AbhishekGhosh1998 commited on
Commit
7ea7007
·
verified ·
1 Parent(s): e4f9ea5

granularity results tiera-rerun-fixed-moonep @ 2026-08-30T05:59:20Z from sp-aiinfra-interactive-aghosh-7jtjl-8xb200-moe-cg-0

Browse files
Files changed (19) hide show
  1. .gitattributes +1 -0
  2. tiera-rerun-fixed-moonep/echo/a1/a1_oneslot_gbs16/G1__echo__one__seq4096__gbs16__cap2p40625/nsys/G1__echo__one__seq4096__gbs16__cap2p40625.nsys-rep +3 -0
  3. tiera-rerun-fixed-moonep/echo/a1/a1_oneslot_gbs16/G1__echo__one__seq4096__gbs16__cap2p40625/nsys/run.env +70 -0
  4. tiera-rerun-fixed-moonep/echo/a1/a1_oneslot_gbs16/G1__echo__one__seq4096__gbs16__cap2p40625/nsys/train.log +984 -0
  5. tiera-rerun-fixed-moonep/echo/a1/a1_oneslot_gbs16/G1__echo__one__seq4096__gbs16__cap2p40625/rep1/mem_samples.csv +232 -0
  6. tiera-rerun-fixed-moonep/echo/a1/a1_oneslot_gbs16/G1__echo__one__seq4096__gbs16__cap2p40625/rep1/run.env +70 -0
  7. tiera-rerun-fixed-moonep/echo/a1/a1_oneslot_gbs16/G1__echo__one__seq4096__gbs16__cap2p40625/rep1/train.log +1023 -0
  8. tiera-rerun-fixed-moonep/echo/a1/a1_oneslot_gbs16/G1__echo__one__seq4096__gbs16__cap2p40625/rep2/mem_samples.csv +232 -0
  9. tiera-rerun-fixed-moonep/echo/a1/a1_oneslot_gbs16/G1__echo__one__seq4096__gbs16__cap2p40625/rep2/run.env +70 -0
  10. tiera-rerun-fixed-moonep/echo/a1/a1_oneslot_gbs16/G1__echo__one__seq4096__gbs16__cap2p40625/rep2/train.log +1023 -0
  11. tiera-rerun-fixed-moonep/echo/a1/a1_oneslot_gbs16/G1__echo__one__seq4096__gbs16__cap2p40625/rep3/mem_samples.csv +240 -0
  12. tiera-rerun-fixed-moonep/echo/a1/a1_oneslot_gbs16/G1__echo__one__seq4096__gbs16__cap2p40625/rep3/run.env +70 -0
  13. tiera-rerun-fixed-moonep/echo/a1/a1_oneslot_gbs16/G1__echo__one__seq4096__gbs16__cap2p40625/rep3/train.log +1023 -0
  14. tiera-rerun-fixed-moonep/echo/a1/a1_oneslot_gbs16/results.tsv +4 -0
  15. tiera-rerun-fixed-moonep/echo/a1/manifest.txt +19 -0
  16. tiera-rerun-fixed-moonep/echo/a1/pip_freeze.txt +317 -0
  17. tiera-rerun-fixed-moonep/echo/a1/provenance.txt +61 -0
  18. tiera-rerun-fixed-moonep/hf_watcher.log +3 -0
  19. tiera-rerun-fixed-moonep/study.log +16 -0
.gitattributes CHANGED
@@ -68,3 +68,4 @@ tierb-fixed-moonep-full/moonep/tierbtrace/tierbtrace_gbs8/DSV2_LITE_L24__moonep_
68
  tierb-fixed-moonep-full/moonep/tierbtrace/tierbtrace_gbs8/QWEN3_30B__moonep__matched__seq4096__gbs8__capnone__async/nsys/QWEN3_30B__moonep__matched__seq4096__gbs8__capnone__async.nsys-rep filter=lfs diff=lfs merge=lfs -text
69
  tierb-fixed-moonep-full/moonep/tierbtrace/tierbtrace_gbs8/QWEN2_57B__moonep__matched__seq4096__gbs8__capnone__async/nsys/QWEN2_57B__moonep__matched__seq4096__gbs8__capnone__async.nsys-rep filter=lfs diff=lfs merge=lfs -text
70
  tiera-rerun-fixed-moonep/echo/tierbprobe/tierbprobe_gbs16/G4__echo__matched__seq4096__gbs16__cap1p068359375__forcebal/rep1/train.log filter=lfs diff=lfs merge=lfs -text
 
 
68
  tierb-fixed-moonep-full/moonep/tierbtrace/tierbtrace_gbs8/QWEN3_30B__moonep__matched__seq4096__gbs8__capnone__async/nsys/QWEN3_30B__moonep__matched__seq4096__gbs8__capnone__async.nsys-rep filter=lfs diff=lfs merge=lfs -text
69
  tierb-fixed-moonep-full/moonep/tierbtrace/tierbtrace_gbs8/QWEN2_57B__moonep__matched__seq4096__gbs8__capnone__async/nsys/QWEN2_57B__moonep__matched__seq4096__gbs8__capnone__async.nsys-rep filter=lfs diff=lfs merge=lfs -text
70
  tiera-rerun-fixed-moonep/echo/tierbprobe/tierbprobe_gbs16/G4__echo__matched__seq4096__gbs16__cap1p068359375__forcebal/rep1/train.log filter=lfs diff=lfs merge=lfs -text
71
+ tiera-rerun-fixed-moonep/echo/a1/a1_oneslot_gbs16/G1__echo__one__seq4096__gbs16__cap2p40625/nsys/G1__echo__one__seq4096__gbs16__cap2p40625.nsys-rep filter=lfs diff=lfs merge=lfs -text
tiera-rerun-fixed-moonep/echo/a1/a1_oneslot_gbs16/G1__echo__one__seq4096__gbs16__cap2p40625/nsys/G1__echo__one__seq4096__gbs16__cap2p40625.nsys-rep ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:689469b0717042c2d60846a25f6ad6c548c21bebb933f2f7ab36ad5c234917d8
3
+ size 14366810
tiera-rerun-fixed-moonep/echo/a1/a1_oneslot_gbs16/G1__echo__one__seq4096__gbs16__cap2p40625/nsys/run.env ADDED
@@ -0,0 +1,70 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ backend=echo
2
+ config=G1
3
+ family=tierA
4
+ repo=/workspace/Megatron-MoE-Echo
5
+ branch=bracing-regions-through-kernels
6
+ commit=9a236793
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=one
42
+ slots_per_rank=1
43
+ capacity_mode=static
44
+ capacity=2.40625
45
+ cuda_graph=1
46
+ probe_mode=0
47
+ async_grad_reduce=0
48
+ branch_role=timing
49
+ memory_dump_mode=none
50
+ memory_snapshot_path=
51
+ cuda_graph_warmup_steps=3
52
+ recompute=none
53
+ force_balance=0
54
+ train_iters=10
55
+ ep_size=8
56
+ echo_dispatch_overlap=1
57
+ echo_dispatch_recompute=0
58
+ ultraep_quota_min_tokens=
59
+ ultraep_weight_sync_join=
60
+ profile_enable=1
61
+ kernel_trace=1
62
+ fp8_mode=mxfp8
63
+ fp8_recipe=mxfp8
64
+ training_seed=1234
65
+ learning_rate=1e-4
66
+ min_learning_rate=1.0e-5
67
+ lr_warmup_iters=5
68
+ state_dump_enable=0
69
+ state_dump_dir=
70
+ moonep_assert_no_pending_reduce=0
tiera-rerun-fixed-moonep/echo/a1/a1_oneslot_gbs16/G1__echo__one__seq4096__gbs16__cap2p40625/nsys/train.log ADDED
@@ -0,0 +1,984 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ==============================================================
2
+ G1 / echo (slots/rank=1 policy=one)
3
+ E=8 K=2 H'=14336 m=1
4
+ L=16 seq=4096 gbs=16 (2 microbatches)
5
+ cap_mode=static cap=2.40625 cg=1 recompute=none
6
+ memory_dump=none snapshots=/workspace/tiera-rerun-fixed-moonep/echo/a1/a1_oneslot_gbs16/G1__echo__one__seq4096__gbs16__cap2p40625/nsys/memory_snapshots/snapshot.pickle
7
+ repo=/workspace/Megatron-MoE-Echo @ 9a236793
8
+ ==============================================================
9
+ W0830 05:53:26.265000 555662 torch/distributed/run.py:766]
10
+ W0830 05:53:26.265000 555662 torch/distributed/run.py:766] *****************************************
11
+ W0830 05:53:26.265000 555662 torch/distributed/run.py:766] Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed.
12
+ W0830 05:53:26.265000 555662 torch/distributed/run.py:766] *****************************************
13
+ using world size: 8, data-parallel size: 8, context-parallel size: 1, hierarchical context-parallel sizes: None, tensor-model-parallel size: 1, pipeline-model-parallel size: 1
14
+ WARNING: overriding default arguments for tokenizer_type:GPT2BPETokenizer with tokenizer_type:GPT2BPETokenizer
15
+ Number of virtual stages per pipeline stage: None
16
+ accumulate and all-reduce gradients in fp32 for bfloat16 data type.
17
+ using torch.bfloat16 for parameters ...
18
+ /workspace/Megatron-MoE-Echo/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
19
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
20
+ ------------------------ arguments ------------------------
21
+ account_for_embedding_in_pipeline_split ......... False
22
+ account_for_loss_in_pipeline_split .............. False
23
+ accumulate_allreduce_grads_in_fp32 .............. True
24
+ adam_beta1 ...................................... 0.9
25
+ adam_beta2 ...................................... 0.999
26
+ adam_eps ........................................ 1e-08
27
+ add_bias_linear ................................. False
28
+ add_position_embedding .......................... False
29
+ add_qkv_bias .................................... False
30
+ adlr_autoresume ................................. False
31
+ adlr_autoresume_interval ........................ 1000
32
+ align_grad_reduce ............................... True
33
+ align_param_gather .............................. False
34
+ app_tag_run_name ................................ None
35
+ app_tag_run_version ............................. 0.0.0
36
+ apply_layernorm_1p .............................. False
37
+ apply_query_key_layer_scaling ................... False
38
+ apply_residual_connection_post_layernorm ........ False
39
+ apply_rope_fusion ............................... True
40
+ async_save ...................................... None
41
+ async_tensor_model_parallel_allreduce ........... True
42
+ attention_backend ............................... AttnBackend.auto
43
+ attention_dropout ............................... 0.0
44
+ attention_softmax_in_fp32 ....................... False
45
+ auto_detect_ckpt_format ......................... False
46
+ barrier_with_L1_time ............................ True
47
+ bert_binary_head ................................ True
48
+ bert_embedder_type .............................. megatron
49
+ bert_load ....................................... None
50
+ bf16 ............................................ True
51
+ bias_dropout_fusion ............................. True
52
+ bias_gelu_fusion ................................ False
53
+ bias_swiglu_fusion .............................. True
54
+ biencoder_projection_dim ........................ 0
55
+ biencoder_shared_query_context_model ............ False
56
+ block_data_path ................................. None
57
+ calc_ft_timeouts ................................ False
58
+ calculate_per_token_loss ........................ False
59
+ check_for_large_grads ........................... False
60
+ check_for_nan_in_loss_and_grad .................. False
61
+ check_for_spiky_loss ............................ False
62
+ check_weight_hash_across_dp_replicas_interval ... None
63
+ ckpt_assume_constant_structure .................. False
64
+ ckpt_convert_format ............................. None
65
+ ckpt_convert_save ............................... None
66
+ ckpt_convert_update_legacy_dist_opt_format ...... False
67
+ ckpt_format ..................................... torch_dist
68
+ ckpt_fully_parallel_load ........................ False
69
+ ckpt_fully_parallel_save ........................ True
70
+ ckpt_fully_parallel_save_deprecated ............. False
71
+ ckpt_step ....................................... None
72
+ classes_fraction ................................ 1.0
73
+ clip_grad ....................................... 1.0
74
+ clone_scatter_output_in_embedding ............... True
75
+ config_logger_dir ...............................
76
+ consumed_train_samples .......................... 0
77
+ consumed_valid_samples .......................... 0
78
+ context_parallel_size ........................... 1
79
+ cp_comm_type .................................... ['p2p']
80
+ create_attention_mask_in_dataloader ............. True
81
+ cross_entropy_fusion_impl ....................... native
82
+ cross_entropy_loss_fusion ....................... False
83
+ cuda_graph_scope ................................ full_iteration
84
+ cuda_graph_warmup_steps ......................... 3
85
+ data_args_path .................................. None
86
+ data_cache_path ................................. None
87
+ data_parallel_random_init ....................... False
88
+ data_parallel_sharding_strategy ................. no_shard
89
+ data_parallel_size .............................. 8
90
+ data_path ....................................... ['/workspace/data/wikitext/wikitext103_gpt2_text_document']
91
+ data_per_class_fraction ......................... 1.0
92
+ data_sharding ................................... True
93
+ dataloader_type ................................. single
94
+ ddp_average_in_collective ....................... False
95
+ ddp_bucket_size ................................. None
96
+ ddp_num_buckets ................................. None
97
+ ddp_pad_buckets_for_high_nccl_busbw ............. False
98
+ decoder_first_pipeline_num_layers ............... None
99
+ decoder_last_pipeline_num_layers ................ None
100
+ decoder_num_layers .............................. None
101
+ decoder_seq_length .............................. None
102
+ decoupled_lr .................................... None
103
+ decoupled_min_lr ................................ None
104
+ decrease_batch_size_if_needed ................... False
105
+ defer_embedding_wgrad_compute ................... False
106
+ delay_wgrad_compute ............................. False
107
+ deprecated_use_mcore_models ..................... True
108
+ deterministic_mode .............................. False
109
+ dino_bottleneck_size ............................ 256
110
+ dino_freeze_last_layer .......................... 1
111
+ dino_head_hidden_size ........................... 2048
112
+ dino_local_crops_number ......................... 10
113
+ dino_local_img_size ............................. 96
114
+ dino_norm_last_layer ............................ False
115
+ dino_teacher_temp ............................... 0.07
116
+ dino_warmup_teacher_temp ........................ 0.04
117
+ dino_warmup_teacher_temp_epochs ................. 30
118
+ disable_bf16_reduced_precision_matmul ........... False
119
+ disable_mamba_mem_eff_path ...................... False
120
+ disable_straggler_on_startup .................... False
121
+ dist_ckpt_format_deprecated ..................... None
122
+ dist_ckpt_strictness ............................ assume_ok_unexpected
123
+ distribute_saved_activations .................... False
124
+ distributed_backend ............................. nccl
125
+ distributed_timeout_minutes ..................... 10
126
+ embedding_init_method_std ....................... None
127
+ embedding_path .................................. None
128
+ empty_unused_memory_level ....................... 0
129
+ enable_cuda_graph ............................... True
130
+ enable_experimental ............................. False
131
+ enable_ft_package ............................... False
132
+ enable_gloo_process_groups ...................... True
133
+ enable_msc ...................................... True
134
+ enable_one_logger ............................... True
135
+ encoder_num_layers .............................. 16
136
+ encoder_seq_length .............................. 4096
137
+ end_weight_decay ................................ 0.1
138
+ eod_mask_loss ................................... False
139
+ error_injection_rate ............................ 0
140
+ error_injection_type ............................ transient_error
141
+ eval_interval ................................... 1000000
142
+ eval_iters ...................................... 0
143
+ evidence_data_path .............................. None
144
+ exit_duration_in_mins ........................... None
145
+ exit_interval ................................... None
146
+ exit_on_missing_checkpoint ...................... False
147
+ exit_signal_handler ............................. False
148
+ exp_avg_dtype ................................... torch.float32
149
+ exp_avg_sq_dtype ................................ torch.float32
150
+ expert_model_parallel_size ...................... 8
151
+ expert_tensor_parallel_size ..................... 1
152
+ export_force_local_attention .................... False
153
+ export_kd_cfg ................................... None
154
+ export_kd_teacher_ckpt_format ................... None
155
+ export_kd_teacher_load .......................... None
156
+ export_kv_cache_quant ........................... False
157
+ export_legacy_megatron .......................... False
158
+ export_model_type ............................... GPTModel
159
+ export_moe_apply_probs_on_input ................. False
160
+ export_qk_l2_norm ............................... False
161
+ export_quant_cfg ................................ None
162
+ export_real_quant_cfg ........................... None
163
+ export_te_mcore_model ........................... False
164
+ external_cuda_graph ............................. False
165
+ ffn_hidden_size ................................. 14336
166
+ finetune ........................................ False
167
+ finetune_data_split ............................. train
168
+ finetune_hf_dataset ............................. None
169
+ first_last_layers_bf16 .......................... False
170
+ flash_decode .................................... False
171
+ fp16 ............................................ False
172
+ fp16_lm_cross_entropy ........................... False
173
+ fp32_residual_connection ........................ False
174
+ fp8 ............................................. e4m3
175
+ fp8_amax_compute_algo ........................... most_recent
176
+ fp8_amax_history_len ............................ 1
177
+ fp8_interval .................................... 1
178
+ fp8_margin ...................................... 0
179
+ fp8_param_gather ................................ False
180
+ fp8_recipe ...................................... mxfp8
181
+ fp8_wgrad ....................................... True
182
+ fsdp_double_buffer .............................. False
183
+ global_batch_size ............................... 16
184
+ grad_reduce_in_bf16 ............................. False
185
+ gradient_accumulation_fusion .................... True
186
+ gradient_reduce_div_fusion ...................... True
187
+ group_query_attention ........................... True
188
+ head_lr_mult .................................... 1.0
189
+ heterogeneous_layers_config_encoded_json ........ None
190
+ heterogeneous_layers_config_path ................ None
191
+ hidden_dropout .................................. 0.0
192
+ hidden_size ..................................... 4096
193
+ hierarchical_context_parallel_sizes ............. None
194
+ high_priority_stream_groups ..................... []
195
+ hybrid_attention_ratio .......................... 0.0
196
+ hybrid_mlp_ratio ................................ 0.0
197
+ hybrid_override_pattern ......................... None
198
+ hysteresis ...................................... 2
199
+ ict_head_size ................................... None
200
+ ict_load ........................................ None
201
+ img_h ........................................... 224
202
+ img_w ........................................... 224
203
+ indexer_batch_size .............................. 128
204
+ indexer_log_interval ............................ 1000
205
+ inference_batch_times_seqlen_threshold .......... -1
206
+ inference_dynamic_batching ...................... False
207
+ inference_dynamic_batching_buffer_guaranteed_fraction 0.2
208
+ inference_dynamic_batching_buffer_overflow_factor None
209
+ inference_dynamic_batching_buffer_size_gb ....... 40.0
210
+ inference_dynamic_batching_chunk_size ........... 256
211
+ inference_dynamic_batching_max_requests_override None
212
+ inference_dynamic_batching_max_tokens_override .. None
213
+ inference_dynamic_batching_num_cuda_graphs ...... 16
214
+ inference_max_batch_size ........................ 8
215
+ inference_max_seq_length ........................ 2560
216
+ inference_rng_tracker ........................... False
217
+ init_method_std ................................. 0.01
218
+ init_method_xavier_uniform ...................... False
219
+ init_model_with_meta_device ..................... False
220
+ initial_loss_scale .............................. 4294967296
221
+ inprocess_active_world_size ..................... 8
222
+ inprocess_barrier_timeout ....................... 120
223
+ inprocess_completion_timeout .................... 120
224
+ inprocess_empty_cuda_cache ...................... False
225
+ inprocess_granularity ........................... node
226
+ inprocess_hard_timeout .......................... 90
227
+ inprocess_heartbeat_interval .................... 30
228
+ inprocess_heartbeat_timeout ..................... 60
229
+ inprocess_last_call_wait ........................ 1
230
+ inprocess_max_iterations ........................ None
231
+ inprocess_monitor_process_interval .............. 1.0
232
+ inprocess_monitor_thread_interval ............... 1.0
233
+ inprocess_progress_watchdog_interval ............ 1.0
234
+ inprocess_restart ............................... False
235
+ inprocess_soft_timeout .......................... 60
236
+ inprocess_termination_grace_time ................ 1
237
+ is_hybrid_model ................................. False
238
+ iter_per_epoch .................................. 1250
239
+ iterations_to_skip .............................. []
240
+ keep_fp8_transpose_cache_when_using_custom_fsdp . False
241
+ kitchen_config_file ............................. None
242
+ kitchen_recipe_number ........................... None
243
+ kv_channels ..................................... 128
244
+ kv_lora_rank .................................... 32
245
+ lazy_mpu_init ................................... None
246
+ load ............................................ None
247
+ load_main_params_from_ckpt ...................... None
248
+ load_model_opt_format ........................... False
249
+ local_rank ...................................... 0
250
+ log_energy ...................................... False
251
+ log_interval .................................... 1
252
+ log_loss_scale_to_tensorboard ................... True
253
+ log_memory_to_tensorboard ....................... False
254
+ log_num_zeros_in_grad ........................... False
255
+ log_params_norm ................................. False
256
+ log_progress .................................... False
257
+ log_straggler ................................... False
258
+ log_throughput .................................. True
259
+ log_timers_to_tensorboard ....................... False
260
+ log_validation_ppl_to_tensorboard ............... False
261
+ log_world_size_to_tensorboard ................... False
262
+ logging_level ................................... None
263
+ loss_scale ...................................... None
264
+ loss_scale_window ............................... 1000
265
+ lr .............................................. 0.0001
266
+ lr_decay_iters .................................. 8
267
+ lr_decay_samples ................................ None
268
+ lr_decay_style .................................. cosine
269
+ lr_warmup_fraction .............................. None
270
+ lr_warmup_init .................................. 0.0
271
+ lr_warmup_iters ................................. 5
272
+ lr_warmup_samples ............................... 0
273
+ lr_wsd_decay_iters .............................. None
274
+ lr_wsd_decay_samples ............................ None
275
+ lr_wsd_decay_style .............................. exponential
276
+ main_grads_dtype ................................ torch.float32
277
+ main_params_dtype ............................... torch.float32
278
+ make_vocab_size_divisible_by .................... 128
279
+ mamba_head_dim .................................. 64
280
+ mamba_num_groups ................................ 8
281
+ mamba_num_heads ................................. None
282
+ mamba_state_dim ................................. 128
283
+ manual_gc ....................................... False
284
+ manual_gc_eval .................................. True
285
+ manual_gc_interval .............................. 0
286
+ mask_factor ..................................... 1.0
287
+ mask_prob ....................................... 0.15
288
+ mask_type ....................................... random
289
+ masked_softmax_fusion ........................... False
290
+ max_position_embeddings ......................... 32768
291
+ max_tokens_to_oom ............................... 12000
292
+ memory_snapshot_path ............................ snapshot.pickle
293
+ merge_file ...................................... /workspace/data/wikitext/gpt2-merges.txt
294
+ micro_batch_size ................................ 1
295
+ microbatch_group_size_per_vp_stage .............. None
296
+ mid_level_dataset_surplus ....................... 0.005
297
+ min_loss_scale .................................. 1.0
298
+ min_lr .......................................... 1e-05
299
+ mlp_chunks_for_prefill .......................... 1
300
+ mmap_bin_files .................................. True
301
+ mock_data ....................................... False
302
+ moe_apply_probs_on_input ........................ False
303
+ moe_aux_loss_coeff .............................. 0.01
304
+ moe_deepep_num_sms .............................. 20
305
+ moe_echo_enable_random_offloading ............... False
306
+ moe_echo_expert_dispatch_overlap ................ True
307
+ moe_echo_expert_dispatcher_type ................. hybridep
308
+ moe_echo_recompute_expert_dispatch .............. False
309
+ moe_enable_deepep ............................... False
310
+ moe_enable_echo ................................. True
311
+ moe_enable_hybridep ............................. True
312
+ moe_expert_capacity_factor ...................... None
313
+ moe_extended_tp ................................. False
314
+ moe_ffn_hidden_size ............................. 14336
315
+ moe_grouped_gemm ................................ True
316
+ moe_hybridep_num_sms ............................ 16
317
+ moe_input_jitter_eps ............................ None
318
+ moe_layer_freq .................................. 1
319
+ moe_layer_recompute ............................. False
320
+ moe_num_echo_experts ............................ 8
321
+ moe_pad_expert_input_to_capacity ................ False
322
+ moe_per_layer_logging ........................... False
323
+ moe_permute_fusion .............................. False
324
+ moe_received_token_capacity ..................... 2.40625
325
+ moe_router_bias_update_rate ..................... 0.001
326
+ moe_router_dtype ................................ fp32
327
+ moe_router_enable_expert_bias ................... False
328
+ moe_router_force_load_balancing ................. False
329
+ moe_router_group_topk ........................... None
330
+ moe_router_load_balancing_type .................. aux_loss
331
+ moe_router_num_groups ........................... None
332
+ moe_router_padding_for_fp8 ...................... False
333
+ moe_router_pre_softmax .......................... False
334
+ moe_router_score_function ....................... softmax
335
+ moe_router_topk ................................. 2
336
+ moe_router_topk_scaling_factor .................. None
337
+ moe_shared_expert_intermediate_size ............. None
338
+ moe_shared_expert_overlap ....................... False
339
+ moe_token_dispatcher_type ....................... flex
340
+ moe_token_drop_policy ........................... probs
341
+ moe_upcycling_granularity ....................... 1
342
+ moe_use_device_initiated_grouped_gemm ........... True
343
+ moe_use_legacy_grouped_gemm ..................... False
344
+ moe_use_upcycling ............................... False
345
+ moe_z_loss_coeff ................................ None
346
+ mrope_section ................................... None
347
+ mscale .......................................... 1.0
348
+ mscale_all_dim .................................. 0.0
349
+ mtp_loss_scaling_factor ......................... 0.1
350
+ mtp_num_layers .................................. None
351
+ multi_latent_attention .......................... False
352
+ nccl_all_reduce_for_prefill ..................... False
353
+ nccl_communicator_config_path ................... None
354
+ nccl_ub ......................................... False
355
+ no_load_optim ................................... None
356
+ no_load_rng ..................................... None
357
+ no_persist_layer_norm ........................... False
358
+ no_rope_freq .................................... None
359
+ no_save_optim ................................... None
360
+ no_save_rng ..................................... None
361
+ non_persistent_ckpt_type ........................ None
362
+ non_persistent_global_ckpt_dir .................. None
363
+ non_persistent_local_ckpt_algo .................. fully_parallel
364
+ non_persistent_local_ckpt_dir ................... None
365
+ non_persistent_save_interval .................... None
366
+ norm_epsilon .................................... 1e-05
367
+ normalization ................................... RMSNorm
368
+ num_attention_heads ............................. 32
369
+ num_channels .................................... 3
370
+ num_classes ..................................... 1000
371
+ num_dataset_builder_threads ..................... 1
372
+ num_distributed_optimizer_instances ............. 1
373
+ num_experts ..................................... 8
374
+ num_layers ...................................... 16
375
+ num_layers_at_end_in_bf16 ....................... 1
376
+ num_layers_at_start_in_bf16 ..................... 1
377
+ num_layers_per_virtual_pipeline_stage ........... None
378
+ num_query_groups ................................ 8
379
+ num_virtual_stages_per_pipeline_rank ............ None
380
+ num_workers ..................................... 2
381
+ object_storage_cache_path ....................... None
382
+ one_logger_async ................................ False
383
+ one_logger_project .............................. megatron-lm
384
+ one_logger_run_name ............................. None
385
+ onnx_safe ....................................... None
386
+ openai_gelu ..................................... False
387
+ optimizer ....................................... adam
388
+ optimizer_cpu_offload ........................... False
389
+ optimizer_offload_fraction ...................... 1.0
390
+ output_bert_embeddings .......................... False
391
+ overlap_cpu_optimizer_d2h_h2d ................... False
392
+ overlap_grad_reduce ............................. True
393
+ overlap_moe_expert_parallel_comm ................ False
394
+ overlap_p2p_comm ................................ False
395
+ overlap_p2p_comm_warmup_flush ................... False
396
+ overlap_param_gather ............................ True
397
+ overlap_param_gather_with_optimizer_step ........ False
398
+ override_opt_param_scheduler .................... False
399
+ params_dtype .................................... torch.bfloat16
400
+ patch_dim ....................................... 16
401
+ per_split_data_args_path ........................ None
402
+ perform_initialization .......................... True
403
+ pin_cpu_grads ................................... True
404
+ pin_cpu_params .................................. True
405
+ pipeline_model_parallel_comm_backend ............ None
406
+ pipeline_model_parallel_layout .................. None
407
+ pipeline_model_parallel_size .................... 1
408
+ position_embedding_type ......................... rope
409
+ pretrained_checkpoint ........................... None
410
+ profile ......................................... True
411
+ profile_ranks ................................... [0, 1, 2, 3, 4, 5, 6, 7]
412
+ profile_step_end ................................ 8
413
+ profile_step_start .............................. 6
414
+ q_lora_rank ..................................... None
415
+ qk_head_dim ..................................... 128
416
+ qk_l2_norm ...................................... False
417
+ qk_layernorm .................................... False
418
+ qk_pos_emb_head_dim ............................. 64
419
+ query_in_block_prob ............................. 0.1
420
+ rampup_batch_size ............................... None
421
+ rank ............................................ 0
422
+ recompute_granularity ........................... None
423
+ recompute_method ................................ None
424
+ recompute_modules ............................... None
425
+ recompute_num_layers ............................ None
426
+ record_memory_history ........................... False
427
+ relative_attention_max_distance ................. 128
428
+ relative_attention_num_buckets .................. 32
429
+ replication ..................................... False
430
+ replication_factor .............................. 2
431
+ replication_jump ................................ None
432
+ rerun_mode ...................................... validate_results
433
+ reset_attention_mask ............................ False
434
+ reset_position_ids .............................. False
435
+ result_rejected_tracker_filename ................ None
436
+ retriever_report_topk_accuracies ................ []
437
+ retriever_score_scaling ......................... False
438
+ retriever_seq_length ............................ 256
439
+ retro_add_retriever ............................. False
440
+ retro_attention_gate ............................ 1
441
+ retro_cyclic_train_iters ........................ None
442
+ retro_encoder_attention_dropout ................. 0.1
443
+ retro_encoder_hidden_dropout .................... 0.1
444
+ retro_encoder_layers ............................ 2
445
+ retro_num_neighbors ............................. 2
446
+ retro_num_retrieved_chunks ...................... 2
447
+ retro_project_dir ............................... None
448
+ retro_verify_neighbor_count ..................... True
449
+ reuse_grad_buf_for_mxfp8_param_ag ............... False
450
+ rope_scaling_factor ............................. 8.0
451
+ rope_type ....................................... None
452
+ rotary_base ..................................... 1000000
453
+ rotary_interleaved .............................. False
454
+ rotary_percent .................................. 1.0
455
+ rotary_scaling_factor ........................... 1.0
456
+ rotary_seq_len_interpolation_factor ............. None
457
+ run_workload_inspector_server ................... False
458
+ sample_rate ..................................... 1.0
459
+ save ............................................ None
460
+ save_interval ................................... None
461
+ save_retain_interval ............................ None
462
+ scatter_gather_tensors_in_pipeline .............. True
463
+ seed ............................................ 1234
464
+ seq_length ...................................... 4096
465
+ sequence_parallel ............................... False
466
+ sft ............................................. False
467
+ sft_tokenizer_prompt_format ..................... nemotron-h-aligned
468
+ sgd_momentum .................................... 0.9
469
+ short_seq_prob .................................. 0.1
470
+ skip_train ...................................... False
471
+ skipped_train_samples ........................... 0
472
+ spec ............................................ None
473
+ split ........................................... 100,0,0
474
+ squared_relu .................................... False
475
+ start_weight_decay .............................. 0.1
476
+ straggler_ctrlr_port ............................ 65535
477
+ straggler_minmax_count .......................... 1
478
+ suggested_communication_unit_size ............... None
479
+ swiglu .......................................... True
480
+ swin_backbone_type .............................. tiny
481
+ symmetric_ar_type ............................... None
482
+ te_rng_tracker .................................. True
483
+ tensor_model_parallel_size ...................... 1
484
+ tensorboard_dir ................................. None
485
+ tensorboard_log_interval ........................ 1
486
+ tensorboard_queue_size .......................... 1000
487
+ test_data_path .................................. None
488
+ test_mode ....................................... False
489
+ tiktoken_num_special_tokens ..................... 1000
490
+ tiktoken_pattern ................................ None
491
+ tiktoken_special_tokens ......................... None
492
+ timing_log_level ................................ 0
493
+ timing_log_option ............................... minmax
494
+ titles_data_path ................................ None
495
+ tokenizer_model ................................. None
496
+ tokenizer_type .................................. GPT2BPETokenizer
497
+ torch_fsdp2_reshard_after_forward ............... True
498
+ tp_comm_bootstrap_backend ....................... nccl
499
+ tp_comm_bulk_dgrad .............................. True
500
+ tp_comm_bulk_wgrad .............................. True
501
+ tp_comm_overlap ................................. False
502
+ tp_comm_overlap_ag .............................. True
503
+ tp_comm_overlap_cfg ............................. None
504
+ tp_comm_overlap_rs .............................. True
505
+ tp_comm_overlap_rs_dgrad ........................ False
506
+ tp_comm_split_ag ................................ True
507
+ tp_comm_split_rs ................................ True
508
+ train_data_path ................................. None
509
+ train_iters ..................................... 10
510
+ train_samples ................................... None
511
+ train_sync_interval ............................. None
512
+ transformer_impl ................................ transformer_engine
513
+ transformer_pipeline_model_parallel_size ........ 1
514
+ untie_embeddings_and_output_weights ............. True
515
+ use_checkpoint_args ............................. False
516
+ use_checkpoint_opt_param_scheduler .............. False
517
+ use_cpu_initialization .......................... None
518
+ use_custom_fsdp ................................. False
519
+ use_dist_ckpt ................................... True
520
+ use_dist_ckpt_deprecated ........................ False
521
+ use_distributed_optimizer ....................... True
522
+ use_flash_attn .................................. False
523
+ use_legacy_models ............................... False
524
+ use_mp_args_from_checkpoint_args ................ False
525
+ use_one_sent_docs ............................... False
526
+ use_persistent_ckpt_worker ...................... False
527
+ use_precision_aware_optimizer ................... False
528
+ use_pytorch_profiler ............................ False
529
+ use_ring_exchange_p2p ........................... False
530
+ use_rope_scaling ................................ False
531
+ use_rotary_position_embeddings .................. False
532
+ use_sharp ....................................... False
533
+ use_tokenizer_model_from_checkpoint_args ........ True
534
+ use_torch_fsdp2 ................................. False
535
+ use_torch_optimizer_for_cpu_offload ............. False
536
+ use_tp_pp_dp_mapping ............................ False
537
+ v_head_dim ...................................... 128
538
+ valid_data_path ................................. None
539
+ variable_seq_lengths ............................ False
540
+ virtual_pipeline_model_parallel_size ............ None
541
+ vision_backbone_type ............................ vit
542
+ vision_pretraining .............................. False
543
+ vision_pretraining_type ......................... classify
544
+ vocab_extra_ids ................................. 0
545
+ vocab_file ...................................... /workspace/data/wikitext/gpt2-vocab.json
546
+ vocab_size ...................................... None
547
+ wandb_exp_name ..................................
548
+ wandb_project ...................................
549
+ wandb_save_dir ..................................
550
+ weight_decay .................................... 0.1
551
+ weight_decay_incr_style ......................... constant
552
+ wgrad_deferral_limit ............................ 0
553
+ world_size ...................................... 8
554
+ yaml_cfg ........................................ None
555
+ -------------------- end of arguments ---------------------
556
+ INFO:megatron.core.num_microbatches_calculator:setting number of microbatches to constant 2
557
+ > building GPT2BPETokenizer tokenizer ...
558
+ /workspace/Megatron-MoE-Echo/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
559
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
560
+ > padded vocab (size: 50257) with 47 dummy tokens (new size: 50304)
561
+ WARNING:megatron.core.rerun_state_machine:RerunStateMachine initialized in mode RerunMode.VALIDATE_RESULTS
562
+ > initializing torch distributed ...
563
+ /workspace/Megatron-MoE-Echo/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
564
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
565
+ 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
566
+ /workspace/Megatron-MoE-Echo/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
567
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
568
+ /workspace/Megatron-MoE-Echo/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
569
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
570
+ /workspace/Megatron-MoE-Echo/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
571
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
572
+ /workspace/Megatron-MoE-Echo/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
573
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
574
+ /workspace/Megatron-MoE-Echo/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
575
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
576
+ [Gloo] Rank 0 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
577
+ [Gloo] Rank [Gloo] Rank 5 is connected to 7 peer ranks. Expected number of connected peer ranks is : 47 is connected to 7 peer ranks.
578
+ Expected number of connected peer ranks is : 7
579
+ [Gloo] Rank 1 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
580
+ [Gloo] Rank 7 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
581
+ [Gloo] Rank 3 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
582
+ [Gloo] Rank 2 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
583
+ [Gloo] Rank 6 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
584
+ [Gloo] Rank 0 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
585
+ [Gloo] Rank 1 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
586
+ [Gloo] Rank 2 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
587
+ [Gloo] Rank 4 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
588
+ [Gloo] Rank 6 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
589
+ [Gloo] Rank 3 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
590
+ [Gloo] Rank [Gloo] Rank 57 is connected to is connected to 77 peer ranks. peer ranks. Expected number of connected peer ranks is : Expected number of connected peer ranks is : 77
591
+
592
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
593
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
594
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
595
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
596
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
597
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
598
+ > initialized tensor model parallel with size 1
599
+ > initialized pipeline model parallel with size 1
600
+ > setting random seeds to 1234 ...
601
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
602
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
603
+ [rank5]:[W830 05:53:39.896923336 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.
604
+ [rank2]:[W830 05:53:39.901568780 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.
605
+ [rank6]:[W830 05:53:39.902061347 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.
606
+ [rank7]:[W830 05:53:39.902115151 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.
607
+ [rank4]:[W830 05:53:39.902689768 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.
608
+ [rank3]:[W830 05:53:39.902925718 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.
609
+ [rank1]:[W830 05:53:39.903099780 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.
610
+ > compiling dataset index builder ...
611
+ make: Entering directory '/workspace/Megatron-MoE-Echo/megatron/core/datasets'
612
+ make: Nothing to be done for 'default'.
613
+ make: Leaving directory '/workspace/Megatron-MoE-Echo/megatron/core/datasets'
614
+ >>> done with dataset index builder. Compilation time: 0.429 seconds
615
+ WARNING: constraints for invoking optimized fused softmax kernel are not met. We default back to unfused kernel invocations.
616
+ > compiling and loading fused kernels ...
617
+ [rank0]:[W830 05:53:40.509881319 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.
618
+ >>> done with compiling and loading fused kernels. Compilation time: 3.623 seconds
619
+ time to initialize megatron (seconds): 8.937
620
+ [after megatron is initialized] datetime: 2026-08-30 05:53:46
621
+ building GPT model ...
622
+ /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.
623
+ warnings.warn(
624
+ /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.
625
+ warnings.warn(
626
+ /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.
627
+ warnings.warn(
628
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
629
+ warnings.warn(
630
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
631
+ warnings.warn(
632
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
633
+ warnings.warn(
634
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
635
+ warnings.warn(
636
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
637
+ warnings.warn(
638
+ > number of parameters on (tensor, pipeline) model parallel rank (0, 0): 3902410752
639
+ INFO:megatron.core.distributed.distributed_data_parallel:Setting up DistributedDataParallel with config DistributedDataParallelConfig(grad_reduce_in_fp32=True, overlap_grad_reduce=True, overlap_param_gather=True, align_param_gather=False, use_distributed_optimizer=True, num_distributed_optimizer_instances=1, check_for_nan_in_grad=False, check_for_large_grads=False, bucket_size=40000000, pad_buckets_for_high_nccl_busbw=False, average_in_collective=False, fp8_param_gather=False, reuse_grad_buf_for_mxfp8_param_ag=False, use_custom_fsdp=False, data_parallel_sharding_strategy='no_shard', gradient_reduce_div_fusion=True, suggested_communication_unit_size=None, preserve_fp32_weights=True, keep_fp8_transpose_cache_when_using_custom_fsdp=False, nccl_ub=False, fsdp_double_buffer=False)
640
+ INFO:megatron.core.distributed.param_and_grad_buffer:Number of buckets for gradient all-reduce / reduce-scatter: 18
641
+ Params for bucket 1 (206045184 elements, 206045184 padded size):
642
+ module.output_layer.weight
643
+ Params for bucket 2 (41988096 elements, 41988096 padded size):
644
+ module.decoder.final_layernorm.weight
645
+ module.decoder.layers.15.pre_mlp_layernorm.weight
646
+ module.decoder.layers.15.self_attention.linear_qkv.weight
647
+ module.decoder.layers.15.mlp.router.weight
648
+ module.decoder.layers.15.self_attention.linear_qkv.layer_norm_weight
649
+ module.decoder.layers.15.self_attention.linear_proj.weight
650
+ Params for bucket 3 (41984000 elements, 41984000 padded size):
651
+ module.decoder.layers.14.mlp.router.weight
652
+ module.decoder.layers.14.self_attention.linear_qkv.weight
653
+ module.decoder.layers.14.pre_mlp_layernorm.weight
654
+ module.decoder.layers.14.self_attention.linear_qkv.layer_norm_weight
655
+ module.decoder.layers.14.self_attention.linear_proj.weight
656
+ Params for bucket 4 (41984000 elements, 41984000 padded size):
657
+ module.decoder.layers.13.mlp.router.weight
658
+ module.decoder.layers.13.self_attention.linear_proj.weight
659
+ module.decoder.layers.13.self_attention.linear_qkv.layer_norm_weight
660
+ module.decoder.layers.13.pre_mlp_layernorm.weight
661
+ module.decoder.layers.13.self_attention.linear_qkv.weight
662
+ Params for bucket 5 (41984000 elements, 41984000 padded size):
663
+ module.decoder.layers.12.self_attention.linear_qkv.weight
664
+ module.decoder.layers.12.pre_mlp_layernorm.weight
665
+ module.decoder.layers.12.mlp.router.weight
666
+ module.decoder.layers.12.self_attention.linear_qkv.layer_norm_weight
667
+ module.decoder.layers.12.self_attention.linear_proj.weight
668
+ Params for bucket 6 (41984000 elements, 41984000 padded size):
669
+ module.decoder.layers.11.self_attention.linear_qkv.weight
670
+ module.decoder.layers.11.pre_mlp_layernorm.weight
671
+ module.decoder.layers.11.mlp.router.weight
672
+ module.decoder.layers.11.self_attention.linear_proj.weight
673
+ module.decoder.layers.11.self_attention.linear_qkv.layer_norm_weight
674
+ Params for bucket 7 (41984000 elements, 41984000 padded size):
675
+ module.decoder.layers.10.self_attention.linear_qkv.layer_norm_weight
676
+ module.decoder.layers.10.self_attention.linear_proj.weight
677
+ module.decoder.layers.10.mlp.router.weight
678
+ module.decoder.layers.10.self_attention.linear_qkv.weight
679
+ module.decoder.layers.10.pre_mlp_layernorm.weight
680
+ Params for bucket 8 (41984000 elements, 41984000 padded size):
681
+ module.decoder.layers.9.self_attention.linear_qkv.weight
682
+ module.decoder.layers.9.pre_mlp_layernorm.weight
683
+ module.decoder.layers.9.mlp.router.weight
684
+ module.decoder.layers.9.self_attention.linear_proj.weight
685
+ module.decoder.layers.9.self_attention.linear_qkv.layer_norm_weight
686
+ Params for bucket 9 (41984000 elements, 41984000 padded size):
687
+ module.decoder.layers.8.self_attention.linear_proj.weight
688
+ module.decoder.layers.8.self_attention.linear_qkv.layer_norm_weight
689
+ module.decoder.layers.8.mlp.router.weight
690
+ module.decoder.layers.8.self_attention.linear_qkv.weight
691
+ module.decoder.layers.8.pre_mlp_layernorm.weight
692
+ Params for bucket 10 (41984000 elements, 41984000 padded size):
693
+ module.decoder.layers.7.pre_mlp_layernorm.weight
694
+ module.decoder.layers.7.self_attention.linear_qkv.weight
695
+ module.decoder.layers.7.mlp.router.weight
696
+ module.decoder.layers.7.self_attention.linear_proj.weight
697
+ module.decoder.layers.7.self_attention.linear_qkv.layer_norm_weight
698
+ Params for bucket 11 (41984000 elements, 41984000 padded size):
699
+ module.decoder.layers.6.mlp.router.weight
700
+ module.decoder.layers.6.self_attention.linear_proj.weight
701
+ module.decoder.layers.6.self_attention.linear_qkv.layer_norm_weight
702
+ module.decoder.layers.6.pre_mlp_layernorm.weight
703
+ module.decoder.layers.6.self_attention.linear_qkv.weight
704
+ Params for bucket 12 (41984000 elements, 41984000 padded size):
705
+ module.decoder.layers.5.self_attention.linear_qkv.weight
706
+ module.decoder.layers.5.self_attention.linear_proj.weight
707
+ module.decoder.layers.5.pre_mlp_layernorm.weight
708
+ module.decoder.layers.5.mlp.router.weight
709
+ module.decoder.layers.5.self_attention.linear_qkv.layer_norm_weight
710
+ Params for bucket 13 (41984000 elements, 41984000 padded size):
711
+ module.decoder.layers.4.mlp.router.weight
712
+ module.decoder.layers.4.self_attention.linear_proj.weight
713
+ module.decoder.layers.4.self_attention.linear_qkv.layer_norm_weight
714
+ module.decoder.layers.4.self_attention.linear_qkv.weight
715
+ module.decoder.layers.4.pre_mlp_layernorm.weight
716
+ Params for bucket 14 (41984000 elements, 41984000 padded size):
717
+ module.decoder.layers.3.self_attention.linear_qkv.weight
718
+ module.decoder.layers.3.pre_mlp_layernorm.weight
719
+ module.decoder.layers.3.mlp.router.weight
720
+ module.decoder.layers.3.self_attention.linear_qkv.layer_norm_weight
721
+ module.decoder.layers.3.self_attention.linear_proj.weight
722
+ Params for bucket 15 (41984000 elements, 41984000 padded size):
723
+ module.decoder.layers.2.self_attention.linear_qkv.weight
724
+ module.decoder.layers.2.pre_mlp_layernorm.weight
725
+ module.decoder.layers.2.self_attention.linear_qkv.layer_norm_weight
726
+ module.decoder.layers.2.mlp.router.weight
727
+ module.decoder.layers.2.self_attention.linear_proj.weight
728
+ Params for bucket 16 (41984000 elements, 41984000 padded size):
729
+ module.decoder.layers.1.mlp.router.weight
730
+ module.decoder.layers.1.self_attention.linear_qkv.layer_norm_weight
731
+ module.decoder.layers.1.self_attention.linear_proj.weight
732
+ module.decoder.layers.1.pre_mlp_layernorm.weight
733
+ module.decoder.layers.1.self_attention.linear_qkv.weight
734
+ Params for bucket 17 (41984000 elements, 41984000 padded size):
735
+ module.decoder.layers.0.mlp.router.weight
736
+ module.decoder.layers.0.self_attention.linear_qkv.layer_norm_weight
737
+ module.decoder.layers.0.self_attention.linear_qkv.weight
738
+ module.decoder.layers.0.self_attention.linear_proj.weight
739
+ module.decoder.layers.0.pre_mlp_layernorm.weight
740
+ Params for bucket 18 (206045184 elements, 206045184 padded size):
741
+ module.embedding.word_embeddings.weight
742
+ INFO:megatron.core.distributed.param_and_grad_buffer:Number of buckets for gradient all-reduce / reduce-scatter: 32
743
+ Params for bucket 1 (58720256 elements, 58720256 padded size):
744
+ module.decoder.layers.15.mlp.experts.linear_fc2.weight0
745
+ Params for bucket 2 (117440512 elements, 117440512 padded size):
746
+ module.decoder.layers.15.mlp.experts.linear_fc1.weight0
747
+ Params for bucket 3 (58720256 elements, 58720256 padded size):
748
+ module.decoder.layers.14.mlp.experts.linear_fc2.weight0
749
+ Params for bucket 4 (117440512 elements, 117440512 padded size):
750
+ module.decoder.layers.14.mlp.experts.linear_fc1.weight0
751
+ Params for bucket 5 (58720256 elements, 58720256 padded size):
752
+ module.decoder.layers.13.mlp.experts.linear_fc2.weight0
753
+ Params for bucket 6 (117440512 elements, 117440512 padded size):
754
+ module.decoder.layers.13.mlp.experts.linear_fc1.weight0
755
+ Params for bucket 7 (58720256 elements, 58720256 padded size):
756
+ module.decoder.layers.12.mlp.experts.linear_fc2.weight0
757
+ Params for bucket 8 (117440512 elements, 117440512 padded size):
758
+ module.decoder.layers.12.mlp.experts.linear_fc1.weight0
759
+ Params for bucket 9 (58720256 elements, 58720256 padded size):
760
+ module.decoder.layers.11.mlp.experts.linear_fc2.weight0
761
+ Params for bucket 10 (117440512 elements, 117440512 padded size):
762
+ module.decoder.layers.11.mlp.experts.linear_fc1.weight0
763
+ Params for bucket 11 (58720256 elements, 58720256 padded size):
764
+ module.decoder.layers.10.mlp.experts.linear_fc2.weight0
765
+ Params for bucket 12 (117440512 elements, 117440512 padded size):
766
+ module.decoder.layers.10.mlp.experts.linear_fc1.weight0
767
+ Params for bucket 13 (58720256 elements, 58720256 padded size):
768
+ module.decoder.layers.9.mlp.experts.linear_fc2.weight0
769
+ Params for bucket 14 (117440512 elements, 117440512 padded size):
770
+ module.decoder.layers.9.mlp.experts.linear_fc1.weight0
771
+ Params for bucket 15 (58720256 elements, 58720256 padded size):
772
+ module.decoder.layers.8.mlp.experts.linear_fc2.weight0
773
+ Params for bucket 16 (117440512 elements, 117440512 padded size):
774
+ module.decoder.layers.8.mlp.experts.linear_fc1.weight0
775
+ Params for bucket 17 (58720256 elements, 58720256 padded size):
776
+ module.decoder.layers.7.mlp.experts.linear_fc2.weight0
777
+ Params for bucket 18 (117440512 elements, 117440512 padded size):
778
+ module.decoder.layers.7.mlp.experts.linear_fc1.weight0
779
+ Params for bucket 19 (58720256 elements, 58720256 padded size):
780
+ module.decoder.layers.6.mlp.experts.linear_fc2.weight0
781
+ Params for bucket 20 (117440512 elements, 117440512 padded size):
782
+ module.decoder.layers.6.mlp.experts.linear_fc1.weight0
783
+ Params for bucket 21 (58720256 elements, 58720256 padded size):
784
+ module.decoder.layers.5.mlp.experts.linear_fc2.weight0
785
+ Params for bucket 22 (117440512 elements, 117440512 padded size):
786
+ module.decoder.layers.5.mlp.experts.linear_fc1.weight0
787
+ Params for bucket 23 (58720256 elements, 58720256 padded size):
788
+ module.decoder.layers.4.mlp.experts.linear_fc2.weight0
789
+ Params for bucket 24 (117440512 elements, 117440512 padded size):
790
+ module.decoder.layers.4.mlp.experts.linear_fc1.weight0
791
+ Params for bucket 25 (58720256 elements, 58720256 padded size):
792
+ module.decoder.layers.3.mlp.experts.linear_fc2.weight0
793
+ Params for bucket 26 (117440512 elements, 117440512 padded size):
794
+ module.decoder.layers.3.mlp.experts.linear_fc1.weight0
795
+ Params for bucket 27 (58720256 elements, 58720256 padded size):
796
+ module.decoder.layers.2.mlp.experts.linear_fc2.weight0
797
+ Params for bucket 28 (117440512 elements, 117440512 padded size):
798
+ module.decoder.layers.2.mlp.experts.linear_fc1.weight0
799
+ Params for bucket 29 (58720256 elements, 58720256 padded size):
800
+ module.decoder.layers.1.mlp.experts.linear_fc2.weight0
801
+ Params for bucket 30 (117440512 elements, 117440512 padded size):
802
+ module.decoder.layers.1.mlp.experts.linear_fc1.weight0
803
+ Params for bucket 31 (58720256 elements, 58720256 padded size):
804
+ module.decoder.layers.0.mlp.experts.linear_fc2.weight0
805
+ Params for bucket 32 (117440512 elements, 117440512 padded size):
806
+ module.decoder.layers.0.mlp.experts.linear_fc1.weight0
807
+ 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 0x73fa6de53920>, config_logger_dir='')
808
+ INFO:megatron.core.optimizer_param_scheduler:> learning rate decay style: cosine
809
+ [after model, optimizer, and learning rate scheduler are built] datetime: 2026-08-30 05:53:46
810
+ > building train, validation, and test datasets ...
811
+ > datasets target sizes (minimum size):
812
+ train: 160
813
+ validation: 0
814
+ test: 0
815
+ INFO:megatron.core.datasets.blended_megatron_dataset_config:Let split_matrix = [(0, 1.0), None, None]
816
+ > building train, validation, and test datasets for GPT ...
817
+ INFO:megatron.core.datasets.blended_megatron_dataset_builder:Building GPTDataset splits with sizes=(160, 0, 0) and config=GPTDatasetConfig(random_seed=1234, sequence_length=4096, blend=(['/workspace/data/wikitext/wikitext103_gpt2_text_document'], None), blend_per_split=None, split='100,0,0', split_matrix=[(0, 1.0), None, None], num_dataset_builder_threads=1, path_to_cache=None, mmap_bin_files=True, mock=False, tokenizer=<megatron.training.tokenizer.tokenizer._GPT2BPETokenizer object at 0x73fa74c767b0>, 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)
818
+ INFO:megatron.core.datasets.indexed_dataset:Load the _IndexReader from /workspace/data/wikitext/wikitext103_gpt2_text_document.idx
819
+ INFO:megatron.core.datasets.indexed_dataset: Extract the sequence lengths
820
+ INFO:megatron.core.datasets.indexed_dataset: Extract the sequence pointers
821
+ INFO:megatron.core.datasets.indexed_dataset: Extract the document indices
822
+ INFO:megatron.core.datasets.indexed_dataset:> total number of sequences: 1165029
823
+ INFO:megatron.core.datasets.indexed_dataset:> total number of documents: 1801350
824
+ INFO:megatron.core.datasets.gpt_dataset:Build and save the GPTDataset train indices
825
+ INFO:megatron.core.datasets.gpt_dataset:> total number of samples: 29073
826
+ INFO:megatron.core.datasets.gpt_dataset:> total number of epochs: 1
827
+ > finished creating GPT datasets ...
828
+ [after dataloaders are built] datetime: 2026-08-30 05:53:46
829
+ done with setup ...
830
+ training ...
831
+ Overwriting rerun_state_machine.current_iteration from -1 to 0...
832
+ (min, max) time across ranks (ms):
833
+ model-and-optimizer-setup ......................: (231.85, 234.54)
834
+ train/valid/test-data-iterators-setup ..........: (144.29, 160.90)
835
+ [before the start of training step] datetime: 2026-08-30 05:53:46
836
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
837
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
838
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
839
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
840
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
841
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
842
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
843
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
844
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
845
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
846
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
847
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
848
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
849
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
850
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
851
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
852
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
853
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
854
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
855
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
856
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
857
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
858
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
859
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
860
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
861
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
862
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
863
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
864
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
865
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
866
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
867
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
868
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
869
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
870
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
871
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
872
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
873
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
874
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
875
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
876
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
877
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
878
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
879
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
880
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
881
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
882
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
883
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
884
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
885
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
886
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
887
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
888
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
889
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
890
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
891
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
892
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
893
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
894
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
895
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
896
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
897
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
898
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
899
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
900
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
901
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
902
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
903
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
904
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
905
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
906
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
907
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
908
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
909
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
910
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
911
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
912
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
913
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
914
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
915
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
916
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
917
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
918
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
919
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
920
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
921
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
922
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
923
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
924
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
925
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
926
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
927
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
928
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
929
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
930
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
931
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
932
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
933
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
934
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
935
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
936
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
937
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
938
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
939
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
940
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
941
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
942
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
943
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
944
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
945
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
946
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
947
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
948
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
949
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
950
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
951
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
952
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
953
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
954
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
955
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
956
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
957
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
958
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
959
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
960
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
961
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
962
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
963
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
964
+ [2026-08-30 05:54:29] iteration 1/ 10 | consumed samples: 16 | elapsed time per iteration (ms): 43390.1 | throughput per GPU (TFLOP/s/GPU): 7.7 | learning rate: 2.000000E-05 | global batch size: 16 | lm loss: 1.104376E+01 | loss scale: 1.0 | grad norm: 34.299 | number of skipped iterations: 0 | number of nan iterations: 0 |
965
+ Number of parameters in transformer block in billions: 23.22
966
+ Number of parameters in embedding layers in billions: 0.41
967
+ Total number of parameters in billions: 23.63
968
+ Number of parameters in most loaded shard in billions: 23.6320
969
+ Theoretical memory footprints: weight and optimizer=169029.43 MB
970
+ [Rank 0] (after 1 iterations) memory (MB) | allocated: 63795.412109375 | max allocated: 96519.68359375 | reserved: 99120.0 | max reserved: 99120.0
971
+ [2026-08-30 05:54:30] iteration 2/ 10 | consumed samples: 32 | elapsed time per iteration (ms): 599.9 | throughput per GPU (TFLOP/s/GPU): 555.8 | learning rate: 4.000000E-05 | global batch size: 16 | lm loss: 1.104935E+01 | loss scale: 1.0 | grad norm: 35.984 | number of skipped iterations: 0 | number of nan iterations: 0 |
972
+ [2026-08-30 05:54:31] iteration 3/ 10 | consumed samples: 48 | elapsed time per iteration (ms): 579.1 | throughput per GPU (TFLOP/s/GPU): 575.7 | learning rate: 6.000000E-05 | global batch size: 16 | lm loss: 9.657288E+00 | loss scale: 1.0 | grad norm: 40.305 | number of skipped iterations: 0 | number of nan iterations: 0 |
973
+ Capture CUDA graph for training!!!
974
+ CUDA graph capture done!!!
975
+ [2026-08-30 05:54:32] iteration 4/ 10 | consumed samples: 64 | elapsed time per iteration (ms): 1697.5 | throughput per GPU (TFLOP/s/GPU): 196.4 | learning rate: 8.000000E-05 | global batch size: 16 | lm loss: 1.081466E+01 | loss scale: 1.0 | grad norm: 92.896 | number of skipped iterations: 0 | number of nan iterations: 0 |
976
+ [2026-08-30 05:54:33] iteration 5/ 10 | consumed samples: 80 | elapsed time per iteration (ms): 473.3 | throughput per GPU (TFLOP/s/GPU): 704.3 | learning rate: 1.000000E-04 | global batch size: 16 | lm loss: 1.067598E+01 | loss scale: 1.0 | grad norm: 17.037 | number of skipped iterations: 0 | number of nan iterations: 0 |
977
+ [2026-08-30 05:54:33] iteration 6/ 10 | consumed samples: 96 | elapsed time per iteration (ms): 467.1 | throughput per GPU (TFLOP/s/GPU): 713.8 | learning rate: 7.750000E-05 | global batch size: 16 | lm loss: 9.872430E+00 | loss scale: 1.0 | grad norm: 40.021 | number of skipped iterations: 0 | number of nan iterations: 0 |
978
+ [2026-08-30 05:54:37] iteration 7/ 10 | consumed samples: 112 | elapsed time per iteration (ms): 3244.9 | throughput per GPU (TFLOP/s/GPU): 102.7 | learning rate: 3.250000E-05 | global batch size: 16 | lm loss: 9.286396E+00 | loss scale: 1.0 | grad norm: 7.509 | number of skipped iterations: 0 | number of nan iterations: 0 |
979
+ [2026-08-30 05:54:37] iteration 8/ 10 | consumed samples: 128 | elapsed time per iteration (ms): 477.8 | throughput per GPU (TFLOP/s/GPU): 697.7 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 9.027843E+00 | loss scale: 1.0 | grad norm: 21.578 | number of skipped iterations: 0 | number of nan iterations: 0 |
980
+ Capture range started in the application.
981
+ Capture range ended in the application.
982
+ Generating '/tmp/nsys-report-cfc2.qdstrm'
983
+
984
+
tiera-rerun-fixed-moonep/echo/a1/a1_oneslot_gbs16/G1__echo__one__seq4096__gbs16__cap2p40625/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, 629
43
+ 2, 629
44
+ 3, 27
45
+ 4, 629
46
+ 5, 629
47
+ 6, 629
48
+ 7, 629
49
+ 0, 917
50
+ 1, 917
51
+ 2, 913
52
+ 3, 909
53
+ 4, 925
54
+ 5, 929
55
+ 6, 931
56
+ 7, 927
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, 39517
66
+ 1, 39497
67
+ 2, 39449
68
+ 3, 39501
69
+ 4, 39523
70
+ 5, 39537
71
+ 6, 39519
72
+ 7, 39537
73
+ 0, 39839
74
+ 1, 39803
75
+ 2, 39851
76
+ 3, 39817
77
+ 4, 39829
78
+ 5, 39841
79
+ 6, 39817
80
+ 7, 39825
81
+ 0, 42357
82
+ 1, 42355
83
+ 2, 42355
84
+ 3, 42355
85
+ 4, 42355
86
+ 5, 42355
87
+ 6, 42355
88
+ 7, 42355
89
+ 0, 42583
90
+ 1, 42581
91
+ 2, 42581
92
+ 3, 42581
93
+ 4, 42581
94
+ 5, 42581
95
+ 6, 42581
96
+ 7, 42581
97
+ 0, 42761
98
+ 1, 42755
99
+ 2, 42745
100
+ 3, 42739
101
+ 4, 42753
102
+ 5, 42783
103
+ 6, 42801
104
+ 7, 42767
105
+ 0, 43493
106
+ 1, 43491
107
+ 2, 43491
108
+ 3, 43491
109
+ 4, 43491
110
+ 5, 43491
111
+ 6, 43491
112
+ 7, 43491
113
+ 0, 43493
114
+ 1, 43491
115
+ 2, 43491
116
+ 3, 43491
117
+ 4, 43491
118
+ 5, 43491
119
+ 6, 43491
120
+ 7, 43491
121
+ 0, 45265
122
+ 1, 45263
123
+ 2, 45263
124
+ 3, 45263
125
+ 4, 45263
126
+ 5, 45263
127
+ 6, 45263
128
+ 7, 45263
129
+ 0, 45265
130
+ 1, 45263
131
+ 2, 45263
132
+ 3, 45263
133
+ 4, 45263
134
+ 5, 45263
135
+ 6, 45263
136
+ 7, 45263
137
+ 0, 85827
138
+ 1, 85825
139
+ 2, 85825
140
+ 3, 85825
141
+ 4, 85825
142
+ 5, 85825
143
+ 6, 85825
144
+ 7, 85825
145
+ 0, 86905
146
+ 1, 86903
147
+ 2, 86903
148
+ 3, 86903
149
+ 4, 86903
150
+ 5, 86903
151
+ 6, 86903
152
+ 7, 86903
153
+ 0, 86905
154
+ 1, 86903
155
+ 2, 86903
156
+ 3, 86903
157
+ 4, 86903
158
+ 5, 86903
159
+ 6, 86903
160
+ 7, 86903
161
+ 0, 86905
162
+ 1, 86903
163
+ 2, 86903
164
+ 3, 86903
165
+ 4, 86903
166
+ 5, 86903
167
+ 6, 86903
168
+ 7, 86903
169
+ 0, 103941
170
+ 1, 103907
171
+ 2, 103881
172
+ 3, 103907
173
+ 4, 104003
174
+ 5, 103905
175
+ 6, 103919
176
+ 7, 103927
177
+ 0, 105805
178
+ 1, 105675
179
+ 2, 105739
180
+ 3, 105769
181
+ 4, 105829
182
+ 5, 105781
183
+ 6, 105773
184
+ 7, 105753
185
+ 0, 106523
186
+ 1, 106465
187
+ 2, 106465
188
+ 3, 106465
189
+ 4, 106521
190
+ 5, 106465
191
+ 6, 106465
192
+ 7, 106465
193
+ 0, 107099
194
+ 1, 107065
195
+ 2, 107055
196
+ 3, 107039
197
+ 4, 107179
198
+ 5, 107127
199
+ 6, 107143
200
+ 7, 107137
201
+ 0, 131205
202
+ 1, 131765
203
+ 2, 132663
204
+ 3, 130871
205
+ 4, 132857
206
+ 5, 146573
207
+ 6, 133783
208
+ 7, 133781
209
+ 0, 161639
210
+ 1, 161525
211
+ 2, 161751
212
+ 3, 161529
213
+ 4, 161947
214
+ 5, 161525
215
+ 6, 161525
216
+ 7, 161749
217
+ 0, 161639
218
+ 1, 161525
219
+ 2, 161751
220
+ 3, 161529
221
+ 4, 161947
222
+ 5, 161525
223
+ 6, 161525
224
+ 7, 161749
225
+ 0, 3695
226
+ 1, 2098
227
+ 2, 2098
228
+ 3, 2098
229
+ 4, 2098
230
+ 5, 2098
231
+ 6, 2418
232
+ 7, 161749
tiera-rerun-fixed-moonep/echo/a1/a1_oneslot_gbs16/G1__echo__one__seq4096__gbs16__cap2p40625/rep1/run.env ADDED
@@ -0,0 +1,70 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ backend=echo
2
+ config=G1
3
+ family=tierA
4
+ repo=/workspace/Megatron-MoE-Echo
5
+ branch=bracing-regions-through-kernels
6
+ commit=9a236793
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=one
42
+ slots_per_rank=1
43
+ capacity_mode=static
44
+ capacity=2.40625
45
+ cuda_graph=1
46
+ probe_mode=0
47
+ async_grad_reduce=0
48
+ branch_role=timing
49
+ memory_dump_mode=none
50
+ memory_snapshot_path=
51
+ cuda_graph_warmup_steps=3
52
+ recompute=none
53
+ force_balance=0
54
+ train_iters=12
55
+ ep_size=8
56
+ echo_dispatch_overlap=1
57
+ echo_dispatch_recompute=0
58
+ ultraep_quota_min_tokens=
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/echo/a1/a1_oneslot_gbs16/G1__echo__one__seq4096__gbs16__cap2p40625/rep1/train.log ADDED
@@ -0,0 +1,1023 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ==============================================================
2
+ G1 / echo (slots/rank=1 policy=one)
3
+ E=8 K=2 H'=14336 m=1
4
+ L=16 seq=4096 gbs=16 (2 microbatches)
5
+ cap_mode=static cap=2.40625 cg=1 recompute=none
6
+ memory_dump=none snapshots=/workspace/tiera-rerun-fixed-moonep/echo/a1/a1_oneslot_gbs16/G1__echo__one__seq4096__gbs16__cap2p40625/rep1/memory_snapshots/snapshot.pickle
7
+ repo=/workspace/Megatron-MoE-Echo @ 9a236793
8
+ ==============================================================
9
+ W0830 05:49:33.358000 545455 torch/distributed/run.py:766]
10
+ W0830 05:49:33.358000 545455 torch/distributed/run.py:766] *****************************************
11
+ W0830 05:49:33.358000 545455 torch/distributed/run.py:766] Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed.
12
+ W0830 05:49:33.358000 545455 torch/distributed/run.py:766] *****************************************
13
+ /workspace/Megatron-MoE-Echo/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
14
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
15
+ /workspace/Megatron-MoE-Echo/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
16
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
17
+ /workspace/Megatron-MoE-Echo/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
18
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
19
+ /workspace/Megatron-MoE-Echo/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
20
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
21
+ /workspace/Megatron-MoE-Echo/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
22
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
23
+ 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
24
+ /workspace/Megatron-MoE-Echo/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
25
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
26
+ using world size: 8, data-parallel size: 8, context-parallel size: 1, hierarchical context-parallel sizes: None, tensor-model-parallel size: 1, pipeline-model-parallel size: 1
27
+ WARNING: overriding default arguments for tokenizer_type:GPT2BPETokenizer with tokenizer_type:GPT2BPETokenizer
28
+ Number of virtual stages per pipeline stage: None
29
+ accumulate and all-reduce gradients in fp32 for bfloat16 data type.
30
+ using torch.bfloat16 for parameters ...
31
+ /workspace/Megatron-MoE-Echo/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
32
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
33
+ ------------------------ arguments ------------------------
34
+ account_for_embedding_in_pipeline_split ......... False
35
+ account_for_loss_in_pipeline_split .............. False
36
+ accumulate_allreduce_grads_in_fp32 .............. True
37
+ adam_beta1 ...................................... 0.9
38
+ adam_beta2 ...................................... 0.999
39
+ adam_eps ........................................ 1e-08
40
+ add_bias_linear ................................. False
41
+ add_position_embedding .......................... False
42
+ add_qkv_bias .................................... False
43
+ adlr_autoresume ................................. False
44
+ adlr_autoresume_interval ........................ 1000
45
+ align_grad_reduce ............................... True
46
+ align_param_gather .............................. False
47
+ app_tag_run_name ................................ None
48
+ app_tag_run_version ............................. 0.0.0
49
+ apply_layernorm_1p .............................. False
50
+ apply_query_key_layer_scaling ................... False
51
+ apply_residual_connection_post_layernorm ........ False
52
+ apply_rope_fusion ............................... True
53
+ async_save ...................................... None
54
+ async_tensor_model_parallel_allreduce ........... True
55
+ attention_backend ............................... AttnBackend.auto
56
+ attention_dropout ............................... 0.0
57
+ attention_softmax_in_fp32 ....................... False
58
+ auto_detect_ckpt_format ......................... False
59
+ barrier_with_L1_time ............................ True
60
+ bert_binary_head ................................ True
61
+ bert_embedder_type .............................. megatron
62
+ bert_load ....................................... None
63
+ bf16 ............................................ True
64
+ bias_dropout_fusion ............................. True
65
+ bias_gelu_fusion ................................ False
66
+ bias_swiglu_fusion .............................. True
67
+ biencoder_projection_dim ........................ 0
68
+ biencoder_shared_query_context_model ............ False
69
+ block_data_path ................................. None
70
+ calc_ft_timeouts ................................ False
71
+ calculate_per_token_loss ........................ False
72
+ check_for_large_grads ........................... False
73
+ check_for_nan_in_loss_and_grad .................. False
74
+ check_for_spiky_loss ............................ False
75
+ check_weight_hash_across_dp_replicas_interval ... None
76
+ ckpt_assume_constant_structure .................. False
77
+ ckpt_convert_format ............................. None
78
+ ckpt_convert_save ............................... None
79
+ ckpt_convert_update_legacy_dist_opt_format ...... False
80
+ ckpt_format ..................................... torch_dist
81
+ ckpt_fully_parallel_load ........................ False
82
+ ckpt_fully_parallel_save ........................ True
83
+ ckpt_fully_parallel_save_deprecated ............. False
84
+ ckpt_step ....................................... None
85
+ classes_fraction ................................ 1.0
86
+ clip_grad ....................................... 1.0
87
+ clone_scatter_output_in_embedding ............... True
88
+ config_logger_dir ...............................
89
+ consumed_train_samples .......................... 0
90
+ consumed_valid_samples .......................... 0
91
+ context_parallel_size ........................... 1
92
+ cp_comm_type .................................... ['p2p']
93
+ create_attention_mask_in_dataloader ............. True
94
+ cross_entropy_fusion_impl ....................... native
95
+ cross_entropy_loss_fusion ....................... False
96
+ cuda_graph_scope ................................ full_iteration
97
+ cuda_graph_warmup_steps ......................... 3
98
+ data_args_path .................................. None
99
+ data_cache_path ................................. None
100
+ data_parallel_random_init ....................... False
101
+ data_parallel_sharding_strategy ................. no_shard
102
+ data_parallel_size .............................. 8
103
+ data_path ....................................... ['/workspace/data/wikitext/wikitext103_gpt2_text_document']
104
+ data_per_class_fraction ......................... 1.0
105
+ data_sharding ................................... True
106
+ dataloader_type ................................. single
107
+ ddp_average_in_collective ....................... False
108
+ ddp_bucket_size ................................. None
109
+ ddp_num_buckets ................................. None
110
+ ddp_pad_buckets_for_high_nccl_busbw ............. False
111
+ decoder_first_pipeline_num_layers ............... None
112
+ decoder_last_pipeline_num_layers ................ None
113
+ decoder_num_layers .............................. None
114
+ decoder_seq_length .............................. None
115
+ decoupled_lr .................................... None
116
+ decoupled_min_lr ................................ None
117
+ decrease_batch_size_if_needed ................... False
118
+ defer_embedding_wgrad_compute ................... False
119
+ delay_wgrad_compute ............................. False
120
+ deprecated_use_mcore_models ..................... True
121
+ deterministic_mode .............................. False
122
+ dino_bottleneck_size ............................ 256
123
+ dino_freeze_last_layer .......................... 1
124
+ dino_head_hidden_size ........................... 2048
125
+ dino_local_crops_number ......................... 10
126
+ dino_local_img_size ............................. 96
127
+ dino_norm_last_layer ............................ False
128
+ dino_teacher_temp ............................... 0.07
129
+ dino_warmup_teacher_temp ........................ 0.04
130
+ dino_warmup_teacher_temp_epochs ................. 30
131
+ disable_bf16_reduced_precision_matmul ........... False
132
+ disable_mamba_mem_eff_path ...................... False
133
+ disable_straggler_on_startup .................... False
134
+ dist_ckpt_format_deprecated ..................... None
135
+ dist_ckpt_strictness ............................ assume_ok_unexpected
136
+ distribute_saved_activations .................... False
137
+ distributed_backend ............................. nccl
138
+ distributed_timeout_minutes ..................... 10
139
+ embedding_init_method_std ....................... None
140
+ embedding_path .................................. None
141
+ empty_unused_memory_level ....................... 0
142
+ enable_cuda_graph ............................... True
143
+ enable_experimental ............................. False
144
+ enable_ft_package ............................... False
145
+ enable_gloo_process_groups ...................... True
146
+ enable_msc ...................................... True
147
+ enable_one_logger ............................... True
148
+ encoder_num_layers .............................. 16
149
+ encoder_seq_length .............................. 4096
150
+ end_weight_decay ................................ 0.1
151
+ eod_mask_loss ................................... False
152
+ error_injection_rate ............................ 0
153
+ error_injection_type ............................ transient_error
154
+ eval_interval ................................... 1000000
155
+ eval_iters ...................................... 0
156
+ evidence_data_path .............................. None
157
+ exit_duration_in_mins ........................... None
158
+ exit_interval ................................... None
159
+ exit_on_missing_checkpoint ...................... False
160
+ exit_signal_handler ............................. False
161
+ exp_avg_dtype ................................... torch.float32
162
+ exp_avg_sq_dtype ................................ torch.float32
163
+ expert_model_parallel_size ...................... 8
164
+ expert_tensor_parallel_size ..................... 1
165
+ export_force_local_attention .................... False
166
+ export_kd_cfg ................................... None
167
+ export_kd_teacher_ckpt_format ................... None
168
+ export_kd_teacher_load .......................... None
169
+ export_kv_cache_quant ........................... False
170
+ export_legacy_megatron .......................... False
171
+ export_model_type ............................... GPTModel
172
+ export_moe_apply_probs_on_input ................. False
173
+ export_qk_l2_norm ............................... False
174
+ export_quant_cfg ................................ None
175
+ export_real_quant_cfg ........................... None
176
+ export_te_mcore_model ........................... False
177
+ external_cuda_graph ............................. False
178
+ ffn_hidden_size ................................. 14336
179
+ finetune ........................................ False
180
+ finetune_data_split ............................. train
181
+ finetune_hf_dataset ............................. None
182
+ first_last_layers_bf16 .......................... False
183
+ flash_decode .................................... False
184
+ fp16 ............................................ False
185
+ fp16_lm_cross_entropy ........................... False
186
+ fp32_residual_connection ........................ False
187
+ fp8 ............................................. e4m3
188
+ fp8_amax_compute_algo ........................... most_recent
189
+ fp8_amax_history_len ............................ 1
190
+ fp8_interval .................................... 1
191
+ fp8_margin ...................................... 0
192
+ fp8_param_gather ................................ False
193
+ fp8_recipe ...................................... mxfp8
194
+ fp8_wgrad ....................................... True
195
+ fsdp_double_buffer .............................. False
196
+ global_batch_size ............................... 16
197
+ grad_reduce_in_bf16 ............................. False
198
+ gradient_accumulation_fusion .................... True
199
+ gradient_reduce_div_fusion ...................... True
200
+ group_query_attention ........................... True
201
+ head_lr_mult .................................... 1.0
202
+ heterogeneous_layers_config_encoded_json ........ None
203
+ heterogeneous_layers_config_path ................ None
204
+ hidden_dropout .................................. 0.0
205
+ hidden_size ..................................... 4096
206
+ hierarchical_context_parallel_sizes ............. None
207
+ high_priority_stream_groups ..................... []
208
+ hybrid_attention_ratio .......................... 0.0
209
+ hybrid_mlp_ratio ................................ 0.0
210
+ hybrid_override_pattern ......................... None
211
+ hysteresis ...................................... 2
212
+ ict_head_size ................................... None
213
+ ict_load ........................................ None
214
+ img_h ........................................... 224
215
+ img_w ........................................... 224
216
+ indexer_batch_size .............................. 128
217
+ indexer_log_interval ............................ 1000
218
+ inference_batch_times_seqlen_threshold .......... -1
219
+ inference_dynamic_batching ...................... False
220
+ inference_dynamic_batching_buffer_guaranteed_fraction 0.2
221
+ inference_dynamic_batching_buffer_overflow_factor None
222
+ inference_dynamic_batching_buffer_size_gb ....... 40.0
223
+ inference_dynamic_batching_chunk_size ........... 256
224
+ inference_dynamic_batching_max_requests_override None
225
+ inference_dynamic_batching_max_tokens_override .. None
226
+ inference_dynamic_batching_num_cuda_graphs ...... 16
227
+ inference_max_batch_size ........................ 8
228
+ inference_max_seq_length ........................ 2560
229
+ inference_rng_tracker ........................... False
230
+ init_method_std ................................. 0.01
231
+ init_method_xavier_uniform ...................... False
232
+ init_model_with_meta_device ..................... False
233
+ initial_loss_scale .............................. 4294967296
234
+ inprocess_active_world_size ..................... 8
235
+ inprocess_barrier_timeout ....................... 120
236
+ inprocess_completion_timeout .................... 120
237
+ inprocess_empty_cuda_cache ...................... False
238
+ inprocess_granularity ........................... node
239
+ inprocess_hard_timeout .......................... 90
240
+ inprocess_heartbeat_interval .................... 30
241
+ inprocess_heartbeat_timeout ..................... 60
242
+ inprocess_last_call_wait ........................ 1
243
+ inprocess_max_iterations ........................ None
244
+ inprocess_monitor_process_interval .............. 1.0
245
+ inprocess_monitor_thread_interval ............... 1.0
246
+ inprocess_progress_watchdog_interval ............ 1.0
247
+ inprocess_restart ............................... False
248
+ inprocess_soft_timeout .......................... 60
249
+ inprocess_termination_grace_time ................ 1
250
+ is_hybrid_model ................................. False
251
+ iter_per_epoch .................................. 1250
252
+ iterations_to_skip .............................. []
253
+ keep_fp8_transpose_cache_when_using_custom_fsdp . False
254
+ kitchen_config_file ............................. None
255
+ kitchen_recipe_number ........................... None
256
+ kv_channels ..................................... 128
257
+ kv_lora_rank .................................... 32
258
+ lazy_mpu_init ................................... None
259
+ load ............................................ None
260
+ load_main_params_from_ckpt ...................... None
261
+ load_model_opt_format ........................... False
262
+ local_rank ...................................... 0
263
+ log_energy ...................................... False
264
+ log_interval .................................... 1
265
+ log_loss_scale_to_tensorboard ................... True
266
+ log_memory_to_tensorboard ....................... False
267
+ log_num_zeros_in_grad ........................... False
268
+ log_params_norm ................................. False
269
+ log_progress .................................... False
270
+ log_straggler ................................... False
271
+ log_throughput .................................. True
272
+ log_timers_to_tensorboard ....................... False
273
+ log_validation_ppl_to_tensorboard ............... False
274
+ log_world_size_to_tensorboard ................... False
275
+ logging_level ................................... None
276
+ loss_scale ...................................... None
277
+ loss_scale_window ............................... 1000
278
+ lr .............................................. 0.0001
279
+ lr_decay_iters .................................. 8
280
+ lr_decay_samples ................................ None
281
+ lr_decay_style .................................. cosine
282
+ lr_warmup_fraction .............................. None
283
+ lr_warmup_init .................................. 0.0
284
+ lr_warmup_iters ................................. 5
285
+ lr_warmup_samples ............................... 0
286
+ lr_wsd_decay_iters .............................. None
287
+ lr_wsd_decay_samples ............................ None
288
+ lr_wsd_decay_style .............................. exponential
289
+ main_grads_dtype ................................ torch.float32
290
+ main_params_dtype ............................... torch.float32
291
+ make_vocab_size_divisible_by .................... 128
292
+ mamba_head_dim .................................. 64
293
+ mamba_num_groups ................................ 8
294
+ mamba_num_heads ................................. None
295
+ mamba_state_dim ................................. 128
296
+ manual_gc ....................................... False
297
+ manual_gc_eval .................................. True
298
+ manual_gc_interval .............................. 0
299
+ mask_factor ..................................... 1.0
300
+ mask_prob ....................................... 0.15
301
+ mask_type ....................................... random
302
+ masked_softmax_fusion ........................... False
303
+ max_position_embeddings ......................... 32768
304
+ max_tokens_to_oom ............................... 12000
305
+ memory_snapshot_path ............................ snapshot.pickle
306
+ merge_file ...................................... /workspace/data/wikitext/gpt2-merges.txt
307
+ micro_batch_size ................................ 1
308
+ microbatch_group_size_per_vp_stage .............. None
309
+ mid_level_dataset_surplus ....................... 0.005
310
+ min_loss_scale .................................. 1.0
311
+ min_lr .......................................... 1e-05
312
+ mlp_chunks_for_prefill .......................... 1
313
+ mmap_bin_files .................................. True
314
+ mock_data ....................................... False
315
+ moe_apply_probs_on_input ........................ False
316
+ moe_aux_loss_coeff .............................. 0.01
317
+ moe_deepep_num_sms .............................. 20
318
+ moe_echo_enable_random_offloading ............... False
319
+ moe_echo_expert_dispatch_overlap ................ True
320
+ moe_echo_expert_dispatcher_type ................. hybridep
321
+ moe_echo_recompute_expert_dispatch .............. False
322
+ moe_enable_deepep ............................... False
323
+ moe_enable_echo ................................. True
324
+ moe_enable_hybridep ............................. True
325
+ moe_expert_capacity_factor ...................... None
326
+ moe_extended_tp ................................. False
327
+ moe_ffn_hidden_size ............................. 14336
328
+ moe_grouped_gemm ................................ True
329
+ moe_hybridep_num_sms ............................ 16
330
+ moe_input_jitter_eps ............................ None
331
+ moe_layer_freq .................................. 1
332
+ moe_layer_recompute ............................. False
333
+ moe_num_echo_experts ............................ 8
334
+ moe_pad_expert_input_to_capacity ................ False
335
+ moe_per_layer_logging ........................... False
336
+ moe_permute_fusion .............................. False
337
+ moe_received_token_capacity ..................... 2.40625
338
+ moe_router_bias_update_rate ..................... 0.001
339
+ moe_router_dtype ................................ fp32
340
+ moe_router_enable_expert_bias ................... False
341
+ moe_router_force_load_balancing ................. False
342
+ moe_router_group_topk ........................... None
343
+ moe_router_load_balancing_type .................. aux_loss
344
+ moe_router_num_groups ........................... None
345
+ moe_router_padding_for_fp8 ...................... False
346
+ moe_router_pre_softmax .......................... False
347
+ moe_router_score_function ....................... softmax
348
+ moe_router_topk ................................. 2
349
+ moe_router_topk_scaling_factor .................. None
350
+ moe_shared_expert_intermediate_size ............. None
351
+ moe_shared_expert_overlap ....................... False
352
+ moe_token_dispatcher_type ....................... flex
353
+ moe_token_drop_policy ........................... probs
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 ..................................... 8
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 .................................. True
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
+ > padded vocab (size: 50257) with 47 dummy tokens (new size: 50304)
572
+ WARNING:megatron.core.rerun_state_machine:RerunStateMachine initialized in mode RerunMode.VALIDATE_RESULTS
573
+ > initializing torch distributed ...
574
+ /workspace/Megatron-MoE-Echo/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
575
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
576
+ [Gloo] Rank 0 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
577
+ [Gloo] Rank 1 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
578
+ [Gloo] Rank 2 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
579
+ [Gloo] Rank 3 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
580
+ [Gloo] Rank 5 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
581
+ [Gloo] Rank 4 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
582
+ [Gloo] Rank [Gloo] Rank 7 is connected to 67 is connected to peer ranks. 7Expected number of connected peer ranks is : peer ranks. 7Expected number of connected peer ranks is :
583
+ 7
584
+ [Gloo] Rank 2 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
585
+ [Gloo] Rank 1 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
586
+ [Gloo] Rank 0 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
587
+ [Gloo] Rank 3 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
588
+ [Gloo] Rank [Gloo] Rank 54 is connected to is connected to 77 peer ranks. peer ranks. Expected number of connected peer ranks is : Expected number of connected peer ranks is : 77
589
+
590
+ [Gloo] Rank 7 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
591
+ [Gloo] Rank 6 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
592
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
593
+ > initialized tensor model parallel with size 1
594
+ > initialized pipeline model parallel with size 1
595
+ > setting random seeds to 1234 ...
596
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
597
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0[Gloo] Rank
598
+ 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
599
+ [Gloo] Rank 0 is connected to 0[Gloo] Rank peer ranks. [Gloo] Rank Expected number of connected peer ranks is : 000 is connected to is connected to
600
+ 00 peer ranks. peer ranks. Expected number of connected peer ranks is : Expected number of connected peer ranks is : 00
601
+
602
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
603
+ [rank1]:[W830 05:49:50.182830316 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.
604
+ [rank7]:[W830 05:49:50.190825306 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.
605
+ [rank3]:[W830 05:49:50.190916787 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.
606
+ [rank6]:[W830 05:49:50.190942249 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.
607
+ [rank5]:[W830 05:49:50.191275693 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.
608
+ [rank4]:[W830 05:49:50.192393928 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.
609
+ [rank2]:[W830 05:49:50.192725121 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 2] using GPU 2 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
610
+ > compiling dataset index builder ...
611
+ make: Entering directory '/workspace/Megatron-MoE-Echo/megatron/core/datasets'
612
+ make: Nothing to be done for 'default'.
613
+ make: Leaving directory '/workspace/Megatron-MoE-Echo/megatron/core/datasets'
614
+ >>> done with dataset index builder. Compilation time: 0.035 seconds
615
+ WARNING: constraints for invoking optimized fused softmax kernel are not met. We default back to unfused kernel invocations.
616
+ > compiling and loading fused kernels ...
617
+ [rank0]:[W830 05:49:51.350511793 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.
618
+ >>> done with compiling and loading fused kernels. Compilation time: 3.994 seconds
619
+ time to initialize megatron (seconds): 7.215
620
+ [after megatron is initialized] datetime: 2026-08-30 05:49:57
621
+ building GPT model ...
622
+ /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.
623
+ warnings.warn(
624
+ /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.
625
+ warnings.warn(
626
+ /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.
627
+ warnings.warn(
628
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
629
+ warnings.warn(
630
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
631
+ warnings.warn(
632
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
633
+ warnings.warn(
634
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
635
+ warnings.warn(
636
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
637
+ warnings.warn(
638
+ > number of parameters on (tensor, pipeline) model parallel rank (0, 0): 3902410752
639
+ INFO:megatron.core.distributed.distributed_data_parallel:Setting up DistributedDataParallel with config DistributedDataParallelConfig(grad_reduce_in_fp32=True, overlap_grad_reduce=True, overlap_param_gather=True, align_param_gather=False, use_distributed_optimizer=True, num_distributed_optimizer_instances=1, check_for_nan_in_grad=False, check_for_large_grads=False, bucket_size=40000000, pad_buckets_for_high_nccl_busbw=False, average_in_collective=False, fp8_param_gather=False, reuse_grad_buf_for_mxfp8_param_ag=False, use_custom_fsdp=False, data_parallel_sharding_strategy='no_shard', gradient_reduce_div_fusion=True, suggested_communication_unit_size=None, preserve_fp32_weights=True, keep_fp8_transpose_cache_when_using_custom_fsdp=False, nccl_ub=False, fsdp_double_buffer=False)
640
+ INFO:megatron.core.distributed.param_and_grad_buffer:Number of buckets for gradient all-reduce / reduce-scatter: 18
641
+ Params for bucket 1 (206045184 elements, 206045184 padded size):
642
+ module.output_layer.weight
643
+ Params for bucket 2 (41988096 elements, 41988096 padded size):
644
+ module.decoder.final_layernorm.weight
645
+ module.decoder.layers.15.self_attention.linear_qkv.weight
646
+ module.decoder.layers.15.pre_mlp_layernorm.weight
647
+ module.decoder.layers.15.self_attention.linear_qkv.layer_norm_weight
648
+ module.decoder.layers.15.mlp.router.weight
649
+ module.decoder.layers.15.self_attention.linear_proj.weight
650
+ Params for bucket 3 (41984000 elements, 41984000 padded size):
651
+ module.decoder.layers.14.self_attention.linear_proj.weight
652
+ module.decoder.layers.14.self_attention.linear_qkv.layer_norm_weight
653
+ module.decoder.layers.14.mlp.router.weight
654
+ module.decoder.layers.14.self_attention.linear_qkv.weight
655
+ module.decoder.layers.14.pre_mlp_layernorm.weight
656
+ Params for bucket 4 (41984000 elements, 41984000 padded size):
657
+ module.decoder.layers.13.self_attention.linear_qkv.weight
658
+ module.decoder.layers.13.pre_mlp_layernorm.weight
659
+ module.decoder.layers.13.mlp.router.weight
660
+ module.decoder.layers.13.self_attention.linear_proj.weight
661
+ module.decoder.layers.13.self_attention.linear_qkv.layer_norm_weight
662
+ Params for bucket 5 (41984000 elements, 41984000 padded size):
663
+ module.decoder.layers.12.mlp.router.weight
664
+ module.decoder.layers.12.self_attention.linear_proj.weight
665
+ module.decoder.layers.12.self_attention.linear_qkv.layer_norm_weight
666
+ module.decoder.layers.12.pre_mlp_layernorm.weight
667
+ module.decoder.layers.12.self_attention.linear_qkv.weight
668
+ Params for bucket 6 (41984000 elements, 41984000 padded size):
669
+ module.decoder.layers.11.pre_mlp_layernorm.weight
670
+ module.decoder.layers.11.self_attention.linear_qkv.weight
671
+ module.decoder.layers.11.mlp.router.weight
672
+ module.decoder.layers.11.self_attention.linear_qkv.layer_norm_weight
673
+ module.decoder.layers.11.self_attention.linear_proj.weight
674
+ Params for bucket 7 (41984000 elements, 41984000 padded size):
675
+ module.decoder.layers.10.mlp.router.weight
676
+ module.decoder.layers.10.self_attention.linear_qkv.layer_norm_weight
677
+ module.decoder.layers.10.self_attention.linear_proj.weight
678
+ module.decoder.layers.10.self_attention.linear_qkv.weight
679
+ module.decoder.layers.10.pre_mlp_layernorm.weight
680
+ Params for bucket 8 (41984000 elements, 41984000 padded size):
681
+ module.decoder.layers.9.mlp.router.weight
682
+ module.decoder.layers.9.self_attention.linear_proj.weight
683
+ module.decoder.layers.9.self_attention.linear_qkv.layer_norm_weight
684
+ module.decoder.layers.9.self_attention.linear_qkv.weight
685
+ module.decoder.layers.9.pre_mlp_layernorm.weight
686
+ Params for bucket 9 (41984000 elements, 41984000 padded size):
687
+ module.decoder.layers.8.mlp.router.weight
688
+ module.decoder.layers.8.self_attention.linear_proj.weight
689
+ module.decoder.layers.8.self_attention.linear_qkv.layer_norm_weight
690
+ module.decoder.layers.8.pre_mlp_layernorm.weight
691
+ module.decoder.layers.8.self_attention.linear_qkv.weight
692
+ Params for bucket 10 (41984000 elements, 41984000 padded size):
693
+ module.decoder.layers.7.self_attention.linear_qkv.weight
694
+ module.decoder.layers.7.pre_mlp_layernorm.weight
695
+ module.decoder.layers.7.mlp.router.weight
696
+ module.decoder.layers.7.self_attention.linear_qkv.layer_norm_weight
697
+ module.decoder.layers.7.self_attention.linear_proj.weight
698
+ Params for bucket 11 (41984000 elements, 41984000 padded size):
699
+ module.decoder.layers.6.mlp.router.weight
700
+ module.decoder.layers.6.self_attention.linear_proj.weight
701
+ module.decoder.layers.6.self_attention.linear_qkv.layer_norm_weight
702
+ module.decoder.layers.6.pre_mlp_layernorm.weight
703
+ module.decoder.layers.6.self_attention.linear_qkv.weight
704
+ Params for bucket 12 (41984000 elements, 41984000 padded size):
705
+ module.decoder.layers.5.self_attention.linear_qkv.weight
706
+ module.decoder.layers.5.self_attention.linear_proj.weight
707
+ module.decoder.layers.5.pre_mlp_layernorm.weight
708
+ module.decoder.layers.5.mlp.router.weight
709
+ module.decoder.layers.5.self_attention.linear_qkv.layer_norm_weight
710
+ Params for bucket 13 (41984000 elements, 41984000 padded size):
711
+ module.decoder.layers.4.self_attention.linear_qkv.weight
712
+ module.decoder.layers.4.pre_mlp_layernorm.weight
713
+ module.decoder.layers.4.mlp.router.weight
714
+ module.decoder.layers.4.self_attention.linear_proj.weight
715
+ module.decoder.layers.4.self_attention.linear_qkv.layer_norm_weight
716
+ Params for bucket 14 (41984000 elements, 41984000 padded size):
717
+ module.decoder.layers.3.mlp.router.weight
718
+ module.decoder.layers.3.self_attention.linear_qkv.layer_norm_weight
719
+ module.decoder.layers.3.self_attention.linear_proj.weight
720
+ module.decoder.layers.3.self_attention.linear_qkv.weight
721
+ module.decoder.layers.3.pre_mlp_layernorm.weight
722
+ Params for bucket 15 (41984000 elements, 41984000 padded size):
723
+ module.decoder.layers.2.self_attention.linear_proj.weight
724
+ module.decoder.layers.2.self_attention.linear_qkv.layer_norm_weight
725
+ module.decoder.layers.2.pre_mlp_layernorm.weight
726
+ module.decoder.layers.2.mlp.router.weight
727
+ module.decoder.layers.2.self_attention.linear_qkv.weight
728
+ Params for bucket 16 (41984000 elements, 41984000 padded size):
729
+ module.decoder.layers.1.pre_mlp_layernorm.weight
730
+ module.decoder.layers.1.self_attention.linear_qkv.weight
731
+ module.decoder.layers.1.mlp.router.weight
732
+ module.decoder.layers.1.self_attention.linear_qkv.layer_norm_weight
733
+ module.decoder.layers.1.self_attention.linear_proj.weight
734
+ Params for bucket 17 (41984000 elements, 41984000 padded size):
735
+ module.decoder.layers.0.mlp.router.weight
736
+ module.decoder.layers.0.self_attention.linear_qkv.layer_norm_weight
737
+ module.decoder.layers.0.pre_mlp_layernorm.weight
738
+ module.decoder.layers.0.self_attention.linear_qkv.weight
739
+ module.decoder.layers.0.self_attention.linear_proj.weight
740
+ Params for bucket 18 (206045184 elements, 206045184 padded size):
741
+ module.embedding.word_embeddings.weight
742
+ INFO:megatron.core.distributed.param_and_grad_buffer:Number of buckets for gradient all-reduce / reduce-scatter: 32
743
+ Params for bucket 1 (58720256 elements, 58720256 padded size):
744
+ module.decoder.layers.15.mlp.experts.linear_fc2.weight0
745
+ Params for bucket 2 (117440512 elements, 117440512 padded size):
746
+ module.decoder.layers.15.mlp.experts.linear_fc1.weight0
747
+ Params for bucket 3 (58720256 elements, 58720256 padded size):
748
+ module.decoder.layers.14.mlp.experts.linear_fc2.weight0
749
+ Params for bucket 4 (117440512 elements, 117440512 padded size):
750
+ module.decoder.layers.14.mlp.experts.linear_fc1.weight0
751
+ Params for bucket 5 (58720256 elements, 58720256 padded size):
752
+ module.decoder.layers.13.mlp.experts.linear_fc2.weight0
753
+ Params for bucket 6 (117440512 elements, 117440512 padded size):
754
+ module.decoder.layers.13.mlp.experts.linear_fc1.weight0
755
+ Params for bucket 7 (58720256 elements, 58720256 padded size):
756
+ module.decoder.layers.12.mlp.experts.linear_fc2.weight0
757
+ Params for bucket 8 (117440512 elements, 117440512 padded size):
758
+ module.decoder.layers.12.mlp.experts.linear_fc1.weight0
759
+ Params for bucket 9 (58720256 elements, 58720256 padded size):
760
+ module.decoder.layers.11.mlp.experts.linear_fc2.weight0
761
+ Params for bucket 10 (117440512 elements, 117440512 padded size):
762
+ module.decoder.layers.11.mlp.experts.linear_fc1.weight0
763
+ Params for bucket 11 (58720256 elements, 58720256 padded size):
764
+ module.decoder.layers.10.mlp.experts.linear_fc2.weight0
765
+ Params for bucket 12 (117440512 elements, 117440512 padded size):
766
+ module.decoder.layers.10.mlp.experts.linear_fc1.weight0
767
+ Params for bucket 13 (58720256 elements, 58720256 padded size):
768
+ module.decoder.layers.9.mlp.experts.linear_fc2.weight0
769
+ Params for bucket 14 (117440512 elements, 117440512 padded size):
770
+ module.decoder.layers.9.mlp.experts.linear_fc1.weight0
771
+ Params for bucket 15 (58720256 elements, 58720256 padded size):
772
+ module.decoder.layers.8.mlp.experts.linear_fc2.weight0
773
+ Params for bucket 16 (117440512 elements, 117440512 padded size):
774
+ module.decoder.layers.8.mlp.experts.linear_fc1.weight0
775
+ Params for bucket 17 (58720256 elements, 58720256 padded size):
776
+ module.decoder.layers.7.mlp.experts.linear_fc2.weight0
777
+ Params for bucket 18 (117440512 elements, 117440512 padded size):
778
+ module.decoder.layers.7.mlp.experts.linear_fc1.weight0
779
+ Params for bucket 19 (58720256 elements, 58720256 padded size):
780
+ module.decoder.layers.6.mlp.experts.linear_fc2.weight0
781
+ Params for bucket 20 (117440512 elements, 117440512 padded size):
782
+ module.decoder.layers.6.mlp.experts.linear_fc1.weight0
783
+ Params for bucket 21 (58720256 elements, 58720256 padded size):
784
+ module.decoder.layers.5.mlp.experts.linear_fc2.weight0
785
+ Params for bucket 22 (117440512 elements, 117440512 padded size):
786
+ module.decoder.layers.5.mlp.experts.linear_fc1.weight0
787
+ Params for bucket 23 (58720256 elements, 58720256 padded size):
788
+ module.decoder.layers.4.mlp.experts.linear_fc2.weight0
789
+ Params for bucket 24 (117440512 elements, 117440512 padded size):
790
+ module.decoder.layers.4.mlp.experts.linear_fc1.weight0
791
+ Params for bucket 25 (58720256 elements, 58720256 padded size):
792
+ module.decoder.layers.3.mlp.experts.linear_fc2.weight0
793
+ Params for bucket 26 (117440512 elements, 117440512 padded size):
794
+ module.decoder.layers.3.mlp.experts.linear_fc1.weight0
795
+ Params for bucket 27 (58720256 elements, 58720256 padded size):
796
+ module.decoder.layers.2.mlp.experts.linear_fc2.weight0
797
+ Params for bucket 28 (117440512 elements, 117440512 padded size):
798
+ module.decoder.layers.2.mlp.experts.linear_fc1.weight0
799
+ Params for bucket 29 (58720256 elements, 58720256 padded size):
800
+ module.decoder.layers.1.mlp.experts.linear_fc2.weight0
801
+ Params for bucket 30 (117440512 elements, 117440512 padded size):
802
+ module.decoder.layers.1.mlp.experts.linear_fc1.weight0
803
+ Params for bucket 31 (58720256 elements, 58720256 padded size):
804
+ module.decoder.layers.0.mlp.experts.linear_fc2.weight0
805
+ Params for bucket 32 (117440512 elements, 117440512 padded size):
806
+ module.decoder.layers.0.mlp.experts.linear_fc1.weight0
807
+ 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 0x754d07450350>, config_logger_dir='')
808
+ INFO:megatron.core.optimizer_param_scheduler:> learning rate decay style: cosine
809
+ [after model, optimizer, and learning rate scheduler are built] datetime: 2026-08-30 05:49:57
810
+ > building train, validation, and test datasets ...
811
+ > datasets target sizes (minimum size):
812
+ train: 192
813
+ validation: 0
814
+ test: 0
815
+ INFO:megatron.core.datasets.blended_megatron_dataset_config:Let split_matrix = [(0, 1.0), None, None]
816
+ > building train, validation, and test datasets for GPT ...
817
+ 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 0x754d07533470>, 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)
818
+ INFO:megatron.core.datasets.indexed_dataset:Load the _IndexReader from /workspace/data/wikitext/wikitext103_gpt2_text_document.idx
819
+ INFO:megatron.core.datasets.indexed_dataset: Extract the sequence lengths
820
+ INFO:megatron.core.datasets.indexed_dataset: Extract the sequence pointers
821
+ INFO:megatron.core.datasets.indexed_dataset: Extract the document indices
822
+ INFO:megatron.core.datasets.indexed_dataset:> total number of sequences: 1165029
823
+ INFO:megatron.core.datasets.indexed_dataset:> total number of documents: 1801350
824
+ INFO:megatron.core.datasets.gpt_dataset:Load the GPTDataset train indices
825
+ INFO:megatron.core.datasets.gpt_dataset: Load the document index from 44ed1ee5bcb9e5b030387b8d27a6c406-GPTDataset-train-document_index.npy
826
+ INFO:megatron.core.datasets.gpt_dataset: Load the sample index from 44ed1ee5bcb9e5b030387b8d27a6c406-GPTDataset-train-sample_index.npy
827
+ INFO:megatron.core.datasets.gpt_dataset: Load the shuffle index from 44ed1ee5bcb9e5b030387b8d27a6c406-GPTDataset-train-shuffle_index.npy
828
+ INFO:megatron.core.datasets.gpt_dataset:> total number of samples: 29073
829
+ > finished creating GPT datasets ...
830
+ [after dataloaders are built] datetime: 2026-08-30 05:49:57
831
+ done with setup ...
832
+ (min, max) time across ranks (ms):
833
+ model-and-optimizer-setup ......................: (200.23, 201.72)
834
+ train/valid/test-data-iterators-setup ..........: (60.83, 65.92)
835
+ training ...
836
+ Overwriting rerun_state_machine.current_iteration from -1 to 0...
837
+ [before the start of training step] datetime: 2026-08-30 05:49:57
838
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
839
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
840
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
841
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
842
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
843
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
844
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
845
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
846
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
847
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
848
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
849
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
850
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
851
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
852
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
853
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
854
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
855
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
856
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
857
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
858
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
859
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
860
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
861
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
862
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
863
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
864
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
865
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
866
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
867
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
868
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
869
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
870
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
871
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
872
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
873
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
874
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
875
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
876
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
877
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
878
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
879
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
880
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
881
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
882
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
883
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
884
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
885
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
886
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
887
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
888
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
889
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
890
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
891
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
892
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
893
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
894
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
895
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
896
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
897
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
898
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
899
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
900
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
901
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
902
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
903
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
904
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
905
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
906
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
907
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
908
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
909
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
910
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
911
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
912
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
913
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
914
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
915
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
916
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
917
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
918
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
919
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
920
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
921
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
922
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
923
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
924
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
925
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
926
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
927
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
928
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
929
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
930
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
931
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
932
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
933
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
934
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
935
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
936
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
937
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
938
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
939
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
940
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
941
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
942
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
943
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
944
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
945
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
946
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
947
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
948
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
949
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
950
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
951
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
952
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
953
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
954
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
955
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
956
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
957
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
958
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
959
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
960
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
961
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
962
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
963
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
964
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
965
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
966
+ [2026-08-30 05:50:35] iteration 1/ 12 | consumed samples: 16 | elapsed time per iteration (ms): 37651.8 | throughput per GPU (TFLOP/s/GPU): 8.9 | learning rate: 2.000000E-05 | global batch size: 16 | lm loss: 1.104376E+01 | loss scale: 1.0 | grad norm: 34.296 | number of skipped iterations: 0 | number of nan iterations: 0 |
967
+ Number of parameters in transformer block in billions: 23.22
968
+ Number of parameters in embedding layers in billions: 0.41
969
+ Total number of parameters in billions: 23.63
970
+ Number of parameters in most loaded shard in billions: 23.6320
971
+ Theoretical memory footprints: weight and optimizer=169029.43 MB
972
+ [Rank 0] (after 1 iterations) memory (MB) | allocated: 63795.404296875 | max allocated: 96519.67578125 | reserved: 99122.0 | max reserved: 99122.0
973
+ [2026-08-30 05:50:35] iteration 2/ 12 | consumed samples: 32 | elapsed time per iteration (ms): 588.2 | throughput per GPU (TFLOP/s/GPU): 566.8 | learning rate: 4.000000E-05 | global batch size: 16 | lm loss: 1.104935E+01 | loss scale: 1.0 | grad norm: 35.988 | number of skipped iterations: 0 | number of nan iterations: 0 |
974
+ [2026-08-30 05:50:36] iteration 3/ 12 | consumed samples: 48 | elapsed time per iteration (ms): 564.6 | throughput per GPU (TFLOP/s/GPU): 590.5 | learning rate: 6.000000E-05 | global batch size: 16 | lm loss: 9.658228E+00 | loss scale: 1.0 | grad norm: 40.253 | number of skipped iterations: 0 | number of nan iterations: 0 |
975
+ Capture CUDA graph for training!!!
976
+ CUDA graph capture done!!!
977
+ [2026-08-30 05:50:37] iteration 4/ 12 | consumed samples: 64 | elapsed time per iteration (ms): 1674.8 | throughput per GPU (TFLOP/s/GPU): 199.1 | learning rate: 8.000000E-05 | global batch size: 16 | lm loss: 1.082813E+01 | loss scale: 1.0 | grad norm: 92.941 | number of skipped iterations: 0 | number of nan iterations: 0 |
978
+ [2026-08-30 05:50:38] iteration 5/ 12 | consumed samples: 80 | elapsed time per iteration (ms): 467.9 | throughput per GPU (TFLOP/s/GPU): 712.5 | learning rate: 1.000000E-04 | global batch size: 16 | lm loss: 1.067199E+01 | loss scale: 1.0 | grad norm: 16.861 | number of skipped iterations: 0 | number of nan iterations: 0 |
979
+ [2026-08-30 05:50:38] iteration 6/ 12 | consumed samples: 96 | elapsed time per iteration (ms): 466.4 | throughput per GPU (TFLOP/s/GPU): 714.7 | learning rate: 7.750000E-05 | global batch size: 16 | lm loss: 9.873405E+00 | loss scale: 1.0 | grad norm: 40.570 | number of skipped iterations: 0 | number of nan iterations: 0 |
980
+ [2026-08-30 05:50:39] iteration 7/ 12 | consumed samples: 112 | elapsed time per iteration (ms): 469.0 | throughput per GPU (TFLOP/s/GPU): 710.9 | learning rate: 3.250000E-05 | global batch size: 16 | lm loss: 9.268401E+00 | loss scale: 1.0 | grad norm: 7.021 | number of skipped iterations: 0 | number of nan iterations: 0 |
981
+ [2026-08-30 05:50:39] iteration 8/ 12 | consumed samples: 128 | elapsed time per iteration (ms): 468.7 | throughput per GPU (TFLOP/s/GPU): 711.3 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.952490E+00 | loss scale: 1.0 | grad norm: 21.911 | number of skipped iterations: 0 | number of nan iterations: 0 |
982
+ [2026-08-30 05:50:40] iteration 9/ 12 | consumed samples: 144 | elapsed time per iteration (ms): 468.0 | throughput per GPU (TFLOP/s/GPU): 712.3 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.689455E+00 | loss scale: 1.0 | grad norm: 6.281 | number of skipped iterations: 0 | number of nan iterations: 0 |
983
+ [2026-08-30 05:50:40] iteration 10/ 12 | consumed samples: 160 | elapsed time per iteration (ms): 463.5 | throughput per GPU (TFLOP/s/GPU): 719.2 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.558211E+00 | loss scale: 1.0 | grad norm: 5.656 | number of skipped iterations: 0 | number of nan iterations: 0 |
984
+ [2026-08-30 05:50:41] iteration 11/ 12 | consumed samples: 176 | elapsed time per iteration (ms): 466.3 | throughput per GPU (TFLOP/s/GPU): 715.0 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.422763E+00 | loss scale: 1.0 | grad norm: 4.727 | number of skipped iterations: 0 | number of nan iterations: 0 |
985
+ [2026-08-30 05:50:41] iteration 12/ 12 | consumed samples: 192 | elapsed time per iteration (ms): 469.2 | throughput per GPU (TFLOP/s/GPU): 710.5 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.353562E+00 | loss scale: 1.0 | grad norm: 6.615 | number of skipped iterations: 0 | number of nan iterations: 0 |
986
+ [after training is done] datetime: 2026-08-30 05:50:41
987
+ W0830 05:50:42.041000 545455 torch/distributed/elastic/agent/server/api.py:719] Received 15 death signal, shutting down workers
988
+ W0830 05:50:42.042000 545455 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 545544 closing signal SIGTERM
989
+ W0830 05:50:42.043000 545455 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 545545 closing signal SIGTERM
990
+ W0830 05:50:42.044000 545455 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 545546 closing signal SIGTERM
991
+ W0830 05:50:42.046000 545455 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 545547 closing signal SIGTERM
992
+ W0830 05:50:42.047000 545455 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 545548 closing signal SIGTERM
993
+ W0830 05:50:42.048000 545455 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 545549 closing signal SIGTERM
994
+ W0830 05:50:42.049000 545455 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 545550 closing signal SIGTERM
995
+ W0830 05:50:42.050000 545455 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 545551 closing signal SIGTERM
996
+ Traceback (most recent call last):
997
+ File "/usr/local/bin/torchrun", line 8, in <module>
998
+ sys.exit(main())
999
+ ^^^^^^
1000
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/multiprocessing/errors/__init__.py", line 355, in wrapper
1001
+ return f(*args, **kwargs)
1002
+ ^^^^^^^^^^^^^^^^^^
1003
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/run.py", line 892, in main
1004
+ run(args)
1005
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/run.py", line 883, in run
1006
+ elastic_launch(
1007
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/launcher/api.py", line 139, in __call__
1008
+ return launch_agent(self._config, self._entrypoint, list(args))
1009
+ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1010
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/launcher/api.py", line 261, in launch_agent
1011
+ result = agent.run()
1012
+ ^^^^^^^^^^^
1013
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/metrics/api.py", line 138, in wrapper
1014
+ result = f(*args, **kwargs)
1015
+ ^^^^^^^^^^^^^^^^^^
1016
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/agent/server/api.py", line 711, in run
1017
+ result = self._invoke_run(role)
1018
+ ^^^^^^^^^^^^^^^^^^^^^^
1019
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/agent/server/api.py", line 870, in _invoke_run
1020
+ time.sleep(monitor_interval)
1021
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/multiprocessing/api.py", line 84, in _terminate_process_handler
1022
+ raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval)
1023
+ torch.distributed.elastic.multiprocessing.api.SignalException: Process 545455 got signal: 15
tiera-rerun-fixed-moonep/echo/a1/a1_oneslot_gbs16/G1__echo__one__seq4096__gbs16__cap2p40625/rep2/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, 629
43
+ 2, 629
44
+ 3, 629
45
+ 4, 629
46
+ 5, 629
47
+ 6, 629
48
+ 7, 629
49
+ 0, 931
50
+ 1, 929
51
+ 2, 931
52
+ 3, 929
53
+ 4, 937
54
+ 5, 937
55
+ 6, 931
56
+ 7, 935
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, 39479
66
+ 1, 39485
67
+ 2, 39489
68
+ 3, 39479
69
+ 4, 39501
70
+ 5, 39489
71
+ 6, 39509
72
+ 7, 39491
73
+ 0, 39797
74
+ 1, 39813
75
+ 2, 39779
76
+ 3, 39809
77
+ 4, 39803
78
+ 5, 39811
79
+ 6, 39801
80
+ 7, 39799
81
+ 0, 42357
82
+ 1, 42355
83
+ 2, 42355
84
+ 3, 42355
85
+ 4, 42355
86
+ 5, 42355
87
+ 6, 42355
88
+ 7, 42355
89
+ 0, 42583
90
+ 1, 42581
91
+ 2, 42581
92
+ 3, 42581
93
+ 4, 42581
94
+ 5, 42581
95
+ 6, 42581
96
+ 7, 42581
97
+ 0, 42705
98
+ 1, 42727
99
+ 2, 42723
100
+ 3, 42719
101
+ 4, 42743
102
+ 5, 42717
103
+ 6, 42717
104
+ 7, 42733
105
+ 0, 43493
106
+ 1, 43491
107
+ 2, 43491
108
+ 3, 43491
109
+ 4, 43491
110
+ 5, 43491
111
+ 6, 43491
112
+ 7, 43491
113
+ 0, 43493
114
+ 1, 43491
115
+ 2, 43491
116
+ 3, 43491
117
+ 4, 43491
118
+ 5, 43491
119
+ 6, 43491
120
+ 7, 43491
121
+ 0, 45265
122
+ 1, 45263
123
+ 2, 45263
124
+ 3, 45263
125
+ 4, 45263
126
+ 5, 45263
127
+ 6, 45263
128
+ 7, 45263
129
+ 0, 45265
130
+ 1, 45263
131
+ 2, 45263
132
+ 3, 45263
133
+ 4, 45263
134
+ 5, 45263
135
+ 6, 45263
136
+ 7, 45263
137
+ 0, 85827
138
+ 1, 85825
139
+ 2, 85825
140
+ 3, 85825
141
+ 4, 85825
142
+ 5, 85825
143
+ 6, 85825
144
+ 7, 85825
145
+ 0, 86905
146
+ 1, 86903
147
+ 2, 86903
148
+ 3, 86903
149
+ 4, 86903
150
+ 5, 86903
151
+ 6, 86903
152
+ 7, 86903
153
+ 0, 86905
154
+ 1, 86903
155
+ 2, 86903
156
+ 3, 86903
157
+ 4, 86903
158
+ 5, 86903
159
+ 6, 86903
160
+ 7, 86903
161
+ 0, 86905
162
+ 1, 86903
163
+ 2, 86903
164
+ 3, 86903
165
+ 4, 86903
166
+ 5, 86903
167
+ 6, 86903
168
+ 7, 86903
169
+ 0, 104019
170
+ 1, 103961
171
+ 2, 103961
172
+ 3, 103961
173
+ 4, 104017
174
+ 5, 103961
175
+ 6, 103961
176
+ 7, 103961
177
+ 0, 105857
178
+ 1, 105731
179
+ 2, 105757
180
+ 3, 105745
181
+ 4, 105821
182
+ 5, 105807
183
+ 6, 105815
184
+ 7, 105759
185
+ 0, 106523
186
+ 1, 106465
187
+ 2, 106465
188
+ 3, 106465
189
+ 4, 106521
190
+ 5, 106465
191
+ 6, 106465
192
+ 7, 106465
193
+ 0, 107143
194
+ 1, 107059
195
+ 2, 107109
196
+ 3, 107079
197
+ 4, 107169
198
+ 5, 107117
199
+ 6, 107091
200
+ 7, 107157
201
+ 0, 117961
202
+ 1, 133559
203
+ 2, 136051
204
+ 3, 120391
205
+ 4, 120161
206
+ 5, 117851
207
+ 6, 123213
208
+ 7, 117849
209
+ 0, 161639
210
+ 1, 161529
211
+ 2, 161529
212
+ 3, 161525
213
+ 4, 161723
214
+ 5, 161527
215
+ 6, 161525
216
+ 7, 161525
217
+ 0, 161639
218
+ 1, 161529
219
+ 2, 161529
220
+ 3, 161525
221
+ 4, 161723
222
+ 5, 161527
223
+ 6, 161525
224
+ 7, 161525
225
+ 0, 2095
226
+ 1, 2098
227
+ 2, 2098
228
+ 3, 2418
229
+ 4, 161723
230
+ 5, 3698
231
+ 6, 2098
232
+ 7, 2098
tiera-rerun-fixed-moonep/echo/a1/a1_oneslot_gbs16/G1__echo__one__seq4096__gbs16__cap2p40625/rep2/run.env ADDED
@@ -0,0 +1,70 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ backend=echo
2
+ config=G1
3
+ family=tierA
4
+ repo=/workspace/Megatron-MoE-Echo
5
+ branch=bracing-regions-through-kernels
6
+ commit=9a236793
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=one
42
+ slots_per_rank=1
43
+ capacity_mode=static
44
+ capacity=2.40625
45
+ cuda_graph=1
46
+ probe_mode=0
47
+ async_grad_reduce=0
48
+ branch_role=timing
49
+ memory_dump_mode=none
50
+ memory_snapshot_path=
51
+ cuda_graph_warmup_steps=3
52
+ recompute=none
53
+ force_balance=0
54
+ train_iters=12
55
+ ep_size=8
56
+ echo_dispatch_overlap=1
57
+ echo_dispatch_recompute=0
58
+ ultraep_quota_min_tokens=
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/echo/a1/a1_oneslot_gbs16/G1__echo__one__seq4096__gbs16__cap2p40625/rep2/train.log ADDED
@@ -0,0 +1,1023 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ==============================================================
2
+ G1 / echo (slots/rank=1 policy=one)
3
+ E=8 K=2 H'=14336 m=1
4
+ L=16 seq=4096 gbs=16 (2 microbatches)
5
+ cap_mode=static cap=2.40625 cg=1 recompute=none
6
+ memory_dump=none snapshots=/workspace/tiera-rerun-fixed-moonep/echo/a1/a1_oneslot_gbs16/G1__echo__one__seq4096__gbs16__cap2p40625/rep2/memory_snapshots/snapshot.pickle
7
+ repo=/workspace/Megatron-MoE-Echo @ 9a236793
8
+ ==============================================================
9
+ W0830 05:50:48.539000 548836 torch/distributed/run.py:766]
10
+ W0830 05:50:48.539000 548836 torch/distributed/run.py:766] *****************************************
11
+ W0830 05:50:48.539000 548836 torch/distributed/run.py:766] Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed.
12
+ W0830 05:50:48.539000 548836 torch/distributed/run.py:766] *****************************************
13
+ /workspace/Megatron-MoE-Echo/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
14
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
15
+ /workspace/Megatron-MoE-Echo/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
16
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
17
+ /workspace/Megatron-MoE-Echo/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
18
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
19
+ /workspace/Megatron-MoE-Echo/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
20
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
21
+ /workspace/Megatron-MoE-Echo/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
22
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
23
+ /workspace/Megatron-MoE-Echo/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
24
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
25
+ /workspace/Megatron-MoE-Echo/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
26
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
27
+ 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
28
+ 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
29
+ WARNING: overriding default arguments for tokenizer_type:GPT2BPETokenizer with tokenizer_type:GPT2BPETokenizer
30
+ Number of virtual stages per pipeline stage: None
31
+ accumulate and all-reduce gradients in fp32 for bfloat16 data type.
32
+ using torch.bfloat16 for parameters ...
33
+ /workspace/Megatron-MoE-Echo/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
34
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
35
+ ------------------------ arguments ------------------------
36
+ account_for_embedding_in_pipeline_split ......... False
37
+ account_for_loss_in_pipeline_split .............. False
38
+ accumulate_allreduce_grads_in_fp32 .............. True
39
+ adam_beta1 ...................................... 0.9
40
+ adam_beta2 ...................................... 0.999
41
+ adam_eps ........................................ 1e-08
42
+ add_bias_linear ................................. False
43
+ add_position_embedding .......................... False
44
+ add_qkv_bias .................................... False
45
+ adlr_autoresume ................................. False
46
+ adlr_autoresume_interval ........................ 1000
47
+ align_grad_reduce ............................... True
48
+ align_param_gather .............................. False
49
+ app_tag_run_name ................................ None
50
+ app_tag_run_version ............................. 0.0.0
51
+ apply_layernorm_1p .............................. False
52
+ apply_query_key_layer_scaling ................... False
53
+ apply_residual_connection_post_layernorm ........ False
54
+ apply_rope_fusion ............................... True
55
+ async_save ...................................... None
56
+ async_tensor_model_parallel_allreduce ........... True
57
+ attention_backend ............................... AttnBackend.auto
58
+ attention_dropout ............................... 0.0
59
+ attention_softmax_in_fp32 ....................... False
60
+ auto_detect_ckpt_format ......................... False
61
+ barrier_with_L1_time ............................ True
62
+ bert_binary_head ................................ True
63
+ bert_embedder_type .............................. megatron
64
+ bert_load ....................................... None
65
+ bf16 ............................................ True
66
+ bias_dropout_fusion ............................. True
67
+ bias_gelu_fusion ................................ False
68
+ bias_swiglu_fusion .............................. True
69
+ biencoder_projection_dim ........................ 0
70
+ biencoder_shared_query_context_model ............ False
71
+ block_data_path ................................. None
72
+ calc_ft_timeouts ................................ False
73
+ calculate_per_token_loss ........................ False
74
+ check_for_large_grads ........................... False
75
+ check_for_nan_in_loss_and_grad .................. False
76
+ check_for_spiky_loss ............................ False
77
+ check_weight_hash_across_dp_replicas_interval ... None
78
+ ckpt_assume_constant_structure .................. False
79
+ ckpt_convert_format ............................. None
80
+ ckpt_convert_save ............................... None
81
+ ckpt_convert_update_legacy_dist_opt_format ...... False
82
+ ckpt_format ..................................... torch_dist
83
+ ckpt_fully_parallel_load ........................ False
84
+ ckpt_fully_parallel_save ........................ True
85
+ ckpt_fully_parallel_save_deprecated ............. False
86
+ ckpt_step ....................................... None
87
+ classes_fraction ................................ 1.0
88
+ clip_grad ....................................... 1.0
89
+ clone_scatter_output_in_embedding ............... True
90
+ config_logger_dir ...............................
91
+ consumed_train_samples .......................... 0
92
+ consumed_valid_samples .......................... 0
93
+ context_parallel_size ........................... 1
94
+ cp_comm_type .................................... ['p2p']
95
+ create_attention_mask_in_dataloader ............. True
96
+ cross_entropy_fusion_impl ....................... native
97
+ cross_entropy_loss_fusion ....................... False
98
+ cuda_graph_scope ................................ full_iteration
99
+ cuda_graph_warmup_steps ......................... 3
100
+ data_args_path .................................. None
101
+ data_cache_path ................................. None
102
+ data_parallel_random_init ....................... False
103
+ data_parallel_sharding_strategy ................. no_shard
104
+ data_parallel_size .............................. 8
105
+ data_path ....................................... ['/workspace/data/wikitext/wikitext103_gpt2_text_document']
106
+ data_per_class_fraction ......................... 1.0
107
+ data_sharding ................................... True
108
+ dataloader_type ................................. single
109
+ ddp_average_in_collective ....................... False
110
+ ddp_bucket_size ................................. None
111
+ ddp_num_buckets ................................. None
112
+ ddp_pad_buckets_for_high_nccl_busbw ............. False
113
+ decoder_first_pipeline_num_layers ............... None
114
+ decoder_last_pipeline_num_layers ................ None
115
+ decoder_num_layers .............................. None
116
+ decoder_seq_length .............................. None
117
+ decoupled_lr .................................... None
118
+ decoupled_min_lr ................................ None
119
+ decrease_batch_size_if_needed ................... False
120
+ defer_embedding_wgrad_compute ................... False
121
+ delay_wgrad_compute ............................. False
122
+ deprecated_use_mcore_models ..................... True
123
+ deterministic_mode .............................. False
124
+ dino_bottleneck_size ............................ 256
125
+ dino_freeze_last_layer .......................... 1
126
+ dino_head_hidden_size ........................... 2048
127
+ dino_local_crops_number ......................... 10
128
+ dino_local_img_size ............................. 96
129
+ dino_norm_last_layer ............................ False
130
+ dino_teacher_temp ............................... 0.07
131
+ dino_warmup_teacher_temp ........................ 0.04
132
+ dino_warmup_teacher_temp_epochs ................. 30
133
+ disable_bf16_reduced_precision_matmul ........... False
134
+ disable_mamba_mem_eff_path ...................... False
135
+ disable_straggler_on_startup .................... False
136
+ dist_ckpt_format_deprecated ..................... None
137
+ dist_ckpt_strictness ............................ assume_ok_unexpected
138
+ distribute_saved_activations .................... False
139
+ distributed_backend ............................. nccl
140
+ distributed_timeout_minutes ..................... 10
141
+ embedding_init_method_std ....................... None
142
+ embedding_path .................................. None
143
+ empty_unused_memory_level ....................... 0
144
+ enable_cuda_graph ............................... True
145
+ enable_experimental ............................. False
146
+ enable_ft_package ............................... False
147
+ enable_gloo_process_groups ...................... True
148
+ enable_msc ...................................... True
149
+ enable_one_logger ............................... True
150
+ encoder_num_layers .............................. 16
151
+ encoder_seq_length .............................. 4096
152
+ end_weight_decay ................................ 0.1
153
+ eod_mask_loss ................................... False
154
+ error_injection_rate ............................ 0
155
+ error_injection_type ............................ transient_error
156
+ eval_interval ................................... 1000000
157
+ eval_iters ...................................... 0
158
+ evidence_data_path .............................. None
159
+ exit_duration_in_mins ........................... None
160
+ exit_interval ................................... None
161
+ exit_on_missing_checkpoint ...................... False
162
+ exit_signal_handler ............................. False
163
+ exp_avg_dtype ................................... torch.float32
164
+ exp_avg_sq_dtype ................................ torch.float32
165
+ expert_model_parallel_size ...................... 8
166
+ expert_tensor_parallel_size ..................... 1
167
+ export_force_local_attention .................... False
168
+ export_kd_cfg ................................... None
169
+ export_kd_teacher_ckpt_format ................... None
170
+ export_kd_teacher_load .......................... None
171
+ export_kv_cache_quant ........................... False
172
+ export_legacy_megatron .......................... False
173
+ export_model_type ............................... GPTModel
174
+ export_moe_apply_probs_on_input ................. False
175
+ export_qk_l2_norm ............................... False
176
+ export_quant_cfg ................................ None
177
+ export_real_quant_cfg ........................... None
178
+ export_te_mcore_model ........................... False
179
+ external_cuda_graph ............................. False
180
+ ffn_hidden_size ................................. 14336
181
+ finetune ........................................ False
182
+ finetune_data_split ............................. train
183
+ finetune_hf_dataset ............................. None
184
+ first_last_layers_bf16 .......................... False
185
+ flash_decode .................................... False
186
+ fp16 ............................................ False
187
+ fp16_lm_cross_entropy ........................... False
188
+ fp32_residual_connection ........................ False
189
+ fp8 ............................................. e4m3
190
+ fp8_amax_compute_algo ........................... most_recent
191
+ fp8_amax_history_len ............................ 1
192
+ fp8_interval .................................... 1
193
+ fp8_margin ...................................... 0
194
+ fp8_param_gather ................................ False
195
+ fp8_recipe ...................................... mxfp8
196
+ fp8_wgrad ....................................... True
197
+ fsdp_double_buffer .............................. False
198
+ global_batch_size ............................... 16
199
+ grad_reduce_in_bf16 ............................. False
200
+ gradient_accumulation_fusion .................... True
201
+ gradient_reduce_div_fusion ...................... True
202
+ group_query_attention ........................... True
203
+ head_lr_mult .................................... 1.0
204
+ heterogeneous_layers_config_encoded_json ........ None
205
+ heterogeneous_layers_config_path ................ None
206
+ hidden_dropout .................................. 0.0
207
+ hidden_size ..................................... 4096
208
+ hierarchical_context_parallel_sizes ............. None
209
+ high_priority_stream_groups ..................... []
210
+ hybrid_attention_ratio .......................... 0.0
211
+ hybrid_mlp_ratio ................................ 0.0
212
+ hybrid_override_pattern ......................... None
213
+ hysteresis ...................................... 2
214
+ ict_head_size ................................... None
215
+ ict_load ........................................ None
216
+ img_h ........................................... 224
217
+ img_w ........................................... 224
218
+ indexer_batch_size .............................. 128
219
+ indexer_log_interval ............................ 1000
220
+ inference_batch_times_seqlen_threshold .......... -1
221
+ inference_dynamic_batching ...................... False
222
+ inference_dynamic_batching_buffer_guaranteed_fraction 0.2
223
+ inference_dynamic_batching_buffer_overflow_factor None
224
+ inference_dynamic_batching_buffer_size_gb ....... 40.0
225
+ inference_dynamic_batching_chunk_size ........... 256
226
+ inference_dynamic_batching_max_requests_override None
227
+ inference_dynamic_batching_max_tokens_override .. None
228
+ inference_dynamic_batching_num_cuda_graphs ...... 16
229
+ inference_max_batch_size ........................ 8
230
+ inference_max_seq_length ........................ 2560
231
+ inference_rng_tracker ........................... False
232
+ init_method_std ................................. 0.01
233
+ init_method_xavier_uniform ...................... False
234
+ init_model_with_meta_device ..................... False
235
+ initial_loss_scale .............................. 4294967296
236
+ inprocess_active_world_size ..................... 8
237
+ inprocess_barrier_timeout ....................... 120
238
+ inprocess_completion_timeout .................... 120
239
+ inprocess_empty_cuda_cache ...................... False
240
+ inprocess_granularity ........................... node
241
+ inprocess_hard_timeout .......................... 90
242
+ inprocess_heartbeat_interval .................... 30
243
+ inprocess_heartbeat_timeout ..................... 60
244
+ inprocess_last_call_wait ........................ 1
245
+ inprocess_max_iterations ........................ None
246
+ inprocess_monitor_process_interval .............. 1.0
247
+ inprocess_monitor_thread_interval ............... 1.0
248
+ inprocess_progress_watchdog_interval ............ 1.0
249
+ inprocess_restart ............................... False
250
+ inprocess_soft_timeout .......................... 60
251
+ inprocess_termination_grace_time ................ 1
252
+ is_hybrid_model ................................. False
253
+ iter_per_epoch .................................. 1250
254
+ iterations_to_skip .............................. []
255
+ keep_fp8_transpose_cache_when_using_custom_fsdp . False
256
+ kitchen_config_file ............................. None
257
+ kitchen_recipe_number ........................... None
258
+ kv_channels ..................................... 128
259
+ kv_lora_rank .................................... 32
260
+ lazy_mpu_init ................................... None
261
+ load ............................................ None
262
+ load_main_params_from_ckpt ...................... None
263
+ load_model_opt_format ........................... False
264
+ local_rank ...................................... 0
265
+ log_energy ...................................... False
266
+ log_interval .................................... 1
267
+ log_loss_scale_to_tensorboard ................... True
268
+ log_memory_to_tensorboard ....................... False
269
+ log_num_zeros_in_grad ........................... False
270
+ log_params_norm ................................. False
271
+ log_progress .................................... False
272
+ log_straggler ................................... False
273
+ log_throughput .................................. True
274
+ log_timers_to_tensorboard ....................... False
275
+ log_validation_ppl_to_tensorboard ............... False
276
+ log_world_size_to_tensorboard ................... False
277
+ logging_level ................................... None
278
+ loss_scale ...................................... None
279
+ loss_scale_window ............................... 1000
280
+ lr .............................................. 0.0001
281
+ lr_decay_iters .................................. 8
282
+ lr_decay_samples ................................ None
283
+ lr_decay_style .................................. cosine
284
+ lr_warmup_fraction .............................. None
285
+ lr_warmup_init .................................. 0.0
286
+ lr_warmup_iters ................................. 5
287
+ lr_warmup_samples ............................... 0
288
+ lr_wsd_decay_iters .............................. None
289
+ lr_wsd_decay_samples ............................ None
290
+ lr_wsd_decay_style .............................. exponential
291
+ main_grads_dtype ................................ torch.float32
292
+ main_params_dtype ............................... torch.float32
293
+ make_vocab_size_divisible_by .................... 128
294
+ mamba_head_dim .................................. 64
295
+ mamba_num_groups ................................ 8
296
+ mamba_num_heads ................................. None
297
+ mamba_state_dim ................................. 128
298
+ manual_gc ....................................... False
299
+ manual_gc_eval .................................. True
300
+ manual_gc_interval .............................. 0
301
+ mask_factor ..................................... 1.0
302
+ mask_prob ....................................... 0.15
303
+ mask_type ....................................... random
304
+ masked_softmax_fusion ........................... False
305
+ max_position_embeddings ......................... 32768
306
+ max_tokens_to_oom ............................... 12000
307
+ memory_snapshot_path ............................ snapshot.pickle
308
+ merge_file ...................................... /workspace/data/wikitext/gpt2-merges.txt
309
+ micro_batch_size ................................ 1
310
+ microbatch_group_size_per_vp_stage .............. None
311
+ mid_level_dataset_surplus ....................... 0.005
312
+ min_loss_scale .................................. 1.0
313
+ min_lr .......................................... 1e-05
314
+ mlp_chunks_for_prefill .......................... 1
315
+ mmap_bin_files .................................. True
316
+ mock_data ....................................... False
317
+ moe_apply_probs_on_input ........................ False
318
+ moe_aux_loss_coeff .............................. 0.01
319
+ moe_deepep_num_sms .............................. 20
320
+ moe_echo_enable_random_offloading ............... False
321
+ moe_echo_expert_dispatch_overlap ................ True
322
+ moe_echo_expert_dispatcher_type ................. hybridep
323
+ moe_echo_recompute_expert_dispatch .............. False
324
+ moe_enable_deepep ............................... False
325
+ moe_enable_echo ................................. True
326
+ moe_enable_hybridep ............................. True
327
+ moe_expert_capacity_factor ...................... None
328
+ moe_extended_tp ................................. False
329
+ moe_ffn_hidden_size ............................. 14336
330
+ moe_grouped_gemm ................................ True
331
+ moe_hybridep_num_sms ............................ 16
332
+ moe_input_jitter_eps ............................ None
333
+ moe_layer_freq .................................. 1
334
+ moe_layer_recompute ............................. False
335
+ moe_num_echo_experts ............................ 8
336
+ moe_pad_expert_input_to_capacity ................ False
337
+ moe_per_layer_logging ........................... False
338
+ moe_permute_fusion .............................. False
339
+ moe_received_token_capacity ..................... 2.40625
340
+ moe_router_bias_update_rate ..................... 0.001
341
+ moe_router_dtype ................................ fp32
342
+ moe_router_enable_expert_bias ................... False
343
+ moe_router_force_load_balancing ................. False
344
+ moe_router_group_topk ........................... None
345
+ moe_router_load_balancing_type .................. aux_loss
346
+ moe_router_num_groups ........................... None
347
+ moe_router_padding_for_fp8 ...................... False
348
+ moe_router_pre_softmax .......................... False
349
+ moe_router_score_function ....................... softmax
350
+ moe_router_topk ................................. 2
351
+ moe_router_topk_scaling_factor .................. None
352
+ moe_shared_expert_intermediate_size ............. None
353
+ moe_shared_expert_overlap ....................... False
354
+ moe_token_dispatcher_type ....................... flex
355
+ moe_token_drop_policy ........................... probs
356
+ moe_upcycling_granularity ....................... 1
357
+ moe_use_device_initiated_grouped_gemm ........... True
358
+ moe_use_legacy_grouped_gemm ..................... False
359
+ moe_use_upcycling ............................... False
360
+ moe_z_loss_coeff ................................ None
361
+ mrope_section ................................... None
362
+ mscale .......................................... 1.0
363
+ mscale_all_dim .................................. 0.0
364
+ mtp_loss_scaling_factor ......................... 0.1
365
+ mtp_num_layers .................................. None
366
+ multi_latent_attention .......................... False
367
+ nccl_all_reduce_for_prefill ..................... False
368
+ nccl_communicator_config_path ................... None
369
+ nccl_ub ......................................... False
370
+ no_load_optim ................................... None
371
+ no_load_rng ..................................... None
372
+ no_persist_layer_norm ........................... False
373
+ no_rope_freq .................................... None
374
+ no_save_optim ................................... None
375
+ no_save_rng ..................................... None
376
+ non_persistent_ckpt_type ........................ None
377
+ non_persistent_global_ckpt_dir .................. None
378
+ non_persistent_local_ckpt_algo .................. fully_parallel
379
+ non_persistent_local_ckpt_dir ................... None
380
+ non_persistent_save_interval .................... None
381
+ norm_epsilon .................................... 1e-05
382
+ normalization ................................... RMSNorm
383
+ num_attention_heads ............................. 32
384
+ num_channels .................................... 3
385
+ num_classes ..................................... 1000
386
+ num_dataset_builder_threads ..................... 1
387
+ num_distributed_optimizer_instances ............. 1
388
+ num_experts ..................................... 8
389
+ num_layers ...................................... 16
390
+ num_layers_at_end_in_bf16 ....................... 1
391
+ num_layers_at_start_in_bf16 ..................... 1
392
+ num_layers_per_virtual_pipeline_stage ........... None
393
+ num_query_groups ................................ 8
394
+ num_virtual_stages_per_pipeline_rank ............ None
395
+ num_workers ..................................... 2
396
+ object_storage_cache_path ....................... None
397
+ one_logger_async ................................ False
398
+ one_logger_project .............................. megatron-lm
399
+ one_logger_run_name ............................. None
400
+ onnx_safe ....................................... None
401
+ openai_gelu ..................................... False
402
+ optimizer ....................................... adam
403
+ optimizer_cpu_offload ........................... False
404
+ optimizer_offload_fraction ...................... 1.0
405
+ output_bert_embeddings .......................... False
406
+ overlap_cpu_optimizer_d2h_h2d ................... False
407
+ overlap_grad_reduce ............................. True
408
+ overlap_moe_expert_parallel_comm ................ False
409
+ overlap_p2p_comm ................................ False
410
+ overlap_p2p_comm_warmup_flush ................... False
411
+ overlap_param_gather ............................ True
412
+ overlap_param_gather_with_optimizer_step ........ False
413
+ override_opt_param_scheduler .................... False
414
+ params_dtype .................................... torch.bfloat16
415
+ patch_dim ....................................... 16
416
+ per_split_data_args_path ........................ None
417
+ perform_initialization .......................... True
418
+ pin_cpu_grads ................................... True
419
+ pin_cpu_params .................................. True
420
+ pipeline_model_parallel_comm_backend ............ None
421
+ pipeline_model_parallel_layout .................. None
422
+ pipeline_model_parallel_size .................... 1
423
+ position_embedding_type ......................... rope
424
+ pretrained_checkpoint ........................... None
425
+ profile ......................................... False
426
+ profile_ranks ................................... [0]
427
+ profile_step_end ................................ 12
428
+ profile_step_start .............................. 10
429
+ q_lora_rank ..................................... None
430
+ qk_head_dim ..................................... 128
431
+ qk_l2_norm ...................................... False
432
+ qk_layernorm .................................... False
433
+ qk_pos_emb_head_dim ............................. 64
434
+ query_in_block_prob ............................. 0.1
435
+ rampup_batch_size ............................... None
436
+ rank ............................................ 0
437
+ recompute_granularity ........................... None
438
+ recompute_method ................................ None
439
+ recompute_modules ............................... None
440
+ recompute_num_layers ............................ None
441
+ record_memory_history ........................... False
442
+ relative_attention_max_distance ................. 128
443
+ relative_attention_num_buckets .................. 32
444
+ replication ..................................... False
445
+ replication_factor .............................. 2
446
+ replication_jump ................................ None
447
+ rerun_mode ...................................... validate_results
448
+ reset_attention_mask ............................ False
449
+ reset_position_ids .............................. False
450
+ result_rejected_tracker_filename ................ None
451
+ retriever_report_topk_accuracies ................ []
452
+ retriever_score_scaling ......................... False
453
+ retriever_seq_length ............................ 256
454
+ retro_add_retriever ............................. False
455
+ retro_attention_gate ............................ 1
456
+ retro_cyclic_train_iters ........................ None
457
+ retro_encoder_attention_dropout ................. 0.1
458
+ retro_encoder_hidden_dropout .................... 0.1
459
+ retro_encoder_layers ............................ 2
460
+ retro_num_neighbors ............................. 2
461
+ retro_num_retrieved_chunks ...................... 2
462
+ retro_project_dir ............................... None
463
+ retro_verify_neighbor_count ..................... True
464
+ reuse_grad_buf_for_mxfp8_param_ag ............... False
465
+ rope_scaling_factor ............................. 8.0
466
+ rope_type ....................................... None
467
+ rotary_base ..................................... 1000000
468
+ rotary_interleaved .............................. False
469
+ rotary_percent .................................. 1.0
470
+ rotary_scaling_factor ........................... 1.0
471
+ rotary_seq_len_interpolation_factor ............. None
472
+ run_workload_inspector_server ................... False
473
+ sample_rate ..................................... 1.0
474
+ save ............................................ None
475
+ save_interval ................................... None
476
+ save_retain_interval ............................ None
477
+ scatter_gather_tensors_in_pipeline .............. True
478
+ seed ............................................ 1234
479
+ seq_length ...................................... 4096
480
+ sequence_parallel ............................... False
481
+ sft ............................................. False
482
+ sft_tokenizer_prompt_format ..................... nemotron-h-aligned
483
+ sgd_momentum .................................... 0.9
484
+ short_seq_prob .................................. 0.1
485
+ skip_train ...................................... False
486
+ skipped_train_samples ........................... 0
487
+ spec ............................................ None
488
+ split ........................................... 100,0,0
489
+ squared_relu .................................... False
490
+ start_weight_decay .............................. 0.1
491
+ straggler_ctrlr_port ............................ 65535
492
+ straggler_minmax_count .......................... 1
493
+ suggested_communication_unit_size ............... None
494
+ swiglu .......................................... True
495
+ swin_backbone_type .............................. tiny
496
+ symmetric_ar_type ............................... None
497
+ te_rng_tracker .................................. True
498
+ tensor_model_parallel_size ...................... 1
499
+ tensorboard_dir ................................. None
500
+ tensorboard_log_interval ........................ 1
501
+ tensorboard_queue_size .......................... 1000
502
+ test_data_path .................................. None
503
+ test_mode ....................................... False
504
+ tiktoken_num_special_tokens ..................... 1000
505
+ tiktoken_pattern ................................ None
506
+ tiktoken_special_tokens ......................... None
507
+ timing_log_level ................................ 0
508
+ timing_log_option ............................... minmax
509
+ titles_data_path ................................ None
510
+ tokenizer_model ................................. None
511
+ tokenizer_type .................................. GPT2BPETokenizer
512
+ torch_fsdp2_reshard_after_forward ............... True
513
+ tp_comm_bootstrap_backend ....................... nccl
514
+ tp_comm_bulk_dgrad .............................. True
515
+ tp_comm_bulk_wgrad .............................. True
516
+ tp_comm_overlap ................................. False
517
+ tp_comm_overlap_ag .............................. True
518
+ tp_comm_overlap_cfg ............................. None
519
+ tp_comm_overlap_rs .............................. True
520
+ tp_comm_overlap_rs_dgrad ........................ False
521
+ tp_comm_split_ag ................................ True
522
+ tp_comm_split_rs ................................ True
523
+ train_data_path ................................. None
524
+ train_iters ..................................... 12
525
+ train_samples ................................... None
526
+ train_sync_interval ............................. None
527
+ transformer_impl ................................ transformer_engine
528
+ transformer_pipeline_model_parallel_size ........ 1
529
+ untie_embeddings_and_output_weights ............. True
530
+ use_checkpoint_args ............................. False
531
+ use_checkpoint_opt_param_scheduler .............. False
532
+ use_cpu_initialization .......................... None
533
+ use_custom_fsdp ................................. False
534
+ use_dist_ckpt ................................... True
535
+ use_dist_ckpt_deprecated ........................ False
536
+ use_distributed_optimizer ....................... True
537
+ use_flash_attn .................................. False
538
+ use_legacy_models ............................... False
539
+ use_mp_args_from_checkpoint_args ................ False
540
+ use_one_sent_docs ............................... False
541
+ use_persistent_ckpt_worker ...................... False
542
+ use_precision_aware_optimizer ................... False
543
+ use_pytorch_profiler ............................ False
544
+ use_ring_exchange_p2p ........................... False
545
+ use_rope_scaling ................................ False
546
+ use_rotary_position_embeddings .................. False
547
+ use_sharp ....................................... False
548
+ use_tokenizer_model_from_checkpoint_args ........ True
549
+ use_torch_fsdp2 ................................. False
550
+ use_torch_optimizer_for_cpu_offload ............. False
551
+ use_tp_pp_dp_mapping ............................ False
552
+ v_head_dim ...................................... 128
553
+ valid_data_path ................................. None
554
+ variable_seq_lengths ............................ False
555
+ virtual_pipeline_model_parallel_size ............ None
556
+ vision_backbone_type ............................ vit
557
+ vision_pretraining .............................. False
558
+ vision_pretraining_type ......................... classify
559
+ vocab_extra_ids ................................. 0
560
+ vocab_file ...................................... /workspace/data/wikitext/gpt2-vocab.json
561
+ vocab_size ...................................... None
562
+ wandb_exp_name ..................................
563
+ wandb_project ...................................
564
+ wandb_save_dir ..................................
565
+ weight_decay .................................... 0.1
566
+ weight_decay_incr_style ......................... constant
567
+ wgrad_deferral_limit ............................ 0
568
+ world_size ...................................... 8
569
+ yaml_cfg ........................................ None
570
+ -------------------- end of arguments ---------------------
571
+ INFO:megatron.core.num_microbatches_calculator:setting number of microbatches to constant 2
572
+ > building GPT2BPETokenizer tokenizer ...
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
+ [Gloo] Rank 0 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
577
+ [Gloo] Rank 1 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
578
+ [Gloo] Rank 7 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
579
+ [Gloo] Rank 2 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
580
+ [Gloo] Rank [Gloo] Rank [Gloo] Rank [Gloo] Rank 4356 is connected to is connected to is connected to is connected to 7777 peer ranks. peer ranks. peer ranks. peer ranks. Expected number of connected peer ranks is : Expected number of connected peer ranks is : Expected number of connected peer ranks is : Expected number of connected peer ranks is : 7777
581
+
582
+
583
+
584
+ [Gloo] Rank [Gloo] Rank 0 is connected to 37 is connected to peer ranks. 7Expected number of connected peer ranks is : peer ranks. 7Expected number of connected peer ranks is :
585
+ 7
586
+ [Gloo] Rank 1 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
587
+ [Gloo] Rank 4 is connected to [Gloo] Rank 7[Gloo] Rank 2 peer ranks. 6 is connected to Expected number of connected peer ranks is : is connected to 777 peer ranks.
588
+ peer ranks. [Gloo] Rank Expected number of connected peer ranks is : Expected number of connected peer ranks is : 757
589
+ is connected to
590
+ 7 peer ranks. Expected number of connected peer ranks is : 7
591
+ [Gloo] Rank 7 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
592
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
593
+ > initialized tensor model parallel with size 1
594
+ > initialized pipeline model parallel with size 1
595
+ > setting random seeds to 1234 ...
596
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
597
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
598
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
599
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
600
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
601
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
602
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
603
+ [rank1]:[W830 05:51:06.545967647 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.
604
+ [rank5]:[W830 05:51:06.547930287 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.
605
+ [rank6]:[W830 05:51:06.547957777 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.
606
+ [rank2]:[W830 05:51:06.547997406 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.
607
+ [rank3]:[W830 05:51:06.548013686 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.
608
+ [rank4]:[W830 05:51:06.548049882 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.
609
+ [rank7]:[W830 05:51:06.549166171 ProcessGroupNCCL.cpp:4789] [PG ID 0 PG GUID 0 Rank 7] using GPU 7 as device used by this process is currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. You can specify device_id in init_process_group() to force use of a particular device.
610
+ > compiling dataset index builder ...
611
+ make: Entering directory '/workspace/Megatron-MoE-Echo/megatron/core/datasets'
612
+ make: Nothing to be done for 'default'.
613
+ make: Leaving directory '/workspace/Megatron-MoE-Echo/megatron/core/datasets'
614
+ >>> done with dataset index builder. Compilation time: 0.035 seconds
615
+ WARNING: constraints for invoking optimized fused softmax kernel are not met. We default back to unfused kernel invocations.
616
+ > compiling and loading fused kernels ...
617
+ [rank0]:[W830 05:51:06.707500621 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.
618
+ >>> done with compiling and loading fused kernels. Compilation time: 3.782 seconds
619
+ time to initialize megatron (seconds): 7.074
620
+ [after megatron is initialized] datetime: 2026-08-30 05:51:12
621
+ building GPT model ...
622
+ /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.
623
+ warnings.warn(
624
+ /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.
625
+ warnings.warn(
626
+ /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.
627
+ warnings.warn(
628
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
629
+ warnings.warn(
630
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
631
+ warnings.warn(
632
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
633
+ warnings.warn(
634
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
635
+ warnings.warn(
636
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
637
+ warnings.warn(
638
+ > number of parameters on (tensor, pipeline) model parallel rank (0, 0): 3902410752
639
+ INFO:megatron.core.distributed.distributed_data_parallel:Setting up DistributedDataParallel with config DistributedDataParallelConfig(grad_reduce_in_fp32=True, overlap_grad_reduce=True, overlap_param_gather=True, align_param_gather=False, use_distributed_optimizer=True, num_distributed_optimizer_instances=1, check_for_nan_in_grad=False, check_for_large_grads=False, bucket_size=40000000, pad_buckets_for_high_nccl_busbw=False, average_in_collective=False, fp8_param_gather=False, reuse_grad_buf_for_mxfp8_param_ag=False, use_custom_fsdp=False, data_parallel_sharding_strategy='no_shard', gradient_reduce_div_fusion=True, suggested_communication_unit_size=None, preserve_fp32_weights=True, keep_fp8_transpose_cache_when_using_custom_fsdp=False, nccl_ub=False, fsdp_double_buffer=False)
640
+ INFO:megatron.core.distributed.param_and_grad_buffer:Number of buckets for gradient all-reduce / reduce-scatter: 18
641
+ Params for bucket 1 (206045184 elements, 206045184 padded size):
642
+ module.output_layer.weight
643
+ Params for bucket 2 (41988096 elements, 41988096 padded size):
644
+ module.decoder.layers.15.self_attention.linear_qkv.layer_norm_weight
645
+ module.decoder.layers.15.mlp.router.weight
646
+ module.decoder.layers.15.self_attention.linear_proj.weight
647
+ module.decoder.final_layernorm.weight
648
+ module.decoder.layers.15.pre_mlp_layernorm.weight
649
+ module.decoder.layers.15.self_attention.linear_qkv.weight
650
+ Params for bucket 3 (41984000 elements, 41984000 padded size):
651
+ module.decoder.layers.14.self_attention.linear_qkv.weight
652
+ module.decoder.layers.14.pre_mlp_layernorm.weight
653
+ module.decoder.layers.14.mlp.router.weight
654
+ module.decoder.layers.14.self_attention.linear_proj.weight
655
+ module.decoder.layers.14.self_attention.linear_qkv.layer_norm_weight
656
+ Params for bucket 4 (41984000 elements, 41984000 padded size):
657
+ module.decoder.layers.13.mlp.router.weight
658
+ module.decoder.layers.13.self_attention.linear_proj.weight
659
+ module.decoder.layers.13.self_attention.linear_qkv.layer_norm_weight
660
+ module.decoder.layers.13.self_attention.linear_qkv.weight
661
+ module.decoder.layers.13.pre_mlp_layernorm.weight
662
+ Params for bucket 5 (41984000 elements, 41984000 padded size):
663
+ module.decoder.layers.12.self_attention.linear_qkv.weight
664
+ module.decoder.layers.12.pre_mlp_layernorm.weight
665
+ module.decoder.layers.12.mlp.router.weight
666
+ module.decoder.layers.12.self_attention.linear_proj.weight
667
+ module.decoder.layers.12.self_attention.linear_qkv.layer_norm_weight
668
+ Params for bucket 6 (41984000 elements, 41984000 padded size):
669
+ module.decoder.layers.11.mlp.router.weight
670
+ module.decoder.layers.11.self_attention.linear_proj.weight
671
+ module.decoder.layers.11.self_attention.linear_qkv.layer_norm_weight
672
+ module.decoder.layers.11.self_attention.linear_qkv.weight
673
+ module.decoder.layers.11.pre_mlp_layernorm.weight
674
+ Params for bucket 7 (41984000 elements, 41984000 padded size):
675
+ module.decoder.layers.10.pre_mlp_layernorm.weight
676
+ module.decoder.layers.10.self_attention.linear_qkv.weight
677
+ module.decoder.layers.10.mlp.router.weight
678
+ module.decoder.layers.10.self_attention.linear_proj.weight
679
+ module.decoder.layers.10.self_attention.linear_qkv.layer_norm_weight
680
+ Params for bucket 8 (41984000 elements, 41984000 padded size):
681
+ module.decoder.layers.9.mlp.router.weight
682
+ module.decoder.layers.9.self_attention.linear_qkv.layer_norm_weight
683
+ module.decoder.layers.9.self_attention.linear_proj.weight
684
+ module.decoder.layers.9.self_attention.linear_qkv.weight
685
+ module.decoder.layers.9.pre_mlp_layernorm.weight
686
+ Params for bucket 9 (41984000 elements, 41984000 padded size):
687
+ module.decoder.layers.8.self_attention.linear_proj.weight
688
+ module.decoder.layers.8.self_attention.linear_qkv.layer_norm_weight
689
+ module.decoder.layers.8.mlp.router.weight
690
+ module.decoder.layers.8.self_attention.linear_qkv.weight
691
+ module.decoder.layers.8.pre_mlp_layernorm.weight
692
+ Params for bucket 10 (41984000 elements, 41984000 padded size):
693
+ module.decoder.layers.7.mlp.router.weight
694
+ module.decoder.layers.7.self_attention.linear_proj.weight
695
+ module.decoder.layers.7.self_attention.linear_qkv.layer_norm_weight
696
+ module.decoder.layers.7.self_attention.linear_qkv.weight
697
+ module.decoder.layers.7.pre_mlp_layernorm.weight
698
+ Params for bucket 11 (41984000 elements, 41984000 padded size):
699
+ module.decoder.layers.6.pre_mlp_layernorm.weight
700
+ module.decoder.layers.6.self_attention.linear_qkv.weight
701
+ module.decoder.layers.6.mlp.router.weight
702
+ module.decoder.layers.6.self_attention.linear_proj.weight
703
+ module.decoder.layers.6.self_attention.linear_qkv.layer_norm_weight
704
+ Params for bucket 12 (41984000 elements, 41984000 padded size):
705
+ module.decoder.layers.5.mlp.router.weight
706
+ module.decoder.layers.5.self_attention.linear_qkv.layer_norm_weight
707
+ module.decoder.layers.5.self_attention.linear_qkv.weight
708
+ module.decoder.layers.5.self_attention.linear_proj.weight
709
+ module.decoder.layers.5.pre_mlp_layernorm.weight
710
+ Params for bucket 13 (41984000 elements, 41984000 padded size):
711
+ module.decoder.layers.4.mlp.router.weight
712
+ module.decoder.layers.4.self_attention.linear_proj.weight
713
+ module.decoder.layers.4.self_attention.linear_qkv.layer_norm_weight
714
+ module.decoder.layers.4.pre_mlp_layernorm.weight
715
+ module.decoder.layers.4.self_attention.linear_qkv.weight
716
+ Params for bucket 14 (41984000 elements, 41984000 padded size):
717
+ module.decoder.layers.3.self_attention.linear_qkv.weight
718
+ module.decoder.layers.3.pre_mlp_layernorm.weight
719
+ module.decoder.layers.3.mlp.router.weight
720
+ module.decoder.layers.3.self_attention.linear_qkv.layer_norm_weight
721
+ module.decoder.layers.3.self_attention.linear_proj.weight
722
+ Params for bucket 15 (41984000 elements, 41984000 padded size):
723
+ module.decoder.layers.2.mlp.router.weight
724
+ module.decoder.layers.2.self_attention.linear_qkv.layer_norm_weight
725
+ module.decoder.layers.2.self_attention.linear_qkv.weight
726
+ module.decoder.layers.2.self_attention.linear_proj.weight
727
+ module.decoder.layers.2.pre_mlp_layernorm.weight
728
+ Params for bucket 16 (41984000 elements, 41984000 padded size):
729
+ module.decoder.layers.1.self_attention.linear_qkv.layer_norm_weight
730
+ module.decoder.layers.1.self_attention.linear_proj.weight
731
+ module.decoder.layers.1.mlp.router.weight
732
+ module.decoder.layers.1.pre_mlp_layernorm.weight
733
+ module.decoder.layers.1.self_attention.linear_qkv.weight
734
+ Params for bucket 17 (41984000 elements, 41984000 padded size):
735
+ module.decoder.layers.0.self_attention.linear_qkv.weight
736
+ module.decoder.layers.0.self_attention.linear_proj.weight
737
+ module.decoder.layers.0.pre_mlp_layernorm.weight
738
+ module.decoder.layers.0.mlp.router.weight
739
+ module.decoder.layers.0.self_attention.linear_qkv.layer_norm_weight
740
+ Params for bucket 18 (206045184 elements, 206045184 padded size):
741
+ module.embedding.word_embeddings.weight
742
+ INFO:megatron.core.distributed.param_and_grad_buffer:Number of buckets for gradient all-reduce / reduce-scatter: 32
743
+ Params for bucket 1 (58720256 elements, 58720256 padded size):
744
+ module.decoder.layers.15.mlp.experts.linear_fc2.weight0
745
+ Params for bucket 2 (117440512 elements, 117440512 padded size):
746
+ module.decoder.layers.15.mlp.experts.linear_fc1.weight0
747
+ Params for bucket 3 (58720256 elements, 58720256 padded size):
748
+ module.decoder.layers.14.mlp.experts.linear_fc2.weight0
749
+ Params for bucket 4 (117440512 elements, 117440512 padded size):
750
+ module.decoder.layers.14.mlp.experts.linear_fc1.weight0
751
+ Params for bucket 5 (58720256 elements, 58720256 padded size):
752
+ module.decoder.layers.13.mlp.experts.linear_fc2.weight0
753
+ Params for bucket 6 (117440512 elements, 117440512 padded size):
754
+ module.decoder.layers.13.mlp.experts.linear_fc1.weight0
755
+ Params for bucket 7 (58720256 elements, 58720256 padded size):
756
+ module.decoder.layers.12.mlp.experts.linear_fc2.weight0
757
+ Params for bucket 8 (117440512 elements, 117440512 padded size):
758
+ module.decoder.layers.12.mlp.experts.linear_fc1.weight0
759
+ Params for bucket 9 (58720256 elements, 58720256 padded size):
760
+ module.decoder.layers.11.mlp.experts.linear_fc2.weight0
761
+ Params for bucket 10 (117440512 elements, 117440512 padded size):
762
+ module.decoder.layers.11.mlp.experts.linear_fc1.weight0
763
+ Params for bucket 11 (58720256 elements, 58720256 padded size):
764
+ module.decoder.layers.10.mlp.experts.linear_fc2.weight0
765
+ Params for bucket 12 (117440512 elements, 117440512 padded size):
766
+ module.decoder.layers.10.mlp.experts.linear_fc1.weight0
767
+ Params for bucket 13 (58720256 elements, 58720256 padded size):
768
+ module.decoder.layers.9.mlp.experts.linear_fc2.weight0
769
+ Params for bucket 14 (117440512 elements, 117440512 padded size):
770
+ module.decoder.layers.9.mlp.experts.linear_fc1.weight0
771
+ Params for bucket 15 (58720256 elements, 58720256 padded size):
772
+ module.decoder.layers.8.mlp.experts.linear_fc2.weight0
773
+ Params for bucket 16 (117440512 elements, 117440512 padded size):
774
+ module.decoder.layers.8.mlp.experts.linear_fc1.weight0
775
+ Params for bucket 17 (58720256 elements, 58720256 padded size):
776
+ module.decoder.layers.7.mlp.experts.linear_fc2.weight0
777
+ Params for bucket 18 (117440512 elements, 117440512 padded size):
778
+ module.decoder.layers.7.mlp.experts.linear_fc1.weight0
779
+ Params for bucket 19 (58720256 elements, 58720256 padded size):
780
+ module.decoder.layers.6.mlp.experts.linear_fc2.weight0
781
+ Params for bucket 20 (117440512 elements, 117440512 padded size):
782
+ module.decoder.layers.6.mlp.experts.linear_fc1.weight0
783
+ Params for bucket 21 (58720256 elements, 58720256 padded size):
784
+ module.decoder.layers.5.mlp.experts.linear_fc2.weight0
785
+ Params for bucket 22 (117440512 elements, 117440512 padded size):
786
+ module.decoder.layers.5.mlp.experts.linear_fc1.weight0
787
+ Params for bucket 23 (58720256 elements, 58720256 padded size):
788
+ module.decoder.layers.4.mlp.experts.linear_fc2.weight0
789
+ Params for bucket 24 (117440512 elements, 117440512 padded size):
790
+ module.decoder.layers.4.mlp.experts.linear_fc1.weight0
791
+ Params for bucket 25 (58720256 elements, 58720256 padded size):
792
+ module.decoder.layers.3.mlp.experts.linear_fc2.weight0
793
+ Params for bucket 26 (117440512 elements, 117440512 padded size):
794
+ module.decoder.layers.3.mlp.experts.linear_fc1.weight0
795
+ Params for bucket 27 (58720256 elements, 58720256 padded size):
796
+ module.decoder.layers.2.mlp.experts.linear_fc2.weight0
797
+ Params for bucket 28 (117440512 elements, 117440512 padded size):
798
+ module.decoder.layers.2.mlp.experts.linear_fc1.weight0
799
+ Params for bucket 29 (58720256 elements, 58720256 padded size):
800
+ module.decoder.layers.1.mlp.experts.linear_fc2.weight0
801
+ Params for bucket 30 (117440512 elements, 117440512 padded size):
802
+ module.decoder.layers.1.mlp.experts.linear_fc1.weight0
803
+ Params for bucket 31 (58720256 elements, 58720256 padded size):
804
+ module.decoder.layers.0.mlp.experts.linear_fc2.weight0
805
+ Params for bucket 32 (117440512 elements, 117440512 padded size):
806
+ module.decoder.layers.0.mlp.experts.linear_fc1.weight0
807
+ 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 0x746ce8a64440>, config_logger_dir='')
808
+ INFO:megatron.core.optimizer_param_scheduler:> learning rate decay style: cosine
809
+ [after model, optimizer, and learning rate scheduler are built] datetime: 2026-08-30 05:51:12
810
+ > building train, validation, and test datasets ...
811
+ > datasets target sizes (minimum size):
812
+ train: 192
813
+ validation: 0
814
+ test: 0
815
+ INFO:megatron.core.datasets.blended_megatron_dataset_config:Let split_matrix = [(0, 1.0), None, None]
816
+ > building train, validation, and test datasets for GPT ...
817
+ 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 0x746cebcef7d0>, 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)
818
+ INFO:megatron.core.datasets.indexed_dataset:Load the _IndexReader from /workspace/data/wikitext/wikitext103_gpt2_text_document.idx
819
+ INFO:megatron.core.datasets.indexed_dataset: Extract the sequence lengths
820
+ INFO:megatron.core.datasets.indexed_dataset: Extract the sequence pointers
821
+ INFO:megatron.core.datasets.indexed_dataset: Extract the document indices
822
+ INFO:megatron.core.datasets.indexed_dataset:> total number of sequences: 1165029
823
+ INFO:megatron.core.datasets.indexed_dataset:> total number of documents: 1801350
824
+ INFO:megatron.core.datasets.gpt_dataset:Load the GPTDataset train indices
825
+ INFO:megatron.core.datasets.gpt_dataset: Load the document index from 44ed1ee5bcb9e5b030387b8d27a6c406-GPTDataset-train-document_index.npy
826
+ INFO:megatron.core.datasets.gpt_dataset: Load the sample index from 44ed1ee5bcb9e5b030387b8d27a6c406-GPTDataset-train-sample_index.npy
827
+ INFO:megatron.core.datasets.gpt_dataset: Load the shuffle index from 44ed1ee5bcb9e5b030387b8d27a6c406-GPTDataset-train-shuffle_index.npy
828
+ INFO:megatron.core.datasets.gpt_dataset:> total number of samples: 29073
829
+ > finished creating GPT datasets ...
830
+ [after dataloaders are built] datetime: 2026-08-30 05:51:12
831
+ done with setup ...
832
+ training ...
833
+ (min, max) time across ranks (ms):
834
+ model-and-optimizer-setup ......................: (200.71, 204.16)
835
+ train/valid/test-data-iterators-setup ..........: (63.25, 73.75)
836
+ Overwriting rerun_state_machine.current_iteration from -1 to 0...
837
+ [before the start of training step] datetime: 2026-08-30 05:51:12
838
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
839
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
840
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
841
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
842
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
843
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
844
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
845
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
846
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
847
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
848
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
849
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
850
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
851
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
852
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
853
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
854
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
855
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
856
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
857
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
858
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
859
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
860
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
861
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
862
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
863
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
864
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
865
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
866
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
867
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
868
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
869
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
870
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
871
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
872
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
873
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
874
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
875
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
876
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
877
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
878
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
879
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
880
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
881
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
882
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
883
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
884
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
885
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
886
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
887
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
888
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
889
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
890
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
891
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
892
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
893
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
894
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
895
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
896
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
897
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
898
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
899
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
900
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
901
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
902
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
903
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
904
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
905
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
906
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
907
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
908
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
909
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
910
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
911
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
912
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
913
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
914
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
915
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
916
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
917
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
918
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
919
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
920
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
921
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
922
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
923
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
924
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
925
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
926
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
927
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
928
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
929
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
930
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
931
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
932
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
933
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
934
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
935
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
936
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
937
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
938
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
939
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
940
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
941
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
942
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
943
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
944
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
945
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
946
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
947
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
948
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
949
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
950
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
951
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
952
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
953
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
954
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
955
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
956
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
957
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
958
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
959
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
960
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
961
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
962
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
963
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
964
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
965
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
966
+ [2026-08-30 05:51:50] iteration 1/ 12 | consumed samples: 16 | elapsed time per iteration (ms): 37807.3 | throughput per GPU (TFLOP/s/GPU): 8.8 | learning rate: 2.000000E-05 | global batch size: 16 | lm loss: 1.104376E+01 | loss scale: 1.0 | grad norm: 34.296 | number of skipped iterations: 0 | number of nan iterations: 0 |
967
+ Number of parameters in transformer block in billions: 23.22
968
+ Number of parameters in embedding layers in billions: 0.41
969
+ Total number of parameters in billions: 23.63
970
+ Number of parameters in most loaded shard in billions: 23.6320
971
+ Theoretical memory footprints: weight and optimizer=169029.43 MB
972
+ [Rank 0] (after 1 iterations) memory (MB) | allocated: 63795.404296875 | max allocated: 96519.67578125 | reserved: 99122.0 | max reserved: 99122.0
973
+ [2026-08-30 05:51:50] iteration 2/ 12 | consumed samples: 32 | elapsed time per iteration (ms): 567.3 | throughput per GPU (TFLOP/s/GPU): 587.7 | learning rate: 4.000000E-05 | global batch size: 16 | lm loss: 1.104935E+01 | loss scale: 1.0 | grad norm: 35.986 | number of skipped iterations: 0 | number of nan iterations: 0 |
974
+ [2026-08-30 05:51:51] iteration 3/ 12 | consumed samples: 48 | elapsed time per iteration (ms): 557.4 | throughput per GPU (TFLOP/s/GPU): 598.1 | learning rate: 6.000000E-05 | global batch size: 16 | lm loss: 9.658128E+00 | loss scale: 1.0 | grad norm: 40.421 | number of skipped iterations: 0 | number of nan iterations: 0 |
975
+ Capture CUDA graph for training!!!
976
+ CUDA graph capture done!!!
977
+ [2026-08-30 05:51:53] iteration 4/ 12 | consumed samples: 64 | elapsed time per iteration (ms): 1893.1 | throughput per GPU (TFLOP/s/GPU): 176.1 | learning rate: 8.000000E-05 | global batch size: 16 | lm loss: 1.080487E+01 | loss scale: 1.0 | grad norm: 93.349 | number of skipped iterations: 0 | number of nan iterations: 0 |
978
+ [2026-08-30 05:51:53] iteration 5/ 12 | consumed samples: 80 | elapsed time per iteration (ms): 470.3 | throughput per GPU (TFLOP/s/GPU): 708.8 | learning rate: 1.000000E-04 | global batch size: 16 | lm loss: 1.069694E+01 | loss scale: 1.0 | grad norm: 17.487 | number of skipped iterations: 0 | number of nan iterations: 0 |
979
+ [2026-08-30 05:51:54] iteration 6/ 12 | consumed samples: 96 | elapsed time per iteration (ms): 471.5 | throughput per GPU (TFLOP/s/GPU): 707.1 | learning rate: 7.750000E-05 | global batch size: 16 | lm loss: 9.973028E+00 | loss scale: 1.0 | grad norm: 44.239 | number of skipped iterations: 0 | number of nan iterations: 0 |
980
+ [2026-08-30 05:51:54] iteration 7/ 12 | consumed samples: 112 | elapsed time per iteration (ms): 466.1 | throughput per GPU (TFLOP/s/GPU): 715.3 | learning rate: 3.250000E-05 | global batch size: 16 | lm loss: 9.324042E+00 | loss scale: 1.0 | grad norm: 7.313 | number of skipped iterations: 0 | number of nan iterations: 0 |
981
+ [2026-08-30 05:51:55] iteration 8/ 12 | consumed samples: 128 | elapsed time per iteration (ms): 466.9 | throughput per GPU (TFLOP/s/GPU): 714.0 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 9.213482E+00 | loss scale: 1.0 | grad norm: 29.834 | number of skipped iterations: 0 | number of nan iterations: 0 |
982
+ [2026-08-30 05:51:55] iteration 9/ 12 | consumed samples: 144 | elapsed time per iteration (ms): 468.4 | throughput per GPU (TFLOP/s/GPU): 711.8 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.742799E+00 | loss scale: 1.0 | grad norm: 6.456 | number of skipped iterations: 0 | number of nan iterations: 0 |
983
+ [2026-08-30 05:51:56] iteration 10/ 12 | consumed samples: 160 | elapsed time per iteration (ms): 460.6 | throughput per GPU (TFLOP/s/GPU): 723.9 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.629816E+00 | loss scale: 1.0 | grad norm: 5.511 | number of skipped iterations: 0 | number of nan iterations: 0 |
984
+ [2026-08-30 05:51:56] iteration 11/ 12 | consumed samples: 176 | elapsed time per iteration (ms): 466.0 | throughput per GPU (TFLOP/s/GPU): 715.5 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.528298E+00 | loss scale: 1.0 | grad norm: 5.274 | number of skipped iterations: 0 | number of nan iterations: 0 |
985
+ [2026-08-30 05:51:57] iteration 12/ 12 | consumed samples: 192 | elapsed time per iteration (ms): 467.2 | throughput per GPU (TFLOP/s/GPU): 713.6 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.440600E+00 | loss scale: 1.0 | grad norm: 4.965 | number of skipped iterations: 0 | number of nan iterations: 0 |
986
+ [after training is done] datetime: 2026-08-30 05:51:57
987
+ W0830 05:51:57.251000 548836 torch/distributed/elastic/agent/server/api.py:719] Received 15 death signal, shutting down workers
988
+ W0830 05:51:57.252000 548836 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 548925 closing signal SIGTERM
989
+ W0830 05:51:57.253000 548836 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 548926 closing signal SIGTERM
990
+ W0830 05:51:57.254000 548836 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 548927 closing signal SIGTERM
991
+ W0830 05:51:57.256000 548836 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 548928 closing signal SIGTERM
992
+ W0830 05:51:57.258000 548836 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 548929 closing signal SIGTERM
993
+ W0830 05:51:57.259000 548836 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 548930 closing signal SIGTERM
994
+ W0830 05:51:57.261000 548836 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 548931 closing signal SIGTERM
995
+ W0830 05:51:57.263000 548836 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 548932 closing signal SIGTERM
996
+ Traceback (most recent call last):
997
+ File "/usr/local/bin/torchrun", line 8, in <module>
998
+ sys.exit(main())
999
+ ^^^^^^
1000
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/multiprocessing/errors/__init__.py", line 355, in wrapper
1001
+ return f(*args, **kwargs)
1002
+ ^^^^^^^^^^^^^^^^^^
1003
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/run.py", line 892, in main
1004
+ run(args)
1005
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/run.py", line 883, in run
1006
+ elastic_launch(
1007
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/launcher/api.py", line 139, in __call__
1008
+ return launch_agent(self._config, self._entrypoint, list(args))
1009
+ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1010
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/launcher/api.py", line 261, in launch_agent
1011
+ result = agent.run()
1012
+ ^^^^^^^^^^^
1013
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/metrics/api.py", line 138, in wrapper
1014
+ result = f(*args, **kwargs)
1015
+ ^^^^^^^^^^^^^^^^^^
1016
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/agent/server/api.py", line 711, in run
1017
+ result = self._invoke_run(role)
1018
+ ^^^^^^^^^^^^^^^^^^^^^^
1019
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/agent/server/api.py", line 870, in _invoke_run
1020
+ time.sleep(monitor_interval)
1021
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/multiprocessing/api.py", line 84, in _terminate_process_handler
1022
+ raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval)
1023
+ torch.distributed.elastic.multiprocessing.api.SignalException: Process 548836 got signal: 15
tiera-rerun-fixed-moonep/echo/a1/a1_oneslot_gbs16/G1__echo__one__seq4096__gbs16__cap2p40625/rep3/mem_samples.csv ADDED
@@ -0,0 +1,240 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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, 4
45
+ 4, 4
46
+ 5, 4
47
+ 6, 4
48
+ 7, 4
49
+ 0, 821
50
+ 1, 819
51
+ 2, 819
52
+ 3, 821
53
+ 4, 819
54
+ 5, 819
55
+ 6, 819
56
+ 7, 819
57
+ 0, 1537
58
+ 1, 1537
59
+ 2, 1537
60
+ 3, 1537
61
+ 4, 1537
62
+ 5, 1537
63
+ 6, 1537
64
+ 7, 1537
65
+ 0, 2231
66
+ 1, 2231
67
+ 2, 2231
68
+ 3, 2231
69
+ 4, 2231
70
+ 5, 2231
71
+ 6, 2231
72
+ 7, 2231
73
+ 0, 39527
74
+ 1, 39507
75
+ 2, 39549
76
+ 3, 39545
77
+ 4, 39635
78
+ 5, 39591
79
+ 6, 39555
80
+ 7, 39635
81
+ 0, 39825
82
+ 1, 39821
83
+ 2, 39819
84
+ 3, 39819
85
+ 4, 39819
86
+ 5, 39823
87
+ 6, 39835
88
+ 7, 39839
89
+ 0, 42357
90
+ 1, 42355
91
+ 2, 42355
92
+ 3, 42355
93
+ 4, 42355
94
+ 5, 42355
95
+ 6, 42355
96
+ 7, 42355
97
+ 0, 42583
98
+ 1, 42581
99
+ 2, 42581
100
+ 3, 42581
101
+ 4, 42581
102
+ 5, 42581
103
+ 6, 42581
104
+ 7, 42581
105
+ 0, 42711
106
+ 1, 42709
107
+ 2, 42701
108
+ 3, 42707
109
+ 4, 42711
110
+ 5, 42735
111
+ 6, 42711
112
+ 7, 42707
113
+ 0, 43493
114
+ 1, 43491
115
+ 2, 43491
116
+ 3, 43491
117
+ 4, 43491
118
+ 5, 43491
119
+ 6, 43491
120
+ 7, 43491
121
+ 0, 43493
122
+ 1, 43491
123
+ 2, 43491
124
+ 3, 43491
125
+ 4, 43491
126
+ 5, 43491
127
+ 6, 43491
128
+ 7, 43491
129
+ 0, 45265
130
+ 1, 45263
131
+ 2, 45263
132
+ 3, 45263
133
+ 4, 45263
134
+ 5, 45263
135
+ 6, 45263
136
+ 7, 45263
137
+ 0, 45265
138
+ 1, 45263
139
+ 2, 45263
140
+ 3, 45263
141
+ 4, 45263
142
+ 5, 45263
143
+ 6, 45263
144
+ 7, 45263
145
+ 0, 85827
146
+ 1, 85825
147
+ 2, 85825
148
+ 3, 85825
149
+ 4, 85825
150
+ 5, 85825
151
+ 6, 85825
152
+ 7, 85825
153
+ 0, 86905
154
+ 1, 86903
155
+ 2, 86903
156
+ 3, 86903
157
+ 4, 86903
158
+ 5, 86903
159
+ 6, 86903
160
+ 7, 86903
161
+ 0, 86905
162
+ 1, 86903
163
+ 2, 86903
164
+ 3, 86903
165
+ 4, 86903
166
+ 5, 86903
167
+ 6, 86903
168
+ 7, 86903
169
+ 0, 86905
170
+ 1, 86903
171
+ 2, 86903
172
+ 3, 86903
173
+ 4, 86903
174
+ 5, 86903
175
+ 6, 86903
176
+ 7, 86903
177
+ 0, 104407
178
+ 1, 104349
179
+ 2, 104349
180
+ 3, 104349
181
+ 4, 104405
182
+ 5, 104349
183
+ 6, 104349
184
+ 7, 104349
185
+ 0, 105875
186
+ 1, 105817
187
+ 2, 105817
188
+ 3, 105817
189
+ 4, 105873
190
+ 5, 105817
191
+ 6, 105817
192
+ 7, 105817
193
+ 0, 106523
194
+ 1, 106465
195
+ 2, 106465
196
+ 3, 106465
197
+ 4, 106521
198
+ 5, 106465
199
+ 6, 106465
200
+ 7, 106465
201
+ 0, 107157
202
+ 1, 107091
203
+ 2, 107073
204
+ 3, 107129
205
+ 4, 107153
206
+ 5, 107125
207
+ 6, 107129
208
+ 7, 107127
209
+ 0, 134577
210
+ 1, 136049
211
+ 2, 135823
212
+ 3, 136049
213
+ 4, 136245
214
+ 5, 138979
215
+ 6, 144191
216
+ 7, 136881
217
+ 0, 161641
218
+ 1, 161525
219
+ 2, 161527
220
+ 3, 161529
221
+ 4, 161723
222
+ 5, 161525
223
+ 6, 161749
224
+ 7, 161525
225
+ 0, 161641
226
+ 1, 161525
227
+ 2, 161527
228
+ 3, 161529
229
+ 4, 161723
230
+ 5, 161525
231
+ 6, 161749
232
+ 7, 161525
233
+ 0, 2097
234
+ 1, 2098
235
+ 2, 2418
236
+ 3, 161529
237
+ 4, 3698
238
+ 5, 2098
239
+ 6, 2098
240
+ 7, 2098
tiera-rerun-fixed-moonep/echo/a1/a1_oneslot_gbs16/G1__echo__one__seq4096__gbs16__cap2p40625/rep3/run.env ADDED
@@ -0,0 +1,70 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ backend=echo
2
+ config=G1
3
+ family=tierA
4
+ repo=/workspace/Megatron-MoE-Echo
5
+ branch=bracing-regions-through-kernels
6
+ commit=9a236793
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=one
42
+ slots_per_rank=1
43
+ capacity_mode=static
44
+ capacity=2.40625
45
+ cuda_graph=1
46
+ probe_mode=0
47
+ async_grad_reduce=0
48
+ branch_role=timing
49
+ memory_dump_mode=none
50
+ memory_snapshot_path=
51
+ cuda_graph_warmup_steps=3
52
+ recompute=none
53
+ force_balance=0
54
+ train_iters=12
55
+ ep_size=8
56
+ echo_dispatch_overlap=1
57
+ echo_dispatch_recompute=0
58
+ ultraep_quota_min_tokens=
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/echo/a1/a1_oneslot_gbs16/G1__echo__one__seq4096__gbs16__cap2p40625/rep3/train.log ADDED
@@ -0,0 +1,1023 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ==============================================================
2
+ G1 / echo (slots/rank=1 policy=one)
3
+ E=8 K=2 H'=14336 m=1
4
+ L=16 seq=4096 gbs=16 (2 microbatches)
5
+ cap_mode=static cap=2.40625 cg=1 recompute=none
6
+ memory_dump=none snapshots=/workspace/tiera-rerun-fixed-moonep/echo/a1/a1_oneslot_gbs16/G1__echo__one__seq4096__gbs16__cap2p40625/rep3/memory_snapshots/snapshot.pickle
7
+ repo=/workspace/Megatron-MoE-Echo @ 9a236793
8
+ ==============================================================
9
+ W0830 05:52:03.766000 552219 torch/distributed/run.py:766]
10
+ W0830 05:52:03.766000 552219 torch/distributed/run.py:766] *****************************************
11
+ W0830 05:52:03.766000 552219 torch/distributed/run.py:766] Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed.
12
+ W0830 05:52:03.766000 552219 torch/distributed/run.py:766] *****************************************
13
+ /workspace/Megatron-MoE-Echo/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
14
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
15
+ /workspace/Megatron-MoE-Echo/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
16
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
17
+ using world size: 8, data-parallel size: 8, context-parallel size: 1, hierarchical context-parallel sizes: None, tensor-model-parallel size: 1, pipeline-model-parallel size: 1
18
+ WARNING: overriding default arguments for tokenizer_type:GPT2BPETokenizer with tokenizer_type:GPT2BPETokenizer
19
+ Number of virtual stages per pipeline stage: None
20
+ accumulate and all-reduce gradients in fp32 for bfloat16 data type.
21
+ using torch.bfloat16 for parameters ...
22
+ /workspace/Megatron-MoE-Echo/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
23
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
24
+ ------------------------ arguments ------------------------
25
+ account_for_embedding_in_pipeline_split ......... False
26
+ account_for_loss_in_pipeline_split .............. False
27
+ accumulate_allreduce_grads_in_fp32 .............. True
28
+ adam_beta1 ...................................... 0.9
29
+ adam_beta2 ...................................... 0.999
30
+ adam_eps ........................................ 1e-08
31
+ add_bias_linear ................................. False
32
+ add_position_embedding .......................... False
33
+ add_qkv_bias .................................... False
34
+ adlr_autoresume ................................. False
35
+ adlr_autoresume_interval ........................ 1000
36
+ align_grad_reduce ............................... True
37
+ align_param_gather .............................. False
38
+ app_tag_run_name ................................ None
39
+ app_tag_run_version ............................. 0.0.0
40
+ apply_layernorm_1p .............................. False
41
+ apply_query_key_layer_scaling ................... False
42
+ apply_residual_connection_post_layernorm ........ False
43
+ apply_rope_fusion ............................... True
44
+ async_save ...................................... None
45
+ async_tensor_model_parallel_allreduce ........... True
46
+ attention_backend ............................... AttnBackend.auto
47
+ attention_dropout ............................... 0.0
48
+ attention_softmax_in_fp32 ....................... False
49
+ auto_detect_ckpt_format ......................... False
50
+ barrier_with_L1_time ............................ True
51
+ bert_binary_head ................................ True
52
+ bert_embedder_type .............................. megatron
53
+ bert_load ....................................... None
54
+ bf16 ............................................ True
55
+ bias_dropout_fusion ............................. True
56
+ bias_gelu_fusion ................................ False
57
+ bias_swiglu_fusion .............................. True
58
+ biencoder_projection_dim ........................ 0
59
+ biencoder_shared_query_context_model ............ False
60
+ block_data_path ................................. None
61
+ calc_ft_timeouts ................................ False
62
+ calculate_per_token_loss ........................ False
63
+ check_for_large_grads ........................... False
64
+ check_for_nan_in_loss_and_grad .................. False
65
+ check_for_spiky_loss ............................ False
66
+ check_weight_hash_across_dp_replicas_interval ... None
67
+ ckpt_assume_constant_structure .................. False
68
+ ckpt_convert_format ............................. None
69
+ ckpt_convert_save ............................... None
70
+ ckpt_convert_update_legacy_dist_opt_format ...... False
71
+ ckpt_format ..................................... torch_dist
72
+ ckpt_fully_parallel_load ........................ False
73
+ ckpt_fully_parallel_save ........................ True
74
+ ckpt_fully_parallel_save_deprecated ............. False
75
+ ckpt_step ....................................... None
76
+ classes_fraction ................................ 1.0
77
+ clip_grad ....................................... 1.0
78
+ clone_scatter_output_in_embedding ............... True
79
+ config_logger_dir ...............................
80
+ consumed_train_samples .......................... 0
81
+ consumed_valid_samples .......................... 0
82
+ context_parallel_size ........................... 1
83
+ cp_comm_type .................................... ['p2p']
84
+ create_attention_mask_in_dataloader ............. True
85
+ cross_entropy_fusion_impl ....................... native
86
+ cross_entropy_loss_fusion ....................... False
87
+ cuda_graph_scope ................................ full_iteration
88
+ cuda_graph_warmup_steps ......................... 3
89
+ data_args_path .................................. None
90
+ data_cache_path ................................. None
91
+ data_parallel_random_init ....................... False
92
+ data_parallel_sharding_strategy ................. no_shard
93
+ data_parallel_size .............................. 8
94
+ data_path ....................................... ['/workspace/data/wikitext/wikitext103_gpt2_text_document']
95
+ data_per_class_fraction ......................... 1.0
96
+ data_sharding ................................... True
97
+ dataloader_type ................................. single
98
+ ddp_average_in_collective ....................... False
99
+ ddp_bucket_size ................................. None
100
+ ddp_num_buckets ................................. None
101
+ ddp_pad_buckets_for_high_nccl_busbw ............. False
102
+ decoder_first_pipeline_num_layers ............... None
103
+ decoder_last_pipeline_num_layers ................ None
104
+ decoder_num_layers .............................. None
105
+ decoder_seq_length .............................. None
106
+ decoupled_lr .................................... None
107
+ decoupled_min_lr ................................ None
108
+ decrease_batch_size_if_needed ................... False
109
+ defer_embedding_wgrad_compute ................... False
110
+ delay_wgrad_compute ............................. False
111
+ deprecated_use_mcore_models ..................... True
112
+ deterministic_mode .............................. False
113
+ dino_bottleneck_size ............................ 256
114
+ dino_freeze_last_layer .......................... 1
115
+ dino_head_hidden_size ........................... 2048
116
+ dino_local_crops_number ......................... 10
117
+ dino_local_img_size ............................. 96
118
+ dino_norm_last_layer ............................ False
119
+ dino_teacher_temp ............................... 0.07
120
+ dino_warmup_teacher_temp ........................ 0.04
121
+ dino_warmup_teacher_temp_epochs ................. 30
122
+ disable_bf16_reduced_precision_matmul ........... False
123
+ disable_mamba_mem_eff_path ...................... False
124
+ disable_straggler_on_startup .................... False
125
+ dist_ckpt_format_deprecated ..................... None
126
+ dist_ckpt_strictness ............................ assume_ok_unexpected
127
+ distribute_saved_activations .................... False
128
+ distributed_backend ............................. nccl
129
+ distributed_timeout_minutes ..................... 10
130
+ embedding_init_method_std ....................... None
131
+ embedding_path .................................. None
132
+ empty_unused_memory_level ....................... 0
133
+ enable_cuda_graph ............................... True
134
+ enable_experimental ............................. False
135
+ enable_ft_package ............................... False
136
+ enable_gloo_process_groups ...................... True
137
+ enable_msc ...................................... True
138
+ enable_one_logger ............................... True
139
+ encoder_num_layers .............................. 16
140
+ encoder_seq_length .............................. 4096
141
+ end_weight_decay ................................ 0.1
142
+ eod_mask_loss ................................... False
143
+ error_injection_rate ............................ 0
144
+ error_injection_type ............................ transient_error
145
+ eval_interval ................................... 1000000
146
+ eval_iters ...................................... 0
147
+ evidence_data_path .............................. None
148
+ exit_duration_in_mins ........................... None
149
+ exit_interval ................................... None
150
+ exit_on_missing_checkpoint ...................... False
151
+ exit_signal_handler ............................. False
152
+ exp_avg_dtype ................................... torch.float32
153
+ exp_avg_sq_dtype ................................ torch.float32
154
+ expert_model_parallel_size ...................... 8
155
+ expert_tensor_parallel_size ..................... 1
156
+ export_force_local_attention .................... False
157
+ export_kd_cfg ................................... None
158
+ export_kd_teacher_ckpt_format ................... None
159
+ export_kd_teacher_load .......................... None
160
+ export_kv_cache_quant ........................... False
161
+ export_legacy_megatron .......................... False
162
+ export_model_type ............................... GPTModel
163
+ export_moe_apply_probs_on_input ................. False
164
+ export_qk_l2_norm ............................... False
165
+ export_quant_cfg ................................ None
166
+ export_real_quant_cfg ........................... None
167
+ export_te_mcore_model ........................... False
168
+ external_cuda_graph ............................. False
169
+ ffn_hidden_size ................................. 14336
170
+ finetune ........................................ False
171
+ finetune_data_split ............................. train
172
+ finetune_hf_dataset ............................. None
173
+ first_last_layers_bf16 .......................... False
174
+ flash_decode .................................... False
175
+ fp16 ............................................ False
176
+ fp16_lm_cross_entropy ........................... False
177
+ fp32_residual_connection ........................ False
178
+ fp8 ............................................. e4m3
179
+ fp8_amax_compute_algo ........................... most_recent
180
+ fp8_amax_history_len ............................ 1
181
+ fp8_interval .................................... 1
182
+ fp8_margin ...................................... 0
183
+ fp8_param_gather ................................ False
184
+ fp8_recipe ...................................... mxfp8
185
+ fp8_wgrad ....................................... True
186
+ fsdp_double_buffer .............................. False
187
+ global_batch_size ............................... 16
188
+ grad_reduce_in_bf16 ............................. False
189
+ gradient_accumulation_fusion .................... True
190
+ gradient_reduce_div_fusion ...................... True
191
+ group_query_attention ........................... True
192
+ head_lr_mult .................................... 1.0
193
+ heterogeneous_layers_config_encoded_json ........ None
194
+ heterogeneous_layers_config_path ................ None
195
+ hidden_dropout .................................. 0.0
196
+ hidden_size ..................................... 4096
197
+ hierarchical_context_parallel_sizes ............. None
198
+ high_priority_stream_groups ..................... []
199
+ hybrid_attention_ratio .......................... 0.0
200
+ hybrid_mlp_ratio ................................ 0.0
201
+ hybrid_override_pattern ......................... None
202
+ hysteresis ...................................... 2
203
+ ict_head_size ................................... None
204
+ ict_load ........................................ None
205
+ img_h ........................................... 224
206
+ img_w ........................................... 224
207
+ indexer_batch_size .............................. 128
208
+ indexer_log_interval ............................ 1000
209
+ inference_batch_times_seqlen_threshold .......... -1
210
+ inference_dynamic_batching ...................... False
211
+ inference_dynamic_batching_buffer_guaranteed_fraction 0.2
212
+ inference_dynamic_batching_buffer_overflow_factor None
213
+ inference_dynamic_batching_buffer_size_gb ....... 40.0
214
+ inference_dynamic_batching_chunk_size ........... 256
215
+ inference_dynamic_batching_max_requests_override None
216
+ inference_dynamic_batching_max_tokens_override .. None
217
+ inference_dynamic_batching_num_cuda_graphs ...... 16
218
+ inference_max_batch_size ........................ 8
219
+ inference_max_seq_length ........................ 2560
220
+ inference_rng_tracker ........................... False
221
+ init_method_std ................................. 0.01
222
+ init_method_xavier_uniform ...................... False
223
+ init_model_with_meta_device ..................... False
224
+ initial_loss_scale .............................. 4294967296
225
+ inprocess_active_world_size ..................... 8
226
+ inprocess_barrier_timeout ....................... 120
227
+ inprocess_completion_timeout .................... 120
228
+ inprocess_empty_cuda_cache ...................... False
229
+ inprocess_granularity ........................... node
230
+ inprocess_hard_timeout .......................... 90
231
+ inprocess_heartbeat_interval .................... 30
232
+ inprocess_heartbeat_timeout ..................... 60
233
+ inprocess_last_call_wait ........................ 1
234
+ inprocess_max_iterations ........................ None
235
+ inprocess_monitor_process_interval .............. 1.0
236
+ inprocess_monitor_thread_interval ............... 1.0
237
+ inprocess_progress_watchdog_interval ............ 1.0
238
+ inprocess_restart ............................... False
239
+ inprocess_soft_timeout .......................... 60
240
+ inprocess_termination_grace_time ................ 1
241
+ is_hybrid_model ................................. False
242
+ iter_per_epoch .................................. 1250
243
+ iterations_to_skip .............................. []
244
+ keep_fp8_transpose_cache_when_using_custom_fsdp . False
245
+ kitchen_config_file ............................. None
246
+ kitchen_recipe_number ........................... None
247
+ kv_channels ..................................... 128
248
+ kv_lora_rank .................................... 32
249
+ lazy_mpu_init ................................... None
250
+ load ............................................ None
251
+ load_main_params_from_ckpt ...................... None
252
+ load_model_opt_format ........................... False
253
+ local_rank ...................................... 0
254
+ log_energy ...................................... False
255
+ log_interval .................................... 1
256
+ log_loss_scale_to_tensorboard ................... True
257
+ log_memory_to_tensorboard ....................... False
258
+ log_num_zeros_in_grad ........................... False
259
+ log_params_norm ................................. False
260
+ log_progress .................................... False
261
+ log_straggler ................................... False
262
+ log_throughput .................................. True
263
+ log_timers_to_tensorboard ....................... False
264
+ log_validation_ppl_to_tensorboard ............... False
265
+ log_world_size_to_tensorboard ................... False
266
+ logging_level ................................... None
267
+ loss_scale ...................................... None
268
+ loss_scale_window ............................... 1000
269
+ lr .............................................. 0.0001
270
+ lr_decay_iters .................................. 8
271
+ lr_decay_samples ................................ None
272
+ lr_decay_style .................................. cosine
273
+ lr_warmup_fraction .............................. None
274
+ lr_warmup_init .................................. 0.0
275
+ lr_warmup_iters ................................. 5
276
+ lr_warmup_samples ............................... 0
277
+ lr_wsd_decay_iters .............................. None
278
+ lr_wsd_decay_samples ............................ None
279
+ lr_wsd_decay_style .............................. exponential
280
+ main_grads_dtype ................................ torch.float32
281
+ main_params_dtype ............................... torch.float32
282
+ make_vocab_size_divisible_by .................... 128
283
+ mamba_head_dim .................................. 64
284
+ mamba_num_groups ................................ 8
285
+ mamba_num_heads ................................. None
286
+ mamba_state_dim ................................. 128
287
+ manual_gc ....................................... False
288
+ manual_gc_eval .................................. True
289
+ manual_gc_interval .............................. 0
290
+ mask_factor ..................................... 1.0
291
+ mask_prob ....................................... 0.15
292
+ mask_type ....................................... random
293
+ masked_softmax_fusion ........................... False
294
+ max_position_embeddings ......................... 32768
295
+ max_tokens_to_oom ............................... 12000
296
+ memory_snapshot_path ............................ snapshot.pickle
297
+ merge_file ...................................... /workspace/data/wikitext/gpt2-merges.txt
298
+ micro_batch_size ................................ 1
299
+ microbatch_group_size_per_vp_stage .............. None
300
+ mid_level_dataset_surplus ....................... 0.005
301
+ min_loss_scale .................................. 1.0
302
+ min_lr .......................................... 1e-05
303
+ mlp_chunks_for_prefill .......................... 1
304
+ mmap_bin_files .................................. True
305
+ mock_data ....................................... False
306
+ moe_apply_probs_on_input ........................ False
307
+ moe_aux_loss_coeff .............................. 0.01
308
+ moe_deepep_num_sms .............................. 20
309
+ moe_echo_enable_random_offloading ............... False
310
+ moe_echo_expert_dispatch_overlap ................ True
311
+ moe_echo_expert_dispatcher_type ................. hybridep
312
+ moe_echo_recompute_expert_dispatch .............. False
313
+ moe_enable_deepep ............................... False
314
+ moe_enable_echo ................................. True
315
+ moe_enable_hybridep ............................. True
316
+ moe_expert_capacity_factor ...................... None
317
+ moe_extended_tp ................................. False
318
+ moe_ffn_hidden_size ............................. 14336
319
+ moe_grouped_gemm ................................ True
320
+ moe_hybridep_num_sms ............................ 16
321
+ moe_input_jitter_eps ............................ None
322
+ moe_layer_freq .................................. 1
323
+ moe_layer_recompute ............................. False
324
+ moe_num_echo_experts ............................ 8
325
+ moe_pad_expert_input_to_capacity ................ False
326
+ moe_per_layer_logging ........................... False
327
+ moe_permute_fusion .............................. False
328
+ moe_received_token_capacity ..................... 2.40625
329
+ moe_router_bias_update_rate ..................... 0.001
330
+ moe_router_dtype ................................ fp32
331
+ moe_router_enable_expert_bias ................... False
332
+ moe_router_force_load_balancing ................. False
333
+ moe_router_group_topk ........................... None
334
+ moe_router_load_balancing_type .................. aux_loss
335
+ moe_router_num_groups ........................... None
336
+ moe_router_padding_for_fp8 ...................... False
337
+ moe_router_pre_softmax .......................... False
338
+ moe_router_score_function ....................... softmax
339
+ moe_router_topk ................................. 2
340
+ moe_router_topk_scaling_factor .................. None
341
+ moe_shared_expert_intermediate_size ............. None
342
+ moe_shared_expert_overlap ....................... False
343
+ moe_token_dispatcher_type ....................... flex
344
+ moe_token_drop_policy ........................... probs
345
+ moe_upcycling_granularity ....................... 1
346
+ moe_use_device_initiated_grouped_gemm ........... True
347
+ moe_use_legacy_grouped_gemm ..................... False
348
+ moe_use_upcycling ............................... False
349
+ moe_z_loss_coeff ................................ None
350
+ mrope_section ................................... None
351
+ mscale .......................................... 1.0
352
+ mscale_all_dim .................................. 0.0
353
+ mtp_loss_scaling_factor ......................... 0.1
354
+ mtp_num_layers .................................. None
355
+ multi_latent_attention .......................... False
356
+ nccl_all_reduce_for_prefill ..................... False
357
+ nccl_communicator_config_path ................... None
358
+ nccl_ub ......................................... False
359
+ no_load_optim ................................... None
360
+ no_load_rng ..................................... None
361
+ no_persist_layer_norm ........................... False
362
+ no_rope_freq .................................... None
363
+ no_save_optim ................................... None
364
+ no_save_rng ..................................... None
365
+ non_persistent_ckpt_type ........................ None
366
+ non_persistent_global_ckpt_dir .................. None
367
+ non_persistent_local_ckpt_algo .................. fully_parallel
368
+ non_persistent_local_ckpt_dir ................... None
369
+ non_persistent_save_interval .................... None
370
+ norm_epsilon .................................... 1e-05
371
+ normalization ................................... RMSNorm
372
+ num_attention_heads ............................. 32
373
+ num_channels .................................... 3
374
+ num_classes ..................................... 1000
375
+ num_dataset_builder_threads ..................... 1
376
+ num_distributed_optimizer_instances ............. 1
377
+ num_experts ..................................... 8
378
+ num_layers ...................................... 16
379
+ num_layers_at_end_in_bf16 ....................... 1
380
+ num_layers_at_start_in_bf16 ..................... 1
381
+ num_layers_per_virtual_pipeline_stage ........... None
382
+ num_query_groups ................................ 8
383
+ num_virtual_stages_per_pipeline_rank ............ None
384
+ num_workers ..................................... 2
385
+ object_storage_cache_path ....................... None
386
+ one_logger_async ................................ False
387
+ one_logger_project .............................. megatron-lm
388
+ one_logger_run_name ............................. None
389
+ onnx_safe ....................................... None
390
+ openai_gelu ..................................... False
391
+ optimizer ....................................... adam
392
+ optimizer_cpu_offload ........................... False
393
+ optimizer_offload_fraction ...................... 1.0
394
+ output_bert_embeddings .......................... False
395
+ overlap_cpu_optimizer_d2h_h2d ................... False
396
+ overlap_grad_reduce ............................. True
397
+ overlap_moe_expert_parallel_comm ................ False
398
+ overlap_p2p_comm ................................ False
399
+ overlap_p2p_comm_warmup_flush ................... False
400
+ overlap_param_gather ............................ True
401
+ overlap_param_gather_with_optimizer_step ........ False
402
+ override_opt_param_scheduler .................... False
403
+ params_dtype .................................... torch.bfloat16
404
+ patch_dim ....................................... 16
405
+ per_split_data_args_path ........................ None
406
+ perform_initialization .......................... True
407
+ pin_cpu_grads ................................... True
408
+ pin_cpu_params .................................. True
409
+ pipeline_model_parallel_comm_backend ............ None
410
+ pipeline_model_parallel_layout .................. None
411
+ pipeline_model_parallel_size .................... 1
412
+ position_embedding_type ......................... rope
413
+ pretrained_checkpoint ........................... None
414
+ profile ......................................... False
415
+ profile_ranks ................................... [0]
416
+ profile_step_end ................................ 12
417
+ profile_step_start .............................. 10
418
+ q_lora_rank ..................................... None
419
+ qk_head_dim ..................................... 128
420
+ qk_l2_norm ...................................... False
421
+ qk_layernorm .................................... False
422
+ qk_pos_emb_head_dim ............................. 64
423
+ query_in_block_prob ............................. 0.1
424
+ rampup_batch_size ............................... None
425
+ rank ............................................ 0
426
+ recompute_granularity ........................... None
427
+ recompute_method ................................ None
428
+ recompute_modules ............................... None
429
+ recompute_num_layers ............................ None
430
+ record_memory_history ........................... False
431
+ relative_attention_max_distance ................. 128
432
+ relative_attention_num_buckets .................. 32
433
+ replication ..................................... False
434
+ replication_factor .............................. 2
435
+ replication_jump ................................ None
436
+ rerun_mode ...................................... validate_results
437
+ reset_attention_mask ............................ False
438
+ reset_position_ids .............................. False
439
+ result_rejected_tracker_filename ................ None
440
+ retriever_report_topk_accuracies ................ []
441
+ retriever_score_scaling ......................... False
442
+ retriever_seq_length ............................ 256
443
+ retro_add_retriever ............................. False
444
+ retro_attention_gate ............................ 1
445
+ retro_cyclic_train_iters ........................ None
446
+ retro_encoder_attention_dropout ................. 0.1
447
+ retro_encoder_hidden_dropout .................... 0.1
448
+ retro_encoder_layers ............................ 2
449
+ retro_num_neighbors ............................. 2
450
+ retro_num_retrieved_chunks ...................... 2
451
+ retro_project_dir ............................... None
452
+ retro_verify_neighbor_count ..................... True
453
+ reuse_grad_buf_for_mxfp8_param_ag ............... False
454
+ rope_scaling_factor ............................. 8.0
455
+ rope_type ....................................... None
456
+ rotary_base ..................................... 1000000
457
+ rotary_interleaved .............................. False
458
+ rotary_percent .................................. 1.0
459
+ rotary_scaling_factor ........................... 1.0
460
+ rotary_seq_len_interpolation_factor ............. None
461
+ run_workload_inspector_server ................... False
462
+ sample_rate ..................................... 1.0
463
+ save ............................................ None
464
+ save_interval ................................... None
465
+ save_retain_interval ............................ None
466
+ scatter_gather_tensors_in_pipeline .............. True
467
+ seed ............................................ 1234
468
+ seq_length ...................................... 4096
469
+ sequence_parallel ............................... False
470
+ sft ............................................. False
471
+ sft_tokenizer_prompt_format ..................... nemotron-h-aligned
472
+ sgd_momentum .................................... 0.9
473
+ short_seq_prob .................................. 0.1
474
+ skip_train ...................................... False
475
+ skipped_train_samples ........................... 0
476
+ spec ............................................ None
477
+ split ........................................... 100,0,0
478
+ squared_relu .................................... False
479
+ start_weight_decay .............................. 0.1
480
+ straggler_ctrlr_port ............................ 65535
481
+ straggler_minmax_count .......................... 1
482
+ suggested_communication_unit_size ............... None
483
+ swiglu .......................................... True
484
+ swin_backbone_type .............................. tiny
485
+ symmetric_ar_type ............................... None
486
+ te_rng_tracker .................................. True
487
+ tensor_model_parallel_size ...................... 1
488
+ tensorboard_dir ................................. None
489
+ tensorboard_log_interval ........................ 1
490
+ tensorboard_queue_size .......................... 1000
491
+ test_data_path .................................. None
492
+ test_mode ....................................... False
493
+ tiktoken_num_special_tokens ..................... 1000
494
+ tiktoken_pattern ................................ None
495
+ tiktoken_special_tokens ......................... None
496
+ timing_log_level ................................ 0
497
+ timing_log_option ............................... minmax
498
+ titles_data_path ................................ None
499
+ tokenizer_model ................................. None
500
+ tokenizer_type .................................. GPT2BPETokenizer
501
+ torch_fsdp2_reshard_after_forward ............... True
502
+ tp_comm_bootstrap_backend ....................... nccl
503
+ tp_comm_bulk_dgrad .............................. True
504
+ tp_comm_bulk_wgrad .............................. True
505
+ tp_comm_overlap ................................. False
506
+ tp_comm_overlap_ag .............................. True
507
+ tp_comm_overlap_cfg ............................. None
508
+ tp_comm_overlap_rs .............................. True
509
+ tp_comm_overlap_rs_dgrad ........................ False
510
+ tp_comm_split_ag ................................ True
511
+ tp_comm_split_rs ................................ True
512
+ train_data_path ................................. None
513
+ train_iters ..................................... 12
514
+ train_samples ................................... None
515
+ train_sync_interval ............................. None
516
+ transformer_impl ................................ transformer_engine
517
+ transformer_pipeline_model_parallel_size ........ 1
518
+ untie_embeddings_and_output_weights ............. True
519
+ use_checkpoint_args ............................. False
520
+ use_checkpoint_opt_param_scheduler .............. False
521
+ use_cpu_initialization .......................... None
522
+ use_custom_fsdp ................................. False
523
+ use_dist_ckpt ................................... True
524
+ use_dist_ckpt_deprecated ........................ False
525
+ use_distributed_optimizer ....................... True
526
+ use_flash_attn .................................. False
527
+ use_legacy_models ............................... False
528
+ use_mp_args_from_checkpoint_args ................ False
529
+ use_one_sent_docs ............................... False
530
+ use_persistent_ckpt_worker ...................... False
531
+ use_precision_aware_optimizer ................... False
532
+ use_pytorch_profiler ............................ False
533
+ use_ring_exchange_p2p ........................... False
534
+ use_rope_scaling ................................ False
535
+ use_rotary_position_embeddings .................. False
536
+ use_sharp ....................................... False
537
+ use_tokenizer_model_from_checkpoint_args ........ True
538
+ use_torch_fsdp2 ................................. False
539
+ use_torch_optimizer_for_cpu_offload ............. False
540
+ use_tp_pp_dp_mapping ............................ False
541
+ v_head_dim ...................................... 128
542
+ valid_data_path ................................. None
543
+ variable_seq_lengths ............................ False
544
+ virtual_pipeline_model_parallel_size ............ None
545
+ vision_backbone_type ............................ vit
546
+ vision_pretraining .............................. False
547
+ vision_pretraining_type ......................... classify
548
+ vocab_extra_ids ................................. 0
549
+ vocab_file ...................................... /workspace/data/wikitext/gpt2-vocab.json
550
+ vocab_size ...................................... None
551
+ wandb_exp_name ..................................
552
+ wandb_project ...................................
553
+ wandb_save_dir ..................................
554
+ weight_decay .................................... 0.1
555
+ weight_decay_incr_style ......................... constant
556
+ wgrad_deferral_limit ............................ 0
557
+ world_size ...................................... 8
558
+ yaml_cfg ........................................ None
559
+ -------------------- end of arguments ---------------------
560
+ INFO:megatron.core.num_microbatches_calculator:setting number of microbatches to constant 2
561
+ > building GPT2BPETokenizer tokenizer ...
562
+ 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
563
+ /workspace/Megatron-MoE-Echo/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
564
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
565
+ > padded vocab (size: 50257) with 47 dummy tokens (new size: 50304)
566
+ WARNING:megatron.core.rerun_state_machine:RerunStateMachine initialized in mode RerunMode.VALIDATE_RESULTS
567
+ > initializing torch distributed ...
568
+ /workspace/Megatron-MoE-Echo/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
569
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
570
+ /workspace/Megatron-MoE-Echo/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
571
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
572
+ /workspace/Megatron-MoE-Echo/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
573
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
574
+ /workspace/Megatron-MoE-Echo/megatron/training/arguments.py:831: UserWarning: Disabling sequence parallelism because tensor model parallelism is disabled
575
+ warnings.warn("Disabling sequence parallelism because tensor model parallelism is disabled")
576
+ [Gloo] Rank 3 is connected to 7 peer ranks. [Gloo] Rank Expected number of connected peer ranks is : 70 is connected to
577
+ 7 peer ranks. Expected number of connected peer ranks is : 7
578
+ [Gloo] Rank [Gloo] Rank [Gloo] Rank 264 is connected to is connected to is connected to 777 peer ranks. peer ranks. peer ranks. Expected number of connected peer ranks is : Expected number of connected peer ranks is : Expected number of connected peer ranks is : [Gloo] Rank 777[Gloo] Rank 5
579
+
580
+
581
+ 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 : 7
582
+
583
+ [Gloo] Rank 1 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
584
+ [Gloo] Rank [Gloo] Rank [Gloo] Rank 301[Gloo] Rank is connected to is connected to is connected to 2777 is connected to peer ranks. peer ranks. peer ranks. 7Expected number of connected peer ranks is : Expected number of connected peer ranks is : Expected number of connected peer ranks is : peer ranks. 777Expected number of connected peer ranks is :
585
+
586
+
587
+ 7
588
+ [Gloo] Rank 6 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
589
+ [Gloo] Rank 4 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
590
+ [Gloo] Rank 5 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
591
+ [Gloo] Rank 7 is connected to 7 peer ranks. Expected number of connected peer ranks is : 7
592
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
593
+ > initialized tensor model parallel with size 1
594
+ > initialized pipeline model parallel with size 1
595
+ > setting random seeds to 1234 ...
596
+ [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
597
+
598
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
599
+ [Gloo] Rank [Gloo] Rank 00 is connected to is connected to 00 peer ranks. peer ranks. Expected number of connected peer ranks is : Expected number of connected peer ranks is : 00
600
+
601
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
602
+ [Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
603
+ [rank2]:[W830 05:52:21.940654161 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.
604
+ [rank3]:[W830 05:52:21.940746743 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.
605
+ [rank1]:[W830 05:52:21.940769847 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.
606
+ [rank7]:[W830 05:52:21.941145425 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.
607
+ [rank6]:[W830 05:52:21.941329057 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.
608
+ [rank5]:[W830 05:52:21.941388554 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.
609
+ [rank4]:[W830 05:52:21.941404981 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.
610
+ > compiling dataset index builder ...
611
+ make: Entering directory '/workspace/Megatron-MoE-Echo/megatron/core/datasets'
612
+ make: Nothing to be done for 'default'.
613
+ make: Leaving directory '/workspace/Megatron-MoE-Echo/megatron/core/datasets'
614
+ >>> done with dataset index builder. Compilation time: 0.035 seconds
615
+ WARNING: constraints for invoking optimized fused softmax kernel are not met. We default back to unfused kernel invocations.
616
+ > compiling and loading fused kernels ...
617
+ [rank0]:[W830 05:52:21.094123573 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.
618
+ >>> done with compiling and loading fused kernels. Compilation time: 3.760 seconds
619
+ time to initialize megatron (seconds): 7.151
620
+ [after megatron is initialized] datetime: 2026-08-30 05:52:27
621
+ building GPT model ...
622
+ /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.
623
+ warnings.warn(
624
+ /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.
625
+ warnings.warn(
626
+ /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.
627
+ warnings.warn(
628
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
629
+ warnings.warn(
630
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
631
+ warnings.warn(
632
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
633
+ warnings.warn(
634
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
635
+ warnings.warn(
636
+ /usr/local/lib/python3.12/dist-packages/transformer_engine/pytorch/cpu_offload.py:708: DeprecationWarning: Offloading weights is deprecated. Using offload_weights=True does not have any effect.
637
+ warnings.warn(
638
+ > number of parameters on (tensor, pipeline) model parallel rank (0, 0): 3902410752
639
+ INFO:megatron.core.distributed.distributed_data_parallel:Setting up DistributedDataParallel with config DistributedDataParallelConfig(grad_reduce_in_fp32=True, overlap_grad_reduce=True, overlap_param_gather=True, align_param_gather=False, use_distributed_optimizer=True, num_distributed_optimizer_instances=1, check_for_nan_in_grad=False, check_for_large_grads=False, bucket_size=40000000, pad_buckets_for_high_nccl_busbw=False, average_in_collective=False, fp8_param_gather=False, reuse_grad_buf_for_mxfp8_param_ag=False, use_custom_fsdp=False, data_parallel_sharding_strategy='no_shard', gradient_reduce_div_fusion=True, suggested_communication_unit_size=None, preserve_fp32_weights=True, keep_fp8_transpose_cache_when_using_custom_fsdp=False, nccl_ub=False, fsdp_double_buffer=False)
640
+ INFO:megatron.core.distributed.param_and_grad_buffer:Number of buckets for gradient all-reduce / reduce-scatter: 18
641
+ Params for bucket 1 (206045184 elements, 206045184 padded size):
642
+ module.output_layer.weight
643
+ Params for bucket 2 (41988096 elements, 41988096 padded size):
644
+ module.decoder.layers.15.self_attention.linear_qkv.weight
645
+ module.decoder.layers.15.pre_mlp_layernorm.weight
646
+ module.decoder.final_layernorm.weight
647
+ module.decoder.layers.15.mlp.router.weight
648
+ module.decoder.layers.15.self_attention.linear_qkv.layer_norm_weight
649
+ module.decoder.layers.15.self_attention.linear_proj.weight
650
+ Params for bucket 3 (41984000 elements, 41984000 padded size):
651
+ module.decoder.layers.14.mlp.router.weight
652
+ module.decoder.layers.14.pre_mlp_layernorm.weight
653
+ module.decoder.layers.14.self_attention.linear_qkv.weight
654
+ module.decoder.layers.14.self_attention.linear_proj.weight
655
+ module.decoder.layers.14.self_attention.linear_qkv.layer_norm_weight
656
+ Params for bucket 4 (41984000 elements, 41984000 padded size):
657
+ module.decoder.layers.13.mlp.router.weight
658
+ module.decoder.layers.13.self_attention.linear_qkv.layer_norm_weight
659
+ module.decoder.layers.13.self_attention.linear_proj.weight
660
+ module.decoder.layers.13.self_attention.linear_qkv.weight
661
+ module.decoder.layers.13.pre_mlp_layernorm.weight
662
+ Params for bucket 5 (41984000 elements, 41984000 padded size):
663
+ module.decoder.layers.12.pre_mlp_layernorm.weight
664
+ module.decoder.layers.12.self_attention.linear_qkv.layer_norm_weight
665
+ module.decoder.layers.12.self_attention.linear_proj.weight
666
+ module.decoder.layers.12.mlp.router.weight
667
+ module.decoder.layers.12.self_attention.linear_qkv.weight
668
+ Params for bucket 6 (41984000 elements, 41984000 padded size):
669
+ module.decoder.layers.11.self_attention.linear_qkv.weight
670
+ module.decoder.layers.11.pre_mlp_layernorm.weight
671
+ module.decoder.layers.11.mlp.router.weight
672
+ module.decoder.layers.11.self_attention.linear_qkv.layer_norm_weight
673
+ module.decoder.layers.11.self_attention.linear_proj.weight
674
+ Params for bucket 7 (41984000 elements, 41984000 padded size):
675
+ module.decoder.layers.10.self_attention.linear_qkv.weight
676
+ module.decoder.layers.10.pre_mlp_layernorm.weight
677
+ module.decoder.layers.10.mlp.router.weight
678
+ module.decoder.layers.10.self_attention.linear_qkv.layer_norm_weight
679
+ module.decoder.layers.10.self_attention.linear_proj.weight
680
+ Params for bucket 8 (41984000 elements, 41984000 padded size):
681
+ module.decoder.layers.9.mlp.router.weight
682
+ module.decoder.layers.9.self_attention.linear_proj.weight
683
+ module.decoder.layers.9.self_attention.linear_qkv.layer_norm_weight
684
+ module.decoder.layers.9.pre_mlp_layernorm.weight
685
+ module.decoder.layers.9.self_attention.linear_qkv.weight
686
+ Params for bucket 9 (41984000 elements, 41984000 padded size):
687
+ module.decoder.layers.8.self_attention.linear_proj.weight
688
+ module.decoder.layers.8.self_attention.linear_qkv.layer_norm_weight
689
+ module.decoder.layers.8.mlp.router.weight
690
+ module.decoder.layers.8.pre_mlp_layernorm.weight
691
+ module.decoder.layers.8.self_attention.linear_qkv.weight
692
+ Params for bucket 10 (41984000 elements, 41984000 padded size):
693
+ module.decoder.layers.7.mlp.router.weight
694
+ module.decoder.layers.7.self_attention.linear_qkv.weight
695
+ module.decoder.layers.7.pre_mlp_layernorm.weight
696
+ module.decoder.layers.7.self_attention.linear_proj.weight
697
+ module.decoder.layers.7.self_attention.linear_qkv.layer_norm_weight
698
+ Params for bucket 11 (41984000 elements, 41984000 padded size):
699
+ module.decoder.layers.6.self_attention.linear_qkv.weight
700
+ module.decoder.layers.6.pre_mlp_layernorm.weight
701
+ module.decoder.layers.6.mlp.router.weight
702
+ module.decoder.layers.6.self_attention.linear_proj.weight
703
+ module.decoder.layers.6.self_attention.linear_qkv.layer_norm_weight
704
+ Params for bucket 12 (41984000 elements, 41984000 padded size):
705
+ module.decoder.layers.5.self_attention.linear_qkv.layer_norm_weight
706
+ module.decoder.layers.5.mlp.router.weight
707
+ module.decoder.layers.5.self_attention.linear_qkv.weight
708
+ module.decoder.layers.5.pre_mlp_layernorm.weight
709
+ module.decoder.layers.5.self_attention.linear_proj.weight
710
+ Params for bucket 13 (41984000 elements, 41984000 padded size):
711
+ module.decoder.layers.4.mlp.router.weight
712
+ module.decoder.layers.4.self_attention.linear_qkv.layer_norm_weight
713
+ module.decoder.layers.4.self_attention.linear_proj.weight
714
+ module.decoder.layers.4.self_attention.linear_qkv.weight
715
+ module.decoder.layers.4.pre_mlp_layernorm.weight
716
+ Params for bucket 14 (41984000 elements, 41984000 padded size):
717
+ module.decoder.layers.3.self_attention.linear_qkv.weight
718
+ module.decoder.layers.3.pre_mlp_layernorm.weight
719
+ module.decoder.layers.3.mlp.router.weight
720
+ module.decoder.layers.3.self_attention.linear_proj.weight
721
+ module.decoder.layers.3.self_attention.linear_qkv.layer_norm_weight
722
+ Params for bucket 15 (41984000 elements, 41984000 padded size):
723
+ module.decoder.layers.2.self_attention.linear_qkv.weight
724
+ module.decoder.layers.2.pre_mlp_layernorm.weight
725
+ module.decoder.layers.2.mlp.router.weight
726
+ module.decoder.layers.2.self_attention.linear_qkv.layer_norm_weight
727
+ module.decoder.layers.2.self_attention.linear_proj.weight
728
+ Params for bucket 16 (41984000 elements, 41984000 padded size):
729
+ module.decoder.layers.1.pre_mlp_layernorm.weight
730
+ module.decoder.layers.1.self_attention.linear_qkv.weight
731
+ module.decoder.layers.1.mlp.router.weight
732
+ module.decoder.layers.1.self_attention.linear_qkv.layer_norm_weight
733
+ module.decoder.layers.1.self_attention.linear_proj.weight
734
+ Params for bucket 17 (41984000 elements, 41984000 padded size):
735
+ module.decoder.layers.0.self_attention.linear_qkv.layer_norm_weight
736
+ module.decoder.layers.0.pre_mlp_layernorm.weight
737
+ module.decoder.layers.0.self_attention.linear_qkv.weight
738
+ module.decoder.layers.0.mlp.router.weight
739
+ module.decoder.layers.0.self_attention.linear_proj.weight
740
+ Params for bucket 18 (206045184 elements, 206045184 padded size):
741
+ module.embedding.word_embeddings.weight
742
+ INFO:megatron.core.distributed.param_and_grad_buffer:Number of buckets for gradient all-reduce / reduce-scatter: 32
743
+ Params for bucket 1 (58720256 elements, 58720256 padded size):
744
+ module.decoder.layers.15.mlp.experts.linear_fc2.weight0
745
+ Params for bucket 2 (117440512 elements, 117440512 padded size):
746
+ module.decoder.layers.15.mlp.experts.linear_fc1.weight0
747
+ Params for bucket 3 (58720256 elements, 58720256 padded size):
748
+ module.decoder.layers.14.mlp.experts.linear_fc2.weight0
749
+ Params for bucket 4 (117440512 elements, 117440512 padded size):
750
+ module.decoder.layers.14.mlp.experts.linear_fc1.weight0
751
+ Params for bucket 5 (58720256 elements, 58720256 padded size):
752
+ module.decoder.layers.13.mlp.experts.linear_fc2.weight0
753
+ Params for bucket 6 (117440512 elements, 117440512 padded size):
754
+ module.decoder.layers.13.mlp.experts.linear_fc1.weight0
755
+ Params for bucket 7 (58720256 elements, 58720256 padded size):
756
+ module.decoder.layers.12.mlp.experts.linear_fc2.weight0
757
+ Params for bucket 8 (117440512 elements, 117440512 padded size):
758
+ module.decoder.layers.12.mlp.experts.linear_fc1.weight0
759
+ Params for bucket 9 (58720256 elements, 58720256 padded size):
760
+ module.decoder.layers.11.mlp.experts.linear_fc2.weight0
761
+ Params for bucket 10 (117440512 elements, 117440512 padded size):
762
+ module.decoder.layers.11.mlp.experts.linear_fc1.weight0
763
+ Params for bucket 11 (58720256 elements, 58720256 padded size):
764
+ module.decoder.layers.10.mlp.experts.linear_fc2.weight0
765
+ Params for bucket 12 (117440512 elements, 117440512 padded size):
766
+ module.decoder.layers.10.mlp.experts.linear_fc1.weight0
767
+ Params for bucket 13 (58720256 elements, 58720256 padded size):
768
+ module.decoder.layers.9.mlp.experts.linear_fc2.weight0
769
+ Params for bucket 14 (117440512 elements, 117440512 padded size):
770
+ module.decoder.layers.9.mlp.experts.linear_fc1.weight0
771
+ Params for bucket 15 (58720256 elements, 58720256 padded size):
772
+ module.decoder.layers.8.mlp.experts.linear_fc2.weight0
773
+ Params for bucket 16 (117440512 elements, 117440512 padded size):
774
+ module.decoder.layers.8.mlp.experts.linear_fc1.weight0
775
+ Params for bucket 17 (58720256 elements, 58720256 padded size):
776
+ module.decoder.layers.7.mlp.experts.linear_fc2.weight0
777
+ Params for bucket 18 (117440512 elements, 117440512 padded size):
778
+ module.decoder.layers.7.mlp.experts.linear_fc1.weight0
779
+ Params for bucket 19 (58720256 elements, 58720256 padded size):
780
+ module.decoder.layers.6.mlp.experts.linear_fc2.weight0
781
+ Params for bucket 20 (117440512 elements, 117440512 padded size):
782
+ module.decoder.layers.6.mlp.experts.linear_fc1.weight0
783
+ Params for bucket 21 (58720256 elements, 58720256 padded size):
784
+ module.decoder.layers.5.mlp.experts.linear_fc2.weight0
785
+ Params for bucket 22 (117440512 elements, 117440512 padded size):
786
+ module.decoder.layers.5.mlp.experts.linear_fc1.weight0
787
+ Params for bucket 23 (58720256 elements, 58720256 padded size):
788
+ module.decoder.layers.4.mlp.experts.linear_fc2.weight0
789
+ Params for bucket 24 (117440512 elements, 117440512 padded size):
790
+ module.decoder.layers.4.mlp.experts.linear_fc1.weight0
791
+ Params for bucket 25 (58720256 elements, 58720256 padded size):
792
+ module.decoder.layers.3.mlp.experts.linear_fc2.weight0
793
+ Params for bucket 26 (117440512 elements, 117440512 padded size):
794
+ module.decoder.layers.3.mlp.experts.linear_fc1.weight0
795
+ Params for bucket 27 (58720256 elements, 58720256 padded size):
796
+ module.decoder.layers.2.mlp.experts.linear_fc2.weight0
797
+ Params for bucket 28 (117440512 elements, 117440512 padded size):
798
+ module.decoder.layers.2.mlp.experts.linear_fc1.weight0
799
+ Params for bucket 29 (58720256 elements, 58720256 padded size):
800
+ module.decoder.layers.1.mlp.experts.linear_fc2.weight0
801
+ Params for bucket 30 (117440512 elements, 117440512 padded size):
802
+ module.decoder.layers.1.mlp.experts.linear_fc1.weight0
803
+ Params for bucket 31 (58720256 elements, 58720256 padded size):
804
+ module.decoder.layers.0.mlp.experts.linear_fc2.weight0
805
+ Params for bucket 32 (117440512 elements, 117440512 padded size):
806
+ module.decoder.layers.0.mlp.experts.linear_fc1.weight0
807
+ 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 0x772934b8c470>, config_logger_dir='')
808
+ INFO:megatron.core.optimizer_param_scheduler:> learning rate decay style: cosine
809
+ [after model, optimizer, and learning rate scheduler are built] datetime: 2026-08-30 05:52:27
810
+ > building train, validation, and test datasets ...
811
+ > datasets target sizes (minimum size):
812
+ train: 192
813
+ validation: 0
814
+ test: 0
815
+ INFO:megatron.core.datasets.blended_megatron_dataset_config:Let split_matrix = [(0, 1.0), None, None]
816
+ > building train, validation, and test datasets for GPT ...
817
+ 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 0x772935213e30>, 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)
818
+ INFO:megatron.core.datasets.indexed_dataset:Load the _IndexReader from /workspace/data/wikitext/wikitext103_gpt2_text_document.idx
819
+ INFO:megatron.core.datasets.indexed_dataset: Extract the sequence lengths
820
+ INFO:megatron.core.datasets.indexed_dataset: Extract the sequence pointers
821
+ INFO:megatron.core.datasets.indexed_dataset: Extract the document indices
822
+ INFO:megatron.core.datasets.indexed_dataset:> total number of sequences: 1165029
823
+ INFO:megatron.core.datasets.indexed_dataset:> total number of documents: 1801350
824
+ INFO:megatron.core.datasets.gpt_dataset:Load the GPTDataset train indices
825
+ INFO:megatron.core.datasets.gpt_dataset: Load the document index from 44ed1ee5bcb9e5b030387b8d27a6c406-GPTDataset-train-document_index.npy
826
+ INFO:megatron.core.datasets.gpt_dataset: Load the sample index from 44ed1ee5bcb9e5b030387b8d27a6c406-GPTDataset-train-sample_index.npy
827
+ INFO:megatron.core.datasets.gpt_dataset: Load the shuffle index from 44ed1ee5bcb9e5b030387b8d27a6c406-GPTDataset-train-shuffle_index.npy
828
+ INFO:megatron.core.datasets.gpt_dataset:> total number of samples: 29073
829
+ > finished creating GPT datasets ...
830
+ [after dataloaders are built] datetime: 2026-08-30 05:52:27
831
+ done with setup ...
832
+ training ...
833
+ Overwriting rerun_state_machine.current_iteration from -1 to 0...
834
+ (min, max) time across ranks (ms):
835
+ model-and-optimizer-setup ......................: (200.74, 203.31)
836
+ train/valid/test-data-iterators-setup ..........: (62.84, 71.31)
837
+ [before the start of training step] datetime: 2026-08-30 05:52:27
838
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
839
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
840
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
841
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
842
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
843
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
844
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
845
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
846
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
847
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
848
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
849
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
850
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
851
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
852
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
853
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
854
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
855
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
856
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
857
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
858
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
859
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
860
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
861
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
862
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
863
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
864
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
865
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
866
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
867
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
868
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
869
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
870
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
871
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
872
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
873
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
874
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
875
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
876
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
877
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
878
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
879
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
880
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
881
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
882
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
883
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
884
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
885
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
886
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
887
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
888
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
889
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
890
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
891
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
892
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
893
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
894
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
895
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
896
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
897
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
898
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
899
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
900
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
901
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
902
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
903
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
904
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
905
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
906
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
907
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
908
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
909
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
910
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
911
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
912
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
913
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
914
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
915
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
916
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
917
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
918
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
919
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
920
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
921
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
922
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
923
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
924
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
925
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
926
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
927
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
928
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
929
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
930
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
931
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
932
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:482: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
933
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
934
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
935
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
936
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
937
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
938
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
939
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
940
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
941
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
942
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
943
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
944
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
945
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
946
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
947
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
948
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:334: UserWarning: The num_dispatched_tokens is deprecated, it will be removed in the future.
949
+ warnings.warn("The num_dispatched_tokens is deprecated, it will be removed in the future.")
950
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
951
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
952
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
953
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
954
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
955
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
956
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
957
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
958
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
959
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
960
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
961
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
962
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
963
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
964
+ /workspace/DeepEP/deep_ep/hybrid_ep_buffer.py:336: UserWarning: The use_host_meta is deprecated, it will be removed in the future.
965
+ warnings.warn("The use_host_meta is deprecated, it will be removed in the future.")
966
+ [2026-08-30 05:53:05] iteration 1/ 12 | consumed samples: 16 | elapsed time per iteration (ms): 37878.3 | throughput per GPU (TFLOP/s/GPU): 8.8 | learning rate: 2.000000E-05 | global batch size: 16 | lm loss: 1.104376E+01 | loss scale: 1.0 | grad norm: 34.300 | number of skipped iterations: 0 | number of nan iterations: 0 |Number of parameters in transformer block in billions: 23.22
967
+
968
+ Number of parameters in embedding layers in billions: 0.41
969
+ Total number of parameters in billions: 23.63
970
+ Number of parameters in most loaded shard in billions: 23.6320
971
+ Theoretical memory footprints: weight and optimizer=169029.43 MB
972
+ [Rank 0] (after 1 iterations) memory (MB) | allocated: 63795.404296875 | max allocated: 96519.67578125 | reserved: 99122.0 | max reserved: 99122.0
973
+ [2026-08-30 05:53:06] iteration 2/ 12 | consumed samples: 32 | elapsed time per iteration (ms): 570.3 | throughput per GPU (TFLOP/s/GPU): 584.6 | learning rate: 4.000000E-05 | global batch size: 16 | lm loss: 1.104935E+01 | loss scale: 1.0 | grad norm: 35.986 | number of skipped iterations: 0 | number of nan iterations: 0 |
974
+ [2026-08-30 05:53:06] iteration 3/ 12 | consumed samples: 48 | elapsed time per iteration (ms): 557.3 | throughput per GPU (TFLOP/s/GPU): 598.3 | learning rate: 6.000000E-05 | global batch size: 16 | lm loss: 9.657482E+00 | loss scale: 1.0 | grad norm: 40.237 | number of skipped iterations: 0 | number of nan iterations: 0 |
975
+ Capture CUDA graph for training!!!
976
+ CUDA graph capture done!!!
977
+ [2026-08-30 05:53:08] iteration 4/ 12 | consumed samples: 64 | elapsed time per iteration (ms): 1679.9 | throughput per GPU (TFLOP/s/GPU): 198.5 | learning rate: 8.000000E-05 | global batch size: 16 | lm loss: 1.082655E+01 | loss scale: 1.0 | grad norm: 93.367 | number of skipped iterations: 0 | number of nan iterations: 0 |
978
+ [2026-08-30 05:53:09] iteration 5/ 12 | consumed samples: 80 | elapsed time per iteration (ms): 467.8 | throughput per GPU (TFLOP/s/GPU): 712.6 | learning rate: 1.000000E-04 | global batch size: 16 | lm loss: 1.065677E+01 | loss scale: 1.0 | grad norm: 16.818 | number of skipped iterations: 0 | number of nan iterations: 0 |
979
+ [2026-08-30 05:53:09] iteration 6/ 12 | consumed samples: 96 | elapsed time per iteration (ms): 464.7 | throughput per GPU (TFLOP/s/GPU): 717.4 | learning rate: 7.750000E-05 | global batch size: 16 | lm loss: 9.865295E+00 | loss scale: 1.0 | grad norm: 39.999 | number of skipped iterations: 0 | number of nan iterations: 0 |
980
+ [2026-08-30 05:53:10] iteration 7/ 12 | consumed samples: 112 | elapsed time per iteration (ms): 471.1 | throughput per GPU (TFLOP/s/GPU): 707.7 | learning rate: 3.250000E-05 | global batch size: 16 | lm loss: 9.299295E+00 | loss scale: 1.0 | grad norm: 7.166 | number of skipped iterations: 0 | number of nan iterations: 0 |
981
+ [2026-08-30 05:53:10] iteration 8/ 12 | consumed samples: 128 | elapsed time per iteration (ms): 466.9 | throughput per GPU (TFLOP/s/GPU): 714.0 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 9.030855E+00 | loss scale: 1.0 | grad norm: 24.694 | number of skipped iterations: 0 | number of nan iterations: 0 |
982
+ [2026-08-30 05:53:10] iteration 9/ 12 | consumed samples: 144 | elapsed time per iteration (ms): 464.8 | throughput per GPU (TFLOP/s/GPU): 717.3 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.672293E+00 | loss scale: 1.0 | grad norm: 5.930 | number of skipped iterations: 0 | number of nan iterations: 0 |
983
+ [2026-08-30 05:53:11] iteration 10/ 12 | consumed samples: 160 | elapsed time per iteration (ms): 460.5 | throughput per GPU (TFLOP/s/GPU): 724.0 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.545501E+00 | loss scale: 1.0 | grad norm: 5.250 | number of skipped iterations: 0 | number of nan iterations: 0 |
984
+ [2026-08-30 05:53:11] iteration 11/ 12 | consumed samples: 176 | elapsed time per iteration (ms): 470.3 | throughput per GPU (TFLOP/s/GPU): 708.9 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.451258E+00 | loss scale: 1.0 | grad norm: 4.910 | number of skipped iterations: 0 | number of nan iterations: 0 |
985
+ [2026-08-30 05:53:12] iteration 12/ 12 | consumed samples: 192 | elapsed time per iteration (ms): 465.5 | throughput per GPU (TFLOP/s/GPU): 716.1 | learning rate: 1.000000E-05 | global batch size: 16 | lm loss: 8.360296E+00 | loss scale: 1.0 | grad norm: 4.559 | number of skipped iterations: 0 | number of nan iterations: 0 |
986
+ [after training is done] datetime: 2026-08-30 05:53:12
987
+ W0830 05:53:12.454000 552219 torch/distributed/elastic/agent/server/api.py:719] Received 15 death signal, shutting down workers
988
+ W0830 05:53:12.455000 552219 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 552308 closing signal SIGTERM
989
+ W0830 05:53:12.456000 552219 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 552309 closing signal SIGTERM
990
+ W0830 05:53:12.457000 552219 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 552310 closing signal SIGTERM
991
+ W0830 05:53:12.458000 552219 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 552311 closing signal SIGTERM
992
+ W0830 05:53:12.459000 552219 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 552312 closing signal SIGTERM
993
+ W0830 05:53:12.462000 552219 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 552313 closing signal SIGTERM
994
+ W0830 05:53:12.463000 552219 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 552314 closing signal SIGTERM
995
+ W0830 05:53:12.466000 552219 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 552315 closing signal SIGTERM
996
+ Traceback (most recent call last):
997
+ File "/usr/local/bin/torchrun", line 8, in <module>
998
+ sys.exit(main())
999
+ ^^^^^^
1000
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/multiprocessing/errors/__init__.py", line 355, in wrapper
1001
+ return f(*args, **kwargs)
1002
+ ^^^^^^^^^^^^^^^^^^
1003
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/run.py", line 892, in main
1004
+ run(args)
1005
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/run.py", line 883, in run
1006
+ elastic_launch(
1007
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/launcher/api.py", line 139, in __call__
1008
+ return launch_agent(self._config, self._entrypoint, list(args))
1009
+ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1010
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/launcher/api.py", line 261, in launch_agent
1011
+ result = agent.run()
1012
+ ^^^^^^^^^^^
1013
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/metrics/api.py", line 138, in wrapper
1014
+ result = f(*args, **kwargs)
1015
+ ^^^^^^^^^^^^^^^^^^
1016
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/agent/server/api.py", line 711, in run
1017
+ result = self._invoke_run(role)
1018
+ ^^^^^^^^^^^^^^^^^^^^^^
1019
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/agent/server/api.py", line 870, in _invoke_run
1020
+ time.sleep(monitor_interval)
1021
+ File "/usr/local/lib/python3.12/dist-packages/torch/distributed/elastic/multiprocessing/api.py", line 84, in _terminate_process_handler
1022
+ raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval)
1023
+ torch.distributed.elastic.multiprocessing.api.SignalException: Process 552219 got signal: 15
tiera-rerun-fixed-moonep/echo/a1/a1_oneslot_gbs16/results.tsv ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ config backend policy slots seq gbs cap recompute rep status iter_ms_mean iter_ms_min peak_mib run_dir
2
+ G1 echo one 1 4096 16 2.40625 none 1 OK 467.30 463.50 161947 /workspace/tiera-rerun-fixed-moonep/echo/a1/a1_oneslot_gbs16/G1__echo__one__seq4096__gbs16__cap2p40625/rep1
3
+ G1 echo one 1 4096 16 2.40625 none 2 OK 466.99 460.60 161723 /workspace/tiera-rerun-fixed-moonep/echo/a1/a1_oneslot_gbs16/G1__echo__one__seq4096__gbs16__cap2p40625/rep2
4
+ G1 echo one 1 4096 16 2.40625 none 3 OK 466.74 460.50 161749 /workspace/tiera-rerun-fixed-moonep/echo/a1/a1_oneslot_gbs16/G1__echo__one__seq4096__gbs16__cap2p40625/rep3
tiera-rerun-fixed-moonep/echo/a1/manifest.txt ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ timestamp: 2026-08-30T05:49:31+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-echo
6
+ study_scripts_commit: 9925dcc
7
+ campaign: echo
8
+ backend: echo
9
+ phase: a1
10
+ gbs: 16
11
+ seq: 4096
12
+ configs: G1 G2 G3 G4
13
+ capacity: 2.40625
14
+ repeats: 3
15
+ nsys: 1
16
+ megatron_repo: /workspace/Megatron-MoE-Echo
17
+ megatron_branch_required: bracing-regions-through-kernels
18
+ megatron_branch_actual: bracing-regions-through-kernels
19
+ megatron_commit: 9a236793
tiera-rerun-fixed-moonep/echo/a1/pip_freeze.txt ADDED
@@ -0,0 +1,317 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ absl-py==2.3.0
2
+ aiohappyeyeballs @ file:///rapids/aiohappyeyeballs-2.6.1-py3-none-any.whl#sha256=f349ba8f4b75cb25c99c5c2d84e997e485204d2902a9597802b0371f09331fb8
3
+ aiohttp @ file:///rapids/aiohttp-3.12.7-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl#sha256=94f98e0e5a49f89b252e115844f756c04fc8050f38252a32a3dd994ce8121f10
4
+ aiosignal @ file:///rapids/aiosignal-1.3.2-py2.py3-none-any.whl#sha256=45cde58e409a301715980c2b01d0c28bdde3770d8290b5eb2173759d9acb31a5
5
+ annotated-types==0.7.0
6
+ anyio==4.9.0
7
+ apex @ file:///opt/pytorch/apex/dist/apex-0.1-cp312-cp312-linux_x86_64.whl#sha256=a1508e4a386362e9c8f46c92ccada9110e421212b964c694235d0fd5a224b170
8
+ argon2-cffi==25.1.0
9
+ argon2-cffi-bindings==21.2.0
10
+ arrow==1.3.0
11
+ asciitree @ file:///rapids/asciitree-0.3.3-py3-none-any.whl#sha256=3e507ccfaeb417823158091fdb9d26b0205f6a59ba680b479a71db65c73d01d4
12
+ asttokens==3.0.0
13
+ astunparse==1.6.3
14
+ async-lru==2.0.5
15
+ attrs==25.3.0
16
+ audioread==3.0.1
17
+ babel==2.17.0
18
+ beautifulsoup4==4.13.4
19
+ black==25.1.0
20
+ bleach==6.2.0
21
+ blis==0.7.11
22
+ cachetools @ file:///rapids/cachetools-6.0.0-py3-none-any.whl#sha256=82e73ba88f7b30228b5507dce1a1f878498fc669d972aef2dde4f3a3c24f103e
23
+ catalogue==2.0.10
24
+ certifi==2025.4.26
25
+ cffi==1.17.1
26
+ charset-normalizer==3.4.2
27
+ click==8.5.0
28
+ cloudpathlib==0.21.1
29
+ cloudpickle @ file:///rapids/cloudpickle-3.1.1-py3-none-any.whl#sha256=c8c5a44295039331ee9dad40ba100a9c7297b6f988e50e87ccdf3765a668350e
30
+ cmake==3.31.6
31
+ comm==0.2.2
32
+ confection==0.1.5
33
+ contourpy==1.3.2
34
+ cuda-bindings @ file:///rapids/cuda_bindings-12.9.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl#sha256=ff0e28d1e34758654b9c961e1f55e4786e49aee6a4dbceaf3cc24c46c672df7e
35
+ cuda-python @ file:///rapids/cuda_python-12.9.0-py3-none-any.whl#sha256=926acba49b2c0a0374c61b7c98f337c085199cf51cdfe4d6423c4129c20547a7
36
+ cudf @ file:///rapids/cudf-25.4.0-cp312-cp312-linux_x86_64.whl#sha256=b779a8e8c0e8577b9b560895958496da211bfac49cdc4f11a49e8d18cb82e348
37
+ cudf-polars @ file:///rapids/cudf_polars-25.4.0-py3-none-any.whl#sha256=227b6c5147e93ea86282c8f7219118f916769fa2ca0cdd931e97a148aa66b869
38
+ cugraph @ file:///rapids/cugraph-25.4.0-cp312-cp312-linux_x86_64.whl#sha256=b837c2dc2f7b235c702e4f8df64479b8fcbb342f848516d5faf0cdc4b5f31f29
39
+ cugraph-service-client @ file:///rapids/cugraph_service_client-25.4.0-py3-none-any.whl#sha256=7e568c3e41154b9752f5173555ffcc282ac52d72343feeb79653b2e2b1232470
40
+ cugraph-service-server @ file:///rapids/cugraph_service_server-25.4.0-py3-none-any.whl#sha256=a2830dcffafcccb16e12ebf85b98f8f247869d38c01ab4dd8664a1e7439d31be
41
+ cuml @ file:///rapids/cuml-25.4.0-cp312-cp312-linux_x86_64.whl#sha256=1e90a464a18b094e24797f1676688c5d7696c4cf97c934504214a53d5aa09aa5
42
+ cupy-cuda12x @ file:///rapids/cupy_cuda12x-13.3.0-cp312-cp312-linux_x86_64.whl#sha256=6a2dfa0aa1e8189a10d1534276c08f1d2f73b747fd28bfbe7b9f4ef66d9af73f
43
+ cuvs @ file:///rapids/cuvs-25.4.0-cp312-cp312-linux_x86_64.whl#sha256=b12f6b9f80e053a3b38542c2947d2516e468c02cad54a13d8b1a4a61d5aa7914
44
+ cycler==0.12.1
45
+ cymem==2.0.11
46
+ Cython==3.1.1
47
+ dask @ file:///rapids/dask-2025.2.0-py3-none-any.whl#sha256=f0fdeef6ceb0a06569d456c9e704f220f7f54e80f3a6ea42ab98cea6bc642b6e
48
+ dask-cuda @ file:///rapids/dask_cuda-25.4.0-py3-none-any.whl#sha256=442d87da0a95d00ca2a262b64c76e785a32ad905c82000fa2227932f81390e19
49
+ dask-cudf @ file:///rapids/dask_cudf-25.4.0-py3-none-any.whl#sha256=5e7cea0b26daea17da6e0bb457435e24460f76ba740358a3a18e239d6015c95c
50
+ datasets==5.0.1
51
+ debugpy==1.8.14
52
+ decorator==5.2.1
53
+ -e git+https://github.com/deepseek-ai/DeepEP.git@34152ae28f80bcc3ee38d7a12cb2ad87cfd4ea72#egg=deep_ep
54
+ defusedxml==0.7.1
55
+ dill==0.4.1
56
+ distributed @ file:///rapids/distributed-2025.2.0-py3-none-any.whl#sha256=368462084be47092d0a835f6c64c9b852da64dc88302b76d5f93bd5131228c67
57
+ distributed-ucxx @ file:///rapids/distributed_ucxx-0.43.0-py3-none-any.whl#sha256=94f60d0ccb0c94ba8dbfa1eaf455874d18aaab17e2ae218c38590e9fdc2d721d
58
+ dm-tree==0.1.9
59
+ einops==0.8.1
60
+ execnet==2.1.1
61
+ executing==2.2.0
62
+ expecttest==0.3.0
63
+ fasteners @ file:///rapids/fasteners-0.19-py3-none-any.whl#sha256=758819cb5d94cdedf4e836988b74de396ceacb8e2794d21f82d131fd9ee77237
64
+ fastjsonschema==2.21.1
65
+ fastrlock @ file:///rapids/fastrlock-0.8.3-cp312-cp312-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_28_x86_64.whl#sha256=5f13ec08f1adb1aa916c384b05ecb7dbebb8df9ea81abd045f60941c6283a670
66
+ filelock==3.18.0
67
+ flash_attn @ file:///opt/pytorch/flash_attn-2.7.4.post1-cp312-cp312-linux_x86_64.whl#sha256=a10455d6c56b387785b7413fd0bbcb392b42b3d4f4b88fda5d86cc8a9ce9182a
68
+ fonttools==4.58.1
69
+ fqdn==1.5.1
70
+ frozenlist @ file:///rapids/frozenlist-1.6.0-cp312-cp312-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl#sha256=4e1be9111cb6756868ac242b3c2bd1f09d9aea09846e4f5c23715e7afb647103
71
+ fsspec==2025.5.1
72
+ gast==0.6.0
73
+ grpcio==1.62.1
74
+ h11==0.16.0
75
+ hf-xet==1.6.0
76
+ httpcore==1.0.9
77
+ httpx==0.28.1
78
+ huggingface_hub==1.29.0
79
+ hypothesis==6.130.8
80
+ idna==3.10
81
+ importlib_metadata==8.7.0
82
+ iniconfig==2.1.0
83
+ intel-openmp==2021.4.0
84
+ ipykernel==6.29.5
85
+ ipython==9.3.0
86
+ ipython_pygments_lexers==1.1.1
87
+ isoduration==20.11.0
88
+ isort==6.0.1
89
+ jedi==0.19.2
90
+ Jinja2==3.1.6
91
+ joblib==1.5.1
92
+ json5==0.12.0
93
+ jsonpointer==3.0.0
94
+ jsonschema==4.24.0
95
+ jsonschema-specifications==2025.4.1
96
+ jupyter-events==0.12.0
97
+ jupyter-lsp==2.2.5
98
+ jupyter_client==8.6.3
99
+ jupyter_core==5.8.1
100
+ jupyter_server==2.16.0
101
+ jupyter_server_terminals==0.5.3
102
+ jupyterlab==4.4.3
103
+ jupyterlab_code_formatter==3.0.2
104
+ jupyterlab_pygments==0.3.0
105
+ jupyterlab_server==2.27.3
106
+ jupyterlab_tensorboard_pro @ file:///builder/jupyterlab_tensorboard_pro-4.0.0-py2.py3-none-any.whl#sha256=6bfcb5ced3358eb72150f4b2ab91312faef2f0ca14d845e8379d8fa52df6902e
107
+ jupytext==1.17.2
108
+ kiwisolver==1.4.8
109
+ kvikio @ file:///rapids/kvikio-25.4.0-cp312-cp312-linux_x86_64.whl#sha256=8b40fd8ea9e4a2974c04b50b7fc26b06094ac807f9a233281d0bafbaa71b1d85
110
+ langcodes==3.5.0
111
+ language_data==1.3.0
112
+ lazy_loader==0.4
113
+ libcudf @ file:///rapids/libcudf-25.4.0-py3-none-linux_x86_64.whl#sha256=238320b4958c1737bba2a609673e2539a78216b811b5498a68e00395d9b021a4
114
+ libcugraph @ file:///rapids/libcugraph-25.4.0-py3-none-linux_x86_64.whl#sha256=29085e901b2f7569ff2653bb1355fe6fad20f2fcaa585f5649da9867d8df766d
115
+ libcuml @ file:///rapids/libcuml-25.4.0-py3-none-linux_x86_64.whl#sha256=8e2e444c72b0e995e15a48da37ff12fb410a47a2baa69ba86ccc9b1a3db91e8c
116
+ libcuvs @ file:///rapids/libcuvs-25.4.0-py3-none-linux_x86_64.whl#sha256=e2f8db61c92b769c590dd752f43439ddd62b2f6592a3c337362b6c1ef220c3b9
117
+ libkvikio @ file:///rapids/libkvikio-25.4.0-py3-none-linux_x86_64.whl#sha256=da8a5713b05fc478bf40529a3e0fb61a2d605bb5a70ca4d461f96a6005dbe8b7
118
+ libraft @ file:///rapids/libraft-25.4.0-py3-none-linux_x86_64.whl#sha256=b921dec005aa4a321dd156776a18376feb47c96a5b67eec125c8a8e19edd1e24
119
+ librmm @ file:///rapids/librmm-25.4.0-py3-none-linux_x86_64.whl#sha256=ff951811bda729d4fa1109ab906225ae1f489a6ec2eeab5d1b674cc34bd4422f
120
+ librmm-cu12 @ file:///rapids/librmm_cu12-25.4.0-py3-none-any.whl#sha256=29296ef2b0e90679a0db280e86015be12d6ab9f4e62a181aa085f62dbadd06b3
121
+ librosa==0.11.0
122
+ libucxx @ file:///rapids/libucxx-0.43.0-py3-none-linux_x86_64.whl#sha256=f15849d7ac2fb68f0fd8998838a13321ddcfb0c594f7e6cbc401a3bcc4750566
123
+ lightning-thunder @ file:///opt/pytorch/lightning-thunder
124
+ lightning-utilities==0.14.3
125
+ lintrunner==0.12.7
126
+ llvmlite==0.42.0
127
+ locket @ file:///rapids/locket-1.0.0-py2.py3-none-any.whl#sha256=b6c819a722f7b6bd955b80781788e4a66a55628b858d347536b7e81325a3a5e3
128
+ looseversion==1.3.0
129
+ marisa-trie==1.2.1
130
+ Markdown==3.8
131
+ markdown-it-py==3.0.0
132
+ MarkupSafe==3.0.2
133
+ matplotlib==3.10.3
134
+ matplotlib-inline==0.1.7
135
+ mdit-py-plugins==0.4.2
136
+ mdurl==0.1.2
137
+ -e git+ssh://git@github.com/sriram-17-17/my-Megatron-MoE-Echo.git@9a23679390d7ab1fe462bc61fa31d90d393db49c#egg=megatron_core
138
+ mistune==3.1.3
139
+ mkl==2021.1.1
140
+ mkl-devel==2021.1.1
141
+ mkl-include==2021.1.1
142
+ ml_dtypes==0.6.0
143
+ mock==5.2.0
144
+ -e git+ssh://git@github.com/sriram-17-17/MoonEP.git@d8bc6667fad7177e6e09b637d773fa67f08bae93#egg=moonep
145
+ mpmath==1.3.0
146
+ msgpack==1.1.0
147
+ multidict @ file:///rapids/multidict-6.4.4-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl#sha256=a920f9cf2abdf6e493c519492d892c362007f113c94da4c239ae88429835bad1
148
+ multiprocess==0.70.19
149
+ murmurhash==1.0.13
150
+ mypy_extensions==1.1.0
151
+ nbclient==0.10.2
152
+ nbconvert==7.16.6
153
+ nbformat==5.10.4
154
+ nest-asyncio==1.6.0
155
+ networkx==3.5
156
+ ninja==1.11.1.4
157
+ notebook==7.4.3
158
+ notebook_shim==0.2.4
159
+ numba==0.59.1
160
+ numba-cuda @ file:///rapids/numba_cuda-0.4.0-py3-none-any.whl#sha256=5b95a5f42068771e37142944d44fe844fa87f1b2f89051461a4cb0af91bcc28a
161
+ numcodecs @ file:///rapids/numcodecs-0.13.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl#sha256=da2230484e6102e5fa3cc1a5dd37ca1f92dfbd183d91662074d6f7574e3e8f53
162
+ numpy==1.26.4
163
+ nvdlfw_inspect @ git+https://github.com/NVIDIA/nvidia-dlfw-inspect.git@6b60eb0e675606fb2fbbfcb12667ecbec75eaf87
164
+ nvfuser==0.2.27a0+9bf5aca
165
+ nvidia-cuda-cccl-cu12==12.9.27
166
+ nvidia-cudnn-frontend @ file:///opt/pytorch/lightning-thunder/tmp/cudnn_frontend
167
+ nvidia-cutlass-dsl==4.4.2
168
+ nvidia-cutlass-dsl-libs-base==4.4.2
169
+ nvidia-dali-cuda120==1.50.0
170
+ nvidia-ml-py @ file:///rapids/nvidia_ml_py-12.575.51-py3-none-any.whl#sha256=eb8641800d98ce40a22f479873f34b482e214a7e80349c63be51c3919845446e
171
+ nvidia-modelopt==0.29.0
172
+ nvidia-modelopt-core==0.29.0
173
+ nvidia-nvcomp-cu12==4.2.0.14
174
+ nvidia-nvimgcodec-cu12==0.5.0.13
175
+ nvidia-nvjpeg-cu12==12.4.0.16
176
+ nvidia-nvjpeg2k-cu12==0.8.1.40
177
+ nvidia-nvshmem-cu12==3.7.2
178
+ nvidia-nvtiff-cu12==0.5.0.67
179
+ nvidia-resiliency-ext==0.4.0
180
+ nvtx @ file:///rapids/nvtx-0.2.11-cp312-cp312-linux_x86_64.whl#sha256=73521e37cba42b4db0b3cd68fc5eb814e55c500a61eeed6edcdf6746315ac53e
181
+ nx-cugraph @ file:///rapids/nx_cugraph-25.4.0-py3-none-any.whl#sha256=e625444486cc899a77bd48e7da76a2edd95c0006dd95fbf323745feae7f33111
182
+ onnx @ file:///opt/pytorch/pytorch/third_party/onnx
183
+ onnx-ir==1.0.0
184
+ onnxscript==0.7.1
185
+ opt_einsum==3.4.0
186
+ optree==0.16.0
187
+ overrides==7.7.0
188
+ packaging==23.2
189
+ pandas @ file:///rapids/pandas-2.2.3-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl#sha256=fffb8ae78d8af97f849404f21411c95062db1496aeb3e56f146f0355c9989319
190
+ pandocfilters==1.5.1
191
+ parso==0.8.4
192
+ partd @ file:///rapids/partd-1.4.2-py3-none-any.whl#sha256=978e4ac767ec4ba5b86c6eaa52e5a2a3bc748a2ca839e8cc798f1cc6ce6efb0f
193
+ pathspec==0.12.1
194
+ pexpect==4.9.0
195
+ pillow==11.2.1
196
+ platformdirs==4.3.8
197
+ pluggy==1.6.0
198
+ ply @ file:///rapids/ply-3.11-py2.py3-none-any.whl#sha256=096f9b8350b65ebd2fd1346b12452efe5b9607f7482813ffca50c22722a807ce
199
+ polars @ file:///rapids/polars-1.25.2-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl#sha256=f7fcbb4f476784384ccda48757fca4e8c2e2c5a0a3aef3717aaf56aee4e30e09
200
+ polygraphy==0.49.20
201
+ pooch==1.8.2
202
+ preshed==3.0.10
203
+ prometheus_client==0.22.1
204
+ prompt_toolkit==3.0.51
205
+ propcache @ file:///rapids/propcache-0.3.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl#sha256=563f9d8c03ad645597b8d010ef4e9eab359faeb11a0a2ac9f7b4bc8c28ebef25
206
+ protobuf==4.24.4
207
+ psutil==7.0.0
208
+ ptyprocess==0.7.0
209
+ PuLP==3.2.1
210
+ pure_eval==0.2.3
211
+ pyarrow==25.0.1
212
+ pybind11==2.13.6
213
+ pybind11_global==2.13.6
214
+ pycocotools @ git+https://github.com/nvidia/cocoapi.git@08350a157b41f0755a536441b5abec89ab934c5b#subdirectory=PythonAPI
215
+ pycparser==2.22
216
+ pydantic==2.11.5
217
+ pydantic_core==2.33.2
218
+ Pygments==2.19.1
219
+ pylibcudf @ file:///rapids/pylibcudf-25.4.0-cp312-cp312-linux_x86_64.whl#sha256=401b8bfd022d06f84e120779ca7c3d01a28b68762aaa8f7873b10d6b1975cf6a
220
+ pylibcugraph @ file:///rapids/pylibcugraph-25.4.0-cp312-cp312-linux_x86_64.whl#sha256=061c7389a88c815b54fd5d3a3584fb87bf693e5602f5be335260d0ba575ba557
221
+ pylibcugraphops @ file:///rapids/pylibcugraphops-25.4.0-cp312-cp312-linux_x86_64.whl#sha256=a25ef4b22166c80718d5c7e5bddb755e81a2fe340fa5becd0ac36ec3edfb208a
222
+ pylibraft @ file:///rapids/pylibraft-25.4.0-cp312-cp312-linux_x86_64.whl#sha256=f2ecd23fe51f9c2caa87bbbe85cb32fcaff4ecfb52ceb9a9991bc50f82782c2e
223
+ pylibwholegraph @ file:///rapids/pylibwholegraph-25.4.0-cp312-cp312-linux_x86_64.whl#sha256=b82913d1e38bda76d8c272fbd8cc9072d987209d3d47cc93a464a37c2925aca2
224
+ pynvjitlink @ file:///rapids/pynvjitlink-0.3.0-cp312-cp312-linux_x86_64.whl#sha256=066caf1314730643d3b48c9847bd65789defdd6a7c78993176eb3d8bbe3f4919
225
+ pynvml @ file:///rapids/pynvml-12.0.0-py3-none-any.whl#sha256=fdff84b62a27dbe98e08e1a647eb77342bef1aebe0878bcd15e99a83fcbecb9e
226
+ pyparsing==3.2.3
227
+ pytest==8.1.1
228
+ pytest-flakefinder==1.1.0
229
+ pytest-rerunfailures==15.1
230
+ pytest-shard==0.1.2
231
+ pytest-xdist==3.7.0
232
+ python-dateutil==2.9.0.post0
233
+ python-hostlist==2.2.1
234
+ python-json-logger==3.3.0
235
+ pytorch-triton @ file:///tmp/dist/pytorch_triton-3.3.0%2Bgit96316ce52.nvinternal-cp312-cp312-linux_x86_64.whl#sha256=6d6921772b9dfd0243c055e0d054de42a113e92cabcfdd1d3e80a5ad9c08cf7d
236
+ pytz @ file:///rapids/pytz-2023.4-py2.py3-none-any.whl#sha256=f90ef520d95e7c46951105338d918664ebfd6f1d995bd7d153127ce90efafa6a
237
+ PyYAML==6.0.2
238
+ pyzmq==26.4.0
239
+ raft-dask @ file:///rapids/raft_dask-25.4.0-cp312-cp312-linux_x86_64.whl#sha256=36cf37ba6ce35454c4729c2cb597a360f22780eefb11d514a16a07137da5eb6c
240
+ rapids-dask-dependency @ file:///rapids/rapids_dask_dependency-25.4.0a0-py3-none-any.whl#sha256=484faca626e7760112a7d8ad3426f5bdd720b609006d8dca87c2607076a348bd
241
+ rapids-logger @ file:///rapids/rapids_logger-0.1.18-py3-none-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl#sha256=3617b2d86291ac6cb957542895f20fc9737ace29773ca90337e6fb04c73bd120
242
+ referencing==0.36.2
243
+ regex==2024.11.6
244
+ requests==2.32.3
245
+ rfc3339-validator==0.1.4
246
+ rfc3986-validator==0.1.1
247
+ rich==14.0.0
248
+ rmm @ file:///rapids/rmm-25.4.0-cp312-cp312-linux_x86_64.whl#sha256=1824bcab3836296666980a997ddcb9c4bca24dfb873012cfa0a7ecea583ca712
249
+ rpds-py==0.25.1
250
+ safetensors==0.5.3
251
+ scikit-learn==1.6.1
252
+ scipy==1.15.3
253
+ Send2Trash==1.8.3
254
+ setuptools==78.1.1
255
+ shellingham==1.5.4
256
+ six==1.16.0
257
+ smart-open==7.1.0
258
+ sniffio==1.3.1
259
+ sortedcontainers==2.4.0
260
+ soundfile==0.13.1
261
+ soupsieve==2.7
262
+ soxr==0.5.0.post1
263
+ spacy==3.7.5
264
+ spacy-legacy==3.0.12
265
+ spacy-loggers==1.0.5
266
+ srsly==2.5.1
267
+ stack-data==0.6.3
268
+ sympy==1.14.0
269
+ tabulate==0.9.0
270
+ tbb==2021.13.1
271
+ tblib @ file:///rapids/tblib-3.1.0-py3-none-any.whl#sha256=670bb4582578134b3d81a84afa1b016128b429f3d48e6cbbaecc9d15675e984e
272
+ tensorboard==2.16.2
273
+ tensorboard-data-server==0.7.2
274
+ tensorrt @ file:///workspace/TensorRT-10.11.0.33/python/tensorrt-10.11.0.33-cp312-none-linux_x86_64.whl#sha256=c2f18f72758a3d54a71901495c8fda9e7094a8bcdd2a5b6b58967fbf05733b3a
275
+ terminado==0.18.1
276
+ thinc==8.2.5
277
+ threadpoolctl==3.6.0
278
+ thriftpy2 @ file:///rapids/thriftpy2-0.5.2-cp312-cp312-linux_x86_64.whl#sha256=c0c7d939bdc5ccf7739154959938445d5fdf72f1d6e12bd563947a36f67e2147
279
+ tinycss2==1.4.0
280
+ toolz @ file:///rapids/toolz-1.0.0-py3-none-any.whl#sha256=292c8f1c4e7516bf9086f8850935c799a874039c8bcf959d47b600e4c44a6236
281
+ torch @ file:///opt/transfer/torch-2.8.0a0%2B5228986c39.nv25.6-29946474-cp312-cp312-linux_x86_64.whl#sha256=7a3c9ff5288d8102a31296b3d62491f3f47c8820a5c9aaefc7dc0fe20e1e8ad5
282
+ torch_tensorrt @ file:///opt/pytorch/torch_tensorrt/dist/torch_tensorrt-2.8.0a0-cp312-cp312-linux_x86_64.whl#sha256=26f4d7a04e7e6e017a5b917b8b98b83720bb2bb0195f48d25f52d567d286bfe1
283
+ torchao @ file:///opt/pytorch/ao
284
+ torchprofile==0.0.4
285
+ torchvision @ file:///opt/transfer/torchvision-0.22.0a0%2B95f10a4e-29946474-cp312-cp312-manylinux_2_28_x86_64.whl#sha256=746a3c674cc61680a86f4a14b374b21af1b5c21f629c42765b40c9494f6b234d
286
+ tornado==6.5.1
287
+ tqdm==4.67.1
288
+ traitlets==5.14.3
289
+ transformer_engine @ file:///pvcdata/aiinfra/b-aghosh/wheels/transformer_engine-2.10.0.dev0%2B10f6128d-cp312-cp312-linux_x86_64.whl#sha256=35cf5f4b83b2baad864e12e512bb422e0929409ac05610812fa521d4942d0189
290
+ treelite @ file:///rapids/treelite-4.4.1-py3-none-linux_x86_64.whl#sha256=174741724205540a1a5b37da81483aeff55d89548f42b3461cf07acac7ae96eb
291
+ typer==0.16.0
292
+ types-dataclasses==0.6.6
293
+ types-python-dateutil==2.9.0.20250516
294
+ typing-inspection==0.4.1
295
+ typing_extensions==4.14.0
296
+ tzdata @ file:///rapids/tzdata-2025.2-py2.py3-none-any.whl#sha256=1a403fada01ff9221ca8044d701868fa132215d84beb92242d9acd2147f667a8
297
+ ucx-py @ file:///rapids/ucx_py-0.43.0-cp312-cp312-linux_x86_64.whl#sha256=4a452fd8d8c983283ada3d49086e3657dc2ee94fbad08efdd708d15ab6e0b73a
298
+ ucxx @ file:///rapids/ucxx-0.43.0-cp312-cp312-linux_x86_64.whl#sha256=f1b4e55cb45c66fe88a2d8f9c2b87751165f3cfec21fbdc9c47c4ddff75e8a5f
299
+ -e git+https://github.com/Dots-Infra/UltraEP.git@94cab099b44fffa99a82fea99e7c12d89cf65e4f#egg=ultra_ep
300
+ uri-template==1.3.0
301
+ urllib3==2.0.7
302
+ wasabi==1.1.3
303
+ wcwidth==0.2.13
304
+ weasel==0.4.1
305
+ webcolors==24.11.1
306
+ webencodings==0.5.1
307
+ websocket-client==1.8.0
308
+ Werkzeug==3.1.3
309
+ wheel==0.45.1
310
+ wrapt==1.17.2
311
+ xdoctest==1.0.2
312
+ xgboost @ file:///rapids/xgboost-2.1.4-py3-none-manylinux_2_39_x86_64.whl#sha256=25eeca27f4048fc491940b3a0b4b5782ee48c33cb3aec059cef71f8da542930b
313
+ xxhash==4.0.1
314
+ yarl @ file:///rapids/yarl-1.20.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl#sha256=bf099e2432131093cc611623e0b0bcc399b8cddd9a91eded8bfb50402ec35018
315
+ zarr @ file:///rapids/zarr-2.18.7-py3-none-any.whl#sha256=ac3dc4033e9ae4e9d7b5e27c97ea3eaf1003cc0a07f010bd83d5134bf8c4b223
316
+ zict @ file:///rapids/zict-3.0.0-py2.py3-none-any.whl#sha256=5796e36bd0e0cc8cf0fbc1ace6a68912611c1dbd74750a3f3026b9b9d6a327ae
317
+ zipp==3.22.0
tiera-rerun-fixed-moonep/echo/a1/provenance.txt CHANGED
@@ -27,3 +27,64 @@ command : run_ladder.sh --phase a1 --gbs 16 --out /workspace/tiera-rerun-f
27
  7, 120 MHz, 1965 MHz, 1000.00 W, 28
28
 
29
  NVLink topology:
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
27
  7, 120 MHz, 1965 MHz, 1000.00 W, 28
28
 
29
  NVLink topology:
30
+ GPU0 GPU1 GPU2 GPU3 GPU4 GPU5 GPU6 GPU7 NIC0 NIC1 NIC2 NIC3 NIC4 NIC5 NIC6 NIC7 NIC8 CPU Affinity NUMA Affinity GPU NUMA ID
31
+ GPU0 X NV18 NV18 NV18 NV18 NV18 NV18 NV18 PIX NODE SYS SYS SYS SYS SYS SYS NODE 0-27 0 N/A
32
+ GPU1 NV18 X NV18 NV18 NV18 NV18 NV18 NV18 NODE PIX SYS SYS SYS SYS SYS SYS NODE 0-27 0 N/A
33
+ GPU2 NV18 NV18 X NV18 NV18 NV18 NV18 NV18 SYS SYS PIX NODE SYS SYS SYS SYS SYS 28-55 1 N/A
34
+ GPU3 NV18 NV18 NV18 X NV18 NV18 NV18 NV18 SYS SYS NODE PIX SYS SYS SYS SYS SYS 28-55 1 N/A
35
+ GPU4 NV18 NV18 NV18 NV18 X NV18 NV18 NV18 SYS SYS SYS SYS PIX NODE SYS SYS SYS 56-83 2 N/A
36
+ GPU5 NV18 NV18 NV18 NV18 NV18 X NV18 NV18 SYS SYS SYS SYS NODE PIX SYS SYS SYS 56-83 2 N/A
37
+ GPU6 NV18 NV18 NV18 NV18 NV18 NV18 X NV18 SYS SYS SYS SYS SYS SYS PIX NODE SYS 84-111 3 N/A
38
+ GPU7 NV18 NV18 NV18 NV18 NV18 NV18 NV18 X SYS SYS SYS SYS SYS SYS NODE PIX SYS 84-111 3 N/A
39
+ NIC0 PIX NODE SYS SYS SYS SYS SYS SYS X NODE SYS SYS SYS SYS SYS SYS NODE
40
+ NIC1 NODE PIX SYS SYS SYS SYS SYS SYS NODE X SYS SYS SYS SYS SYS SYS NODE
41
+ NIC2 SYS SYS PIX NODE SYS SYS SYS SYS SYS SYS X NODE SYS SYS SYS SYS SYS
42
+ NIC3 SYS SYS NODE PIX SYS SYS SYS SYS SYS SYS NODE X SYS SYS SYS SYS SYS
43
+ NIC4 SYS SYS SYS SYS PIX NODE SYS SYS SYS SYS SYS SYS X NODE SYS SYS SYS
44
+ unavailable
45
+
46
+ ---- toolchain -----------------------------------------------
47
+ nvcc : release 12.9, V12.9.86
48
+ gcc : gcc (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0
49
+ python : 3.12.3
50
+ torch : 2.8.0a0+5228986c39.nv25.06 cuda=12.9 nccl=2.27.3
51
+ torch arch : ['sm_75', 'sm_80', 'sm_86', 'sm_90', 'sm_100', 'sm_120', 'compute_120']
52
+ transformer_engine: 2.10.0.dev0+10f6128d [/usr/local/lib/python3.12/dist-packages/transformer_engine/__init__.py]
53
+ deep_ep : 1.2.1+34152ae [/workspace/DeepEP/deep_ep/__init__.py]
54
+ ultra_ep : 1.0.0 [/workspace/UltraEP/ultra_ep/__init__.py]
55
+ moonep : 0.0.1 [/workspace/MoonEP/moonep/__init__.py]
56
+ megatron : 0.14.0rc5 [None]
57
+ flash_attn : 2.7.4.post1 [/usr/local/lib/python3.12/dist-packages/flash_attn/__init__.py]
58
+ numpy : 1.26.4 [/usr/local/lib/python3.12/dist-packages/numpy/__init__.py]
59
+ huggingface_hub: 1.29.0 [/usr/local/lib/python3.12/dist-packages/huggingface_hub/__init__.py]
60
+
61
+ ---- repositories --------------------------------------------
62
+ Megatron-MoE-Echo bracing-regions-through-kernels 9a23679390d7
63
+ Megatron-UltraEP ultraep-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-echo 9925dccb2715
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/hf_watcher.log CHANGED
@@ -12,3 +12,6 @@ https://huggingface.co/MoE-CUDA-Graph-Analysis/tierb-b200-gbs8/tree/main
12
  Already logged in to Hugging Face as: AbhishekGhosh1998
13
  Watching /workspace/tiera-rerun-fixed-moonep -> HF every 600s (kill to stop)
14
  [2026-08-30T05:49:17+00:00] backup pass
 
 
 
 
12
  Already logged in to Hugging Face as: AbhishekGhosh1998
13
  Watching /workspace/tiera-rerun-fixed-moonep -> HF every 600s (kill to stop)
14
  [2026-08-30T05:49:17+00:00] backup pass
15
+ uploaded: /workspace/tiera-rerun-fixed-moonep -> MoE-CUDA-Graph-Analysis/tierb-b200-gbs8/tiera-rerun-fixed-moonep
16
+ https://huggingface.co/MoE-CUDA-Graph-Analysis/tierb-b200-gbs8/tree/main
17
+ [2026-08-30T05:59:19+00:00] backup pass
tiera-rerun-fixed-moonep/study.log CHANGED
@@ -394,3 +394,19 @@ driver_repo: /workspace/blackwell_run_scripts @ 9925dcc
394
  === STAGE 3 -- Tier A (the causal experiment)
395
 
396
  === STEP echo__a1__gbs16
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
394
  === STAGE 3 -- Tier A (the causal experiment)
395
 
396
  === STEP echo__a1__gbs16
397
+ ==============================================================
398
+ CAMPAIGN: echo backend=echo phase=a1
399
+ Megatron: /workspace/Megatron-MoE-Echo @ bracing-regions-through-kernels
400
+ GBS=16 seq=4096 repeats=3 nsys=1
401
+ results -> /workspace/tiera-rerun-fixed-moonep/echo/a1
402
+ ==============================================================
403
+ --------------------------------------------------------------------------
404
+ [05:49:31] CELL G1__echo__one__seq4096__gbs16__cap2p40625 (slots=1)
405
+ [guard] iter 12/12 (70s)
406
+ [guard] tearing down process group (outcome=DONE)
407
+ [guard] iter 12/12 (70s)
408
+ [guard] tearing down process group (outcome=DONE)
409
+ [guard] iter 12/12 (70s)
410
+ [guard] tearing down process group (outcome=DONE)
411
+ [05:53:17] tracing G1__echo__one__seq4096__gbs16__cap2p40625 (ranks: 0 1 2 3 4 5 6 7)
412
+ [guard] iter 8/10 (80s)