Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
 Community
666
dacorvo HF Staff commited on
Commit
fa9694f
·
verified ·
1 Parent(s): ad17351

Synchronizing local compiler cache.

Browse files
Files changed (14) hide show
  1. .gitattributes +3 -0
  2. neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.0.dev0/qwen2/Qwen/Qwen2.5-72B-Instruct/d7756f7a147b923cb0f7.json +137 -0
  3. neuronxcc-2.21.18209.0+043b1bf7/MODULE_2a3c2e6955495ca5363a+ed72d204/compile_flags.json +1 -0
  4. neuronxcc-2.21.18209.0+043b1bf7/MODULE_2a3c2e6955495ca5363a+ed72d204/model.done +0 -0
  5. neuronxcc-2.21.18209.0+043b1bf7/MODULE_2a3c2e6955495ca5363a+ed72d204/model.hlo_module.pb +3 -0
  6. neuronxcc-2.21.18209.0+043b1bf7/MODULE_2a3c2e6955495ca5363a+ed72d204/model.neff +3 -0
  7. neuronxcc-2.21.18209.0+043b1bf7/MODULE_2d4c725c162596d83fd9+a9d440f5/compile_flags.json +1 -0
  8. neuronxcc-2.21.18209.0+043b1bf7/MODULE_2d4c725c162596d83fd9+a9d440f5/model.done +0 -0
  9. neuronxcc-2.21.18209.0+043b1bf7/MODULE_2d4c725c162596d83fd9+a9d440f5/model.hlo_module.pb +3 -0
  10. neuronxcc-2.21.18209.0+043b1bf7/MODULE_2d4c725c162596d83fd9+a9d440f5/model.neff +3 -0
  11. neuronxcc-2.21.18209.0+043b1bf7/MODULE_2d4c725c162596d83fd9+a9d440f5/wrapped_neff.hlo +3 -0
  12. neuronxcc-2.21.18209.0+043b1bf7/MODULE_33fe5ce4a5a998514a39+a9d440f5/compile_flags.json +1 -0
  13. neuronxcc-2.21.18209.0+043b1bf7/MODULE_33fe5ce4a5a998514a39+a9d440f5/model.hlo_module.pb +3 -0
  14. neuronxcc-2.21.18209.0+043b1bf7/MODULE_33fe5ce4a5a998514a39+a9d440f5/model.log +2 -0
.gitattributes CHANGED
@@ -11686,3 +11686,6 @@ neuronxcc-2.21.18209.0+043b1bf7/MODULE_f2cde388cf9f4159501b+a9d440f5/wrapped_nef
11686
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_040ba3e8ce41452abb14+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11687
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_040ba3e8ce41452abb14+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
11688
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_44bc25621dae578326f5+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
 
 
 
 
11686
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_040ba3e8ce41452abb14+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11687
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_040ba3e8ce41452abb14+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
11688
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_44bc25621dae578326f5+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11689
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_2a3c2e6955495ca5363a+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11690
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_2d4c725c162596d83fd9+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11691
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_2d4c725c162596d83fd9+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.0.dev0/qwen2/Qwen/Qwen2.5-72B-Instruct/d7756f7a147b923cb0f7.json ADDED
@@ -0,0 +1,137 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen2.5-72B-Instruct",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Qwen2ForCausalLM"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 8192,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 29568,
13
+ "layer_types": [
14
+ "full_attention",
15
+ "full_attention",
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
+ "full_attention",
81
+ "full_attention",
82
+ "full_attention",
83
+ "full_attention",
84
+ "full_attention",
85
+ "full_attention",
86
+ "full_attention",
87
+ "full_attention",
88
+ "full_attention",
89
+ "full_attention",
90
+ "full_attention",
91
+ "full_attention",
92
+ "full_attention",
93
+ "full_attention"
94
+ ],
95
+ "max_position_embeddings": 32768,
96
+ "max_window_layers": 70,
97
+ "model_type": "qwen2",
98
+ "neuron": {
99
+ "_serialized_key": "NxDNeuronConfig",
100
+ "batch_size": 4,
101
+ "capacity_factor": null,
102
+ "checkpoint_id": "Qwen/Qwen2.5-72B-Instruct",
103
+ "checkpoint_revision": "495f39366efef23836d0cfae4fbe635880d2be31",
104
+ "continuous_batching": true,
105
+ "enable_bucketing": false,
106
+ "ep_degree": 1,
107
+ "fused_qkv": false,
108
+ "glu_mlp": true,
109
+ "local_ranks_size": 24,
110
+ "max_batch_size": 4,
111
+ "max_context_length": 4096,
112
+ "max_topk": 256,
113
+ "n_active_tokens": 4096,
114
+ "neuronxcc_version": "2.21.18209.0+043b1bf7",
115
+ "on_device_sampling": true,
116
+ "optimum_neuron_version": "0.4.0.dev0",
117
+ "output_logits": false,
118
+ "pp_degree": 1,
119
+ "sequence_length": 4096,
120
+ "speculation_length": 0,
121
+ "start_rank_id": 0,
122
+ "target": "trn1",
123
+ "torch_dtype": "bfloat16",
124
+ "tp_degree": 24
125
+ },
126
+ "num_attention_heads": 64,
127
+ "num_hidden_layers": 80,
128
+ "num_key_value_heads": 8,
129
+ "rms_norm_eps": 1e-06,
130
+ "rope_scaling": null,
131
+ "rope_theta": 1000000.0,
132
+ "sliding_window": null,
133
+ "tie_word_embeddings": false,
134
+ "use_cache": true,
135
+ "use_sliding_window": false,
136
+ "vocab_size": 152064
137
+ }
neuronxcc-2.21.18209.0+043b1bf7/MODULE_2a3c2e6955495ca5363a+ed72d204/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn1", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=1", "--logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.21.18209.0+043b1bf7/MODULE_2a3c2e6955495ca5363a+ed72d204/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_2a3c2e6955495ca5363a+ed72d204/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2f0b08f2f1804caaa3d81a974d54b49ef9dd1f6b0f13debef538a1df486e8fb6
3
+ size 2562127
neuronxcc-2.21.18209.0+043b1bf7/MODULE_2a3c2e6955495ca5363a+ed72d204/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aaa497d8f354063cbcccbaefa15b4f12dba8eeb1f550d49718736f71beef7d44
3
+ size 3165184
neuronxcc-2.21.18209.0+043b1bf7/MODULE_2d4c725c162596d83fd9+a9d440f5/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn1", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=1", "--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.21.18209.0+043b1bf7/MODULE_2d4c725c162596d83fd9+a9d440f5/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_2d4c725c162596d83fd9+a9d440f5/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:afb861a906f80655f6285210046e1145ee1a7fc2c80dbd54eeaf11170ef9a4a4
3
+ size 2066381
neuronxcc-2.21.18209.0+043b1bf7/MODULE_2d4c725c162596d83fd9+a9d440f5/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5ca3270a564f27cfb1432a393c9b9ec8034d563a83f6f21f45d2d765f0602704
3
+ size 11439104
neuronxcc-2.21.18209.0+043b1bf7/MODULE_2d4c725c162596d83fd9+a9d440f5/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e28d0421e6a2f515894d18cb93a2624a9e53cfaa8702cb6236bbc9a27b5f0571
3
+ size 11967498
neuronxcc-2.21.18209.0+043b1bf7/MODULE_33fe5ce4a5a998514a39+a9d440f5/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn1", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=1", "--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.21.18209.0+043b1bf7/MODULE_33fe5ce4a5a998514a39+a9d440f5/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2bd7c29b2997482a3d5447c9d6db21b6fea798133aafea1111aafe85fc80af51
3
+ size 2081711
neuronxcc-2.21.18209.0+043b1bf7/MODULE_33fe5ce4a5a998514a39+a9d440f5/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_33fe5ce4a5a998514a39+a9d440f5.hlo_module.pb', '--output', '/tmp/nxd_model/token_generation_model/_tp0_bk0/model.MODULE_33fe5ce4a5a998514a39+a9d440f5.neff', '--target=trn1', '--auto-cast=none', '--model-type=transformer', '--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ', '-O2', '--lnc=1', '--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt', '--enable-internal-neff-wrapper', '--verbose=35']: 2025-10-08T02:34:44Z 2025-10-08 02:34:44.435146: F hilo/hlo_passes/NeuronHloVerifier.cc:504] [ERROR] [NCC_VRF009] Memory requirement exceeds target architecture's HBM limit. Needed 73378627612 bytes (68 GB) vs. available 17179869184 bytes (16 GB). TIP: Consider using smaller batches or applying model parallelism
2
+