Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
Community
611
dacorvo HF Staff commited on
Commit
f42ad81
·
verified ·
1 Parent(s): 08e7e42

Synchronizing local compiler cache.

Browse files
Files changed (33) hide show
  1. .gitattributes +6 -0
  2. neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.1.dev0/qwen3/Qwen/Qwen3-32B/1ae184e7b95dac097684.json +123 -0
  3. neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.1.dev0/qwen3/Qwen/Qwen3-32B/2bb6e5edef9a6e488094.json +123 -0
  4. neuronxcc-2.21.18209.0+043b1bf7/MODULE_00d39c882e9e437ead19+747527b0/compile_flags.json +1 -0
  5. neuronxcc-2.21.18209.0+043b1bf7/MODULE_00d39c882e9e437ead19+747527b0/model.hlo_module.pb +3 -0
  6. neuronxcc-2.21.18209.0+043b1bf7/MODULE_00d39c882e9e437ead19+747527b0/model.log +2 -0
  7. neuronxcc-2.21.18209.0+043b1bf7/MODULE_0a59544d8a22ab5b06a0+747527b0/compile_flags.json +1 -0
  8. neuronxcc-2.21.18209.0+043b1bf7/MODULE_0a59544d8a22ab5b06a0+747527b0/model.hlo_module.pb +3 -0
  9. neuronxcc-2.21.18209.0+043b1bf7/MODULE_0a59544d8a22ab5b06a0+747527b0/model.log +2 -0
  10. neuronxcc-2.21.18209.0+043b1bf7/MODULE_341e3918a26a3758bc34+877608f3/compile_flags.json +1 -0
  11. neuronxcc-2.21.18209.0+043b1bf7/MODULE_341e3918a26a3758bc34+877608f3/model.done +0 -0
  12. neuronxcc-2.21.18209.0+043b1bf7/MODULE_341e3918a26a3758bc34+877608f3/model.hlo_module.pb +3 -0
  13. neuronxcc-2.21.18209.0+043b1bf7/MODULE_341e3918a26a3758bc34+877608f3/model.neff +3 -0
  14. neuronxcc-2.21.18209.0+043b1bf7/MODULE_3c45b7d14d093b6eb821+747527b0/compile_flags.json +1 -0
  15. neuronxcc-2.21.18209.0+043b1bf7/MODULE_3c45b7d14d093b6eb821+747527b0/model.hlo_module.pb +3 -0
  16. neuronxcc-2.21.18209.0+043b1bf7/MODULE_3c45b7d14d093b6eb821+747527b0/model.log +2 -0
  17. neuronxcc-2.21.18209.0+043b1bf7/MODULE_5ed38095f803237b1507+877608f3/compile_flags.json +1 -0
  18. neuronxcc-2.21.18209.0+043b1bf7/MODULE_5ed38095f803237b1507+877608f3/model.done +0 -0
  19. neuronxcc-2.21.18209.0+043b1bf7/MODULE_5ed38095f803237b1507+877608f3/model.hlo_module.pb +3 -0
  20. neuronxcc-2.21.18209.0+043b1bf7/MODULE_5ed38095f803237b1507+877608f3/model.neff +3 -0
  21. neuronxcc-2.21.18209.0+043b1bf7/MODULE_6b32493175a70d9e16b9+747527b0/compile_flags.json +1 -0
  22. neuronxcc-2.21.18209.0+043b1bf7/MODULE_6b32493175a70d9e16b9+747527b0/model.done +0 -0
  23. neuronxcc-2.21.18209.0+043b1bf7/MODULE_6b32493175a70d9e16b9+747527b0/model.hlo_module.pb +3 -0
  24. neuronxcc-2.21.18209.0+043b1bf7/MODULE_6b32493175a70d9e16b9+747527b0/model.neff +3 -0
  25. neuronxcc-2.21.18209.0+043b1bf7/MODULE_6b32493175a70d9e16b9+747527b0/wrapped_neff.hlo +3 -0
  26. neuronxcc-2.21.18209.0+043b1bf7/MODULE_eb50383f9a879a8375b9+747527b0/compile_flags.json +1 -0
  27. neuronxcc-2.21.18209.0+043b1bf7/MODULE_eb50383f9a879a8375b9+747527b0/model.done +0 -0
  28. neuronxcc-2.21.18209.0+043b1bf7/MODULE_eb50383f9a879a8375b9+747527b0/model.hlo_module.pb +3 -0
  29. neuronxcc-2.21.18209.0+043b1bf7/MODULE_eb50383f9a879a8375b9+747527b0/model.neff +3 -0
  30. neuronxcc-2.21.18209.0+043b1bf7/MODULE_eb50383f9a879a8375b9+747527b0/wrapped_neff.hlo +3 -0
  31. neuronxcc-2.21.18209.0+043b1bf7/MODULE_ffe3bb712e8a45b20f93+747527b0/compile_flags.json +1 -0
  32. neuronxcc-2.21.18209.0+043b1bf7/MODULE_ffe3bb712e8a45b20f93+747527b0/model.hlo_module.pb +3 -0
  33. neuronxcc-2.21.18209.0+043b1bf7/MODULE_ffe3bb712e8a45b20f93+747527b0/model.log +2 -0
.gitattributes CHANGED
@@ -11914,3 +11914,9 @@ neuronxcc-2.21.18209.0+043b1bf7/MODULE_c3aad406cc0df2c16f36+747527b0/wrapped_nef
11914
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_74c11479e8406b0ab839+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11915
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_74c11479e8406b0ab839+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
11916
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_ef2fb26b7dd6c17c8573+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
11914
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_74c11479e8406b0ab839+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11915
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_74c11479e8406b0ab839+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
11916
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_ef2fb26b7dd6c17c8573+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11917
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_341e3918a26a3758bc34+877608f3/model.neff filter=lfs diff=lfs merge=lfs -text
11918
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_5ed38095f803237b1507+877608f3/model.neff filter=lfs diff=lfs merge=lfs -text
11919
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_6b32493175a70d9e16b9+747527b0/model.neff filter=lfs diff=lfs merge=lfs -text
11920
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_6b32493175a70d9e16b9+747527b0/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
11921
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_eb50383f9a879a8375b9+747527b0/model.neff filter=lfs diff=lfs merge=lfs -text
11922
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_eb50383f9a879a8375b9+747527b0/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.1.dev0/qwen3/Qwen/Qwen3-32B/1ae184e7b95dac097684.json ADDED
@@ -0,0 +1,123 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen3-32B",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Qwen3ForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "head_dim": 128,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 5120,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 25600,
15
+ "layer_types": [
16
+ "full_attention",
17
+ "full_attention",
18
+ "full_attention",
19
+ "full_attention",
20
+ "full_attention",
21
+ "full_attention",
22
+ "full_attention",
23
+ "full_attention",
24
+ "full_attention",
25
+ "full_attention",
26
+ "full_attention",
27
+ "full_attention",
28
+ "full_attention",
29
+ "full_attention",
30
+ "full_attention",
31
+ "full_attention",
32
+ "full_attention",
33
+ "full_attention",
34
+ "full_attention",
35
+ "full_attention",
36
+ "full_attention",
37
+ "full_attention",
38
+ "full_attention",
39
+ "full_attention",
40
+ "full_attention",
41
+ "full_attention",
42
+ "full_attention",
43
+ "full_attention",
44
+ "full_attention",
45
+ "full_attention",
46
+ "full_attention",
47
+ "full_attention",
48
+ "full_attention",
49
+ "full_attention",
50
+ "full_attention",
51
+ "full_attention",
52
+ "full_attention",
53
+ "full_attention",
54
+ "full_attention",
55
+ "full_attention",
56
+ "full_attention",
57
+ "full_attention",
58
+ "full_attention",
59
+ "full_attention",
60
+ "full_attention",
61
+ "full_attention",
62
+ "full_attention",
63
+ "full_attention",
64
+ "full_attention",
65
+ "full_attention",
66
+ "full_attention",
67
+ "full_attention",
68
+ "full_attention",
69
+ "full_attention",
70
+ "full_attention",
71
+ "full_attention",
72
+ "full_attention",
73
+ "full_attention",
74
+ "full_attention",
75
+ "full_attention",
76
+ "full_attention",
77
+ "full_attention",
78
+ "full_attention",
79
+ "full_attention"
80
+ ],
81
+ "max_position_embeddings": 40960,
82
+ "max_window_layers": 64,
83
+ "model_type": "qwen3",
84
+ "neuron": {
85
+ "_serialized_key": "NxDNeuronConfig",
86
+ "batch_size": 1,
87
+ "capacity_factor": null,
88
+ "checkpoint_id": "Qwen/Qwen3-32B",
89
+ "checkpoint_revision": "9216db5781bf21249d130ec9da846c4624c16137",
90
+ "continuous_batching": false,
91
+ "enable_bucketing": false,
92
+ "ep_degree": 1,
93
+ "fused_qkv": true,
94
+ "glu_mlp": true,
95
+ "local_ranks_size": 4,
96
+ "max_batch_size": 1,
97
+ "max_context_length": 4096,
98
+ "max_topk": 256,
99
+ "n_active_tokens": 4096,
100
+ "neuronxcc_version": "2.21.18209.0+043b1bf7",
101
+ "on_device_sampling": true,
102
+ "optimum_neuron_version": "0.4.1.dev0",
103
+ "output_logits": false,
104
+ "pp_degree": 1,
105
+ "sequence_length": 4096,
106
+ "speculation_length": 0,
107
+ "start_rank_id": 0,
108
+ "target": "trn2",
109
+ "torch_dtype": "bfloat16",
110
+ "tp_degree": 4
111
+ },
112
+ "num_attention_heads": 64,
113
+ "num_hidden_layers": 64,
114
+ "num_key_value_heads": 8,
115
+ "rms_norm_eps": 1e-06,
116
+ "rope_scaling": null,
117
+ "rope_theta": 1000000,
118
+ "sliding_window": null,
119
+ "tie_word_embeddings": false,
120
+ "use_cache": true,
121
+ "use_sliding_window": false,
122
+ "vocab_size": 151936
123
+ }
neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.1.dev0/qwen3/Qwen/Qwen3-32B/2bb6e5edef9a6e488094.json ADDED
@@ -0,0 +1,123 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen3-32B",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Qwen3ForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "head_dim": 128,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 5120,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 25600,
15
+ "layer_types": [
16
+ "full_attention",
17
+ "full_attention",
18
+ "full_attention",
19
+ "full_attention",
20
+ "full_attention",
21
+ "full_attention",
22
+ "full_attention",
23
+ "full_attention",
24
+ "full_attention",
25
+ "full_attention",
26
+ "full_attention",
27
+ "full_attention",
28
+ "full_attention",
29
+ "full_attention",
30
+ "full_attention",
31
+ "full_attention",
32
+ "full_attention",
33
+ "full_attention",
34
+ "full_attention",
35
+ "full_attention",
36
+ "full_attention",
37
+ "full_attention",
38
+ "full_attention",
39
+ "full_attention",
40
+ "full_attention",
41
+ "full_attention",
42
+ "full_attention",
43
+ "full_attention",
44
+ "full_attention",
45
+ "full_attention",
46
+ "full_attention",
47
+ "full_attention",
48
+ "full_attention",
49
+ "full_attention",
50
+ "full_attention",
51
+ "full_attention",
52
+ "full_attention",
53
+ "full_attention",
54
+ "full_attention",
55
+ "full_attention",
56
+ "full_attention",
57
+ "full_attention",
58
+ "full_attention",
59
+ "full_attention",
60
+ "full_attention",
61
+ "full_attention",
62
+ "full_attention",
63
+ "full_attention",
64
+ "full_attention",
65
+ "full_attention",
66
+ "full_attention",
67
+ "full_attention",
68
+ "full_attention",
69
+ "full_attention",
70
+ "full_attention",
71
+ "full_attention",
72
+ "full_attention",
73
+ "full_attention",
74
+ "full_attention",
75
+ "full_attention",
76
+ "full_attention",
77
+ "full_attention",
78
+ "full_attention",
79
+ "full_attention"
80
+ ],
81
+ "max_position_embeddings": 40960,
82
+ "max_window_layers": 64,
83
+ "model_type": "qwen3",
84
+ "neuron": {
85
+ "_serialized_key": "NxDNeuronConfig",
86
+ "batch_size": 2,
87
+ "capacity_factor": null,
88
+ "checkpoint_id": "Qwen/Qwen3-32B",
89
+ "checkpoint_revision": "9216db5781bf21249d130ec9da846c4624c16137",
90
+ "continuous_batching": true,
91
+ "enable_bucketing": false,
92
+ "ep_degree": 1,
93
+ "fused_qkv": true,
94
+ "glu_mlp": true,
95
+ "local_ranks_size": 4,
96
+ "max_batch_size": 2,
97
+ "max_context_length": 4096,
98
+ "max_topk": 256,
99
+ "n_active_tokens": 4096,
100
+ "neuronxcc_version": "2.21.18209.0+043b1bf7",
101
+ "on_device_sampling": true,
102
+ "optimum_neuron_version": "0.4.1.dev0",
103
+ "output_logits": false,
104
+ "pp_degree": 1,
105
+ "sequence_length": 4096,
106
+ "speculation_length": 0,
107
+ "start_rank_id": 0,
108
+ "target": "trn2",
109
+ "torch_dtype": "bfloat16",
110
+ "tp_degree": 4
111
+ },
112
+ "num_attention_heads": 64,
113
+ "num_hidden_layers": 64,
114
+ "num_key_value_heads": 8,
115
+ "rms_norm_eps": 1e-06,
116
+ "rope_scaling": null,
117
+ "rope_theta": 1000000,
118
+ "sliding_window": null,
119
+ "tie_word_embeddings": false,
120
+ "use_cache": true,
121
+ "use_sliding_window": false,
122
+ "vocab_size": 151936
123
+ }
neuronxcc-2.21.18209.0+043b1bf7/MODULE_00d39c882e9e437ead19+747527b0/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn2", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=2", "--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.21.18209.0+043b1bf7/MODULE_00d39c882e9e437ead19+747527b0/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bd54880c0c95a6079791d4b04e870ebbc4e405f5a961a1ce53420ca32689857e
3
+ size 3371014
neuronxcc-2.21.18209.0+043b1bf7/MODULE_00d39c882e9e437ead19+747527b0/model.log ADDED
@@ -0,0 +1,2 @@
 
 
 
1
+ Failed compilation with ['neuronx-cc', 'compile', '--framework=XLA', '/tmp/nxd_model/token_generation_model/_tp0_bk0/model.MODULE_00d39c882e9e437ead19+747527b0.hlo_module.pb', '--output', '/tmp/nxd_model/token_generation_model/_tp0_bk0/model.MODULE_00d39c882e9e437ead19+747527b0.neff', '--target=trn2', '--auto-cast=none', '--model-type=transformer', '--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ', '-O2', '--lnc=2', '--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt', '--enable-internal-neff-wrapper', '--verbose=35']: 2025-10-17T07:11:57Z 2025-10-17 07:11:57.671531: F hilo/hlo_passes/NeuronHloVerifier.cc:504] [ERROR] [NCC_VRF009] Memory requirement exceeds target architecture's HBM limit. Needed 17455884400 bytes (16 GB) vs. available 17179869184 bytes (16 GB). TIP: Consider using smaller batches or applying model parallelism
2
+
neuronxcc-2.21.18209.0+043b1bf7/MODULE_0a59544d8a22ab5b06a0+747527b0/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn2", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=2", "--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.21.18209.0+043b1bf7/MODULE_0a59544d8a22ab5b06a0+747527b0/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:61a75bca88af13cd0036a4ec2b34c58160f1d04df3a50a44b879a022d3c57214
3
+ size 3150579
neuronxcc-2.21.18209.0+043b1bf7/MODULE_0a59544d8a22ab5b06a0+747527b0/model.log ADDED
@@ -0,0 +1,2 @@
 
 
 
1
+ Failed compilation with ['neuronx-cc', 'compile', '--framework=XLA', '/tmp/nxd_model/token_generation_model/_tp0_bk0/model.MODULE_0a59544d8a22ab5b06a0+747527b0.hlo_module.pb', '--output', '/tmp/nxd_model/token_generation_model/_tp0_bk0/model.MODULE_0a59544d8a22ab5b06a0+747527b0.neff', '--target=trn2', '--auto-cast=none', '--model-type=transformer', '--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ', '-O2', '--lnc=2', '--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt', '--enable-internal-neff-wrapper', '--verbose=35']: 2025-10-17T07:13:35Z 2025-10-17 07:13:34.954102: F hilo/hlo_passes/NeuronHloVerifier.cc:504] [ERROR] [NCC_VRF009] Memory requirement exceeds target architecture's HBM limit. Needed 17187432532 bytes (16 GB) vs. available 17179869184 bytes (16 GB). TIP: Consider using smaller batches or applying model parallelism
2
+
neuronxcc-2.21.18209.0+043b1bf7/MODULE_341e3918a26a3758bc34+877608f3/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn2", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=2", "--logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.21.18209.0+043b1bf7/MODULE_341e3918a26a3758bc34+877608f3/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_341e3918a26a3758bc34+877608f3/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aefa4398901c0d0eff079d32b857ad973d41e0f2b6b620524b6e91f31feb82d9
3
+ size 3227091
neuronxcc-2.21.18209.0+043b1bf7/MODULE_341e3918a26a3758bc34+877608f3/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:07b6209f7697d592c2723ad43c122cb3eb4bea504b52e670ae64af8ba56a71f8
3
+ size 6902784
neuronxcc-2.21.18209.0+043b1bf7/MODULE_3c45b7d14d093b6eb821+747527b0/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn2", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=2", "--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.21.18209.0+043b1bf7/MODULE_3c45b7d14d093b6eb821+747527b0/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:38b0dfaada830ef2c973b0877242920edb447a04792f2a84b338c6467d9629d5
3
+ size 4765744
neuronxcc-2.21.18209.0+043b1bf7/MODULE_3c45b7d14d093b6eb821+747527b0/model.log ADDED
@@ -0,0 +1,2 @@
 
 
 
1
+ Failed compilation with ['neuronx-cc', 'compile', '--framework=XLA', '/tmp/nxd_model/token_generation_model/_tp0_bk0/model.MODULE_3c45b7d14d093b6eb821+747527b0.hlo_module.pb', '--output', '/tmp/nxd_model/token_generation_model/_tp0_bk0/model.MODULE_3c45b7d14d093b6eb821+747527b0.neff', '--target=trn2', '--auto-cast=none', '--model-type=transformer', '--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ', '-O2', '--lnc=2', '--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt', '--enable-internal-neff-wrapper', '--verbose=35']: 2025-10-17T07:08:48Z 2025-10-17 07:08:48.115351: F hilo/hlo_passes/NeuronHloVerifier.cc:504] [ERROR] [NCC_VRF009] Memory requirement exceeds target architecture's HBM limit. Needed 20677306816 bytes (19 GB) vs. available 17179869184 bytes (16 GB). TIP: Consider using smaller batches or applying model parallelism
2
+
neuronxcc-2.21.18209.0+043b1bf7/MODULE_5ed38095f803237b1507+877608f3/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn2", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=2", "--logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.21.18209.0+043b1bf7/MODULE_5ed38095f803237b1507+877608f3/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_5ed38095f803237b1507+877608f3/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:92d3f1fa14f64b1855c351113cc64df41f6fcc515abe962ac49e0539ea102ddc
3
+ size 3373446
neuronxcc-2.21.18209.0+043b1bf7/MODULE_5ed38095f803237b1507+877608f3/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:71a11688935d2c3a924c23a66252ea994d3d90ea4b91876ab75b97059d8ecf17
3
+ size 9769984
neuronxcc-2.21.18209.0+043b1bf7/MODULE_6b32493175a70d9e16b9+747527b0/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn2", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=2", "--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.21.18209.0+043b1bf7/MODULE_6b32493175a70d9e16b9+747527b0/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_6b32493175a70d9e16b9+747527b0/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:88af07e7fe6a3fe427efc73b8006cf5be3ee6586298331009febf6a3016e4ee6
3
+ size 2766437
neuronxcc-2.21.18209.0+043b1bf7/MODULE_6b32493175a70d9e16b9+747527b0/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c997c5f566b7e67afdb7d80bdc2f185a754ffba165d00ba3e296a912f409b67c
3
+ size 18217984
neuronxcc-2.21.18209.0+043b1bf7/MODULE_6b32493175a70d9e16b9+747527b0/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8b508bb1553e8813c82c36c06ea263e2e42adfd0a73b5c11d5687c58a4b559ac
3
+ size 18533668
neuronxcc-2.21.18209.0+043b1bf7/MODULE_eb50383f9a879a8375b9+747527b0/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn2", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=2", "--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.21.18209.0+043b1bf7/MODULE_eb50383f9a879a8375b9+747527b0/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_eb50383f9a879a8375b9+747527b0/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:89c807fe15fa3eeeb6c8996ebc7478fd40e8af6141abcf516e20b76a48a4c688
3
+ size 2953587
neuronxcc-2.21.18209.0+043b1bf7/MODULE_eb50383f9a879a8375b9+747527b0/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8eeea7504b2fef2b6482c3405ecc593cd8d6c023ab18473013ce000589e257ca
3
+ size 21187584
neuronxcc-2.21.18209.0+043b1bf7/MODULE_eb50383f9a879a8375b9+747527b0/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:925dd5583139e414e6b546670b6a757d093f6021f70f0a79495d282d97f1866a
3
+ size 21502820
neuronxcc-2.21.18209.0+043b1bf7/MODULE_ffe3bb712e8a45b20f93+747527b0/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn2", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=2", "--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.21.18209.0+043b1bf7/MODULE_ffe3bb712e8a45b20f93+747527b0/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bf096341d2e88885aeb5fa1949040e4202de6004c555a3a3c8e37b69bbcae7e6
3
+ size 3677670
neuronxcc-2.21.18209.0+043b1bf7/MODULE_ffe3bb712e8a45b20f93+747527b0/model.log ADDED
@@ -0,0 +1,2 @@
 
 
 
1
+ Failed compilation with ['neuronx-cc', 'compile', '--framework=XLA', '/tmp/nxd_model/token_generation_model/_tp0_bk0/model.MODULE_ffe3bb712e8a45b20f93+747527b0.hlo_module.pb', '--output', '/tmp/nxd_model/token_generation_model/_tp0_bk0/model.MODULE_ffe3bb712e8a45b20f93+747527b0.neff', '--target=trn2', '--auto-cast=none', '--model-type=transformer', '--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ', '-O2', '--lnc=2', '--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt', '--enable-internal-neff-wrapper', '--verbose=35']: 2025-10-17T07:10:25Z 2025-10-17 07:10:25.765334: F hilo/hlo_passes/NeuronHloVerifier.cc:504] [ERROR] [NCC_VRF009] Memory requirement exceeds target architecture's HBM limit. Needed 18529691872 bytes (17 GB) vs. available 17179869184 bytes (16 GB). TIP: Consider using smaller batches or applying model parallelism
2
+