:Load the GPTDataset train indices
975
+ INFO:megatron.core.datasets.gpt_dataset: Load the document index from 44ed1ee5bcb9e5b030387b8d27a6c406-GPTDataset-train-document_index.npy
976
+ INFO:megatron.core.datasets.gpt_dataset: Load the sample index from 44ed1ee5bcb9e5b030387b8d27a6c406-GPTDataset-train-sample_index.npy
977
+ INFO:megatron.core.datasets.gpt_dataset: Load the shuffle index from 44ed1ee5bcb9e5b030387b8d27a6c406-GPTDataset-train-shuffle_index.npy
978
+ INFO:megatron.core.datasets.gpt_dataset:> total number of samples: 29073
979
+ > finished creating GPT datasets ...
980
+ [after dataloaders are built] datetime: 2026-08-30 05:12:26
981
+ done with setup ...
982
+ (min, max) time across ranks (ms):
983
+ model-and-optimizer-setup ......................: (3466.83, 3478.95)
984
+ train/valid/test-data-iterators-setup ..........: (75.31, 91.23)
985
+ training ...
986
+ Overwriting rerun_state_machine.current_iteration from -1 to 0...
987
+ [before the start of training step] datetime: 2026-08-30 05:12:26
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:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
1003
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
1004
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1005
+ warnings.warn("The num_dispatched_tokens 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:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1015
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1016
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
1017
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
1018
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
1019
+ warnings.warn("The use_host_meta 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:482: 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:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1035
+ warnings.warn("The num_dispatched_tokens 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:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1051
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1052
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
1053
+ warnings.warn("The use_host_meta 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:334: 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:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
1067
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
1068
+ WARNING:megatron.core.rerun_state_machine:Result validation enabled
1069
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1070
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1071
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1072
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1073
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1074
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1075
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1076
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1077
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1078
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1079
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1080
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1081
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1082
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1083
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1084
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1085
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1086
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1087
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1088
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1089
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1090
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1091
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1092
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1093
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1094
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1095
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1096
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1097
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1098
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1099
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
1100
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
1101
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
1102
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
1103
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
1104
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
1105
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
1106
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
1107
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
1108
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
1109
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
1110
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
1111
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
1112
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
1113
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
1114
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
1115
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
1116
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/capacity_attempts.tsv ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ model backend purpose capacity status peak_mib zero_drop drop_rate_pct required_capacity run_dir
2
+ G1 ultraep measure 1.25 OK 108731 0 0.6209174791971842 1.724609 /workspace/tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G1__ultraep__matched__seq4096__gbs16__cap1p25/rep1
3
+ G1 ultraep validate 1.8125 OK 113767 0 0.032329559326171875 2.309082 /workspace/tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G1__ultraep__matched__seq4096__gbs16__cap1p8125/rep1
4
+ G2 ultraep measure 1.25 OK 110343 1 0.0 1.010101 /workspace/tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G2__ultraep__matched__seq4096__gbs16__cap1p25/rep1
5
+ G2 ultraep validate 1.0625 OK 107987 1 0.0 1.010101 /workspace/tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G2__ultraep__matched__seq4096__gbs16__cap1p0625/rep1
tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/capacity_selection.tsv ADDED
@@ -0,0 +1,2 @@
 
 
 
1
+ model backend status selected_capacity required_capacity drop_rate_pct peak_mib run_dir
2
+ G1 ultraep ROUNDED_CAPACITY_VALIDATION_DROPPED
tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/results.tsv ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ config backend policy slots seq gbs cap recompute rep status iter_ms_mean iter_ms_min peak_mib run_dir
2
+ G1 ultraep matched 1 4096 16 1.25 none 1 OK 893.49 886.90 108731 /workspace/tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G1__ultraep__matched__seq4096__gbs16__cap1p25/rep1
3
+ G1 ultraep matched 1 4096 16 1.8125 none 1 OK 914.84 903.60 113767 /workspace/tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G1__ultraep__matched__seq4096__gbs16__cap1p8125/rep1
4
+ G2 ultraep matched 4 4096 16 1.25 none 1 OK 558.91 532.70 110343 /workspace/tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G2__ultraep__matched__seq4096__gbs16__cap1p25/rep1
5
+ G2 ultraep matched 4 4096 16 1.0625 none 1 OK 551.94 527.00 107987 /workspace/tiera-rerun-fixed-moonep/ultraep/tierbprobe/tierbprobe_gbs16/G2__ultraep__matched__seq4096__gbs16__cap1p0625/rep1