Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
Community
611
dacorvo HF Staff commited on
Commit
c7914f2
·
verified ·
1 Parent(s): 4d05aee

Synchronizing local compiler cache.

Browse files
Files changed (11) hide show
  1. .gitattributes +3 -0
  2. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.3.dev2/qwen3/Qwen/Qwen3-32B/8eeb56a9f862ec73fe06.json +123 -0
  3. neuronxcc-2.21.33363.0+82129205/MODULE_0ca0b7d2c1246aa065a0+24129607/compile_flags.json +1 -0
  4. neuronxcc-2.21.33363.0+82129205/MODULE_0ca0b7d2c1246aa065a0+24129607/model.done +0 -0
  5. neuronxcc-2.21.33363.0+82129205/MODULE_0ca0b7d2c1246aa065a0+24129607/model.hlo_module.pb +3 -0
  6. neuronxcc-2.21.33363.0+82129205/MODULE_0ca0b7d2c1246aa065a0+24129607/model.neff +3 -0
  7. neuronxcc-2.21.33363.0+82129205/MODULE_d9eacd4b781267f13a69+a02c3a36/compile_flags.json +1 -0
  8. neuronxcc-2.21.33363.0+82129205/MODULE_d9eacd4b781267f13a69+a02c3a36/model.done +0 -0
  9. neuronxcc-2.21.33363.0+82129205/MODULE_d9eacd4b781267f13a69+a02c3a36/model.hlo_module.pb +3 -0
  10. neuronxcc-2.21.33363.0+82129205/MODULE_d9eacd4b781267f13a69+a02c3a36/model.neff +3 -0
  11. neuronxcc-2.21.33363.0+82129205/MODULE_d9eacd4b781267f13a69+a02c3a36/wrapped_neff.hlo +3 -0
.gitattributes CHANGED
@@ -13828,3 +13828,6 @@ neuronxcc-2.21.33363.0+82129205/MODULE_df1137f33f25bc4d1628+a02c3a36/wrapped_nef
13828
  neuronxcc-2.21.33363.0+82129205/MODULE_11b9f8bcfae064562f98+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
13829
  neuronxcc-2.21.33363.0+82129205/MODULE_1280d8f2f3240abd5752+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
13830
  neuronxcc-2.21.33363.0+82129205/MODULE_1280d8f2f3240abd5752+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
13828
  neuronxcc-2.21.33363.0+82129205/MODULE_11b9f8bcfae064562f98+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
13829
  neuronxcc-2.21.33363.0+82129205/MODULE_1280d8f2f3240abd5752+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
13830
  neuronxcc-2.21.33363.0+82129205/MODULE_1280d8f2f3240abd5752+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
13831
+ neuronxcc-2.21.33363.0+82129205/MODULE_0ca0b7d2c1246aa065a0+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
13832
+ neuronxcc-2.21.33363.0+82129205/MODULE_d9eacd4b781267f13a69+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
13833
+ neuronxcc-2.21.33363.0+82129205/MODULE_d9eacd4b781267f13a69+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.3.dev2/qwen3/Qwen/Qwen3-32B/8eeb56a9f862ec73fe06.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
+ "dtype": "bfloat16",
11
+ "head_dim": 128,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 5120,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 25600,
16
+ "layer_types": [
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
+ ],
82
+ "max_position_embeddings": 40960,
83
+ "max_window_layers": 64,
84
+ "model_type": "qwen3",
85
+ "neuron": {
86
+ "_serialized_key": "NxDNeuronConfig",
87
+ "batch_size": 16,
88
+ "capacity_factor": null,
89
+ "checkpoint_id": "Qwen/Qwen3-32B",
90
+ "checkpoint_revision": "9216db5781bf21249d130ec9da846c4624c16137",
91
+ "continuous_batching": true,
92
+ "ep_degree": 1,
93
+ "fused_qkv": true,
94
+ "glu_mlp": true,
95
+ "local_ranks_size": 8,
96
+ "max_batch_size": 16,
97
+ "max_context_length": 4096,
98
+ "max_topk": 256,
99
+ "n_active_tokens": 4096,
100
+ "neuronxcc_version": "2.21.33363.0+82129205",
101
+ "on_device_sampling": true,
102
+ "optimum_neuron_version": "0.4.3.dev2",
103
+ "output_logits": false,
104
+ "pp_degree": 1,
105
+ "sequence_length": 4096,
106
+ "speculation_length": 0,
107
+ "start_rank_id": 0,
108
+ "target": "trn1",
109
+ "torch_dtype": "bfloat16",
110
+ "tp_degree": 8
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-2.21.33363.0+82129205/MODULE_0ca0b7d2c1246aa065a0+24129607/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/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.21.33363.0+82129205/MODULE_0ca0b7d2c1246aa065a0+24129607/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_0ca0b7d2c1246aa065a0+24129607/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6de4f79601f0ac1aab7f5c70ade3e55ef29dbd2fbc94d4f45f55714d52d0dc71
3
+ size 2009056
neuronxcc-2.21.33363.0+82129205/MODULE_0ca0b7d2c1246aa065a0+24129607/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dcb11e65d66209dd80104d53a68f91bc154206bb67ff2a3ad2447235238bdda1
3
+ size 4199424
neuronxcc-2.21.33363.0+82129205/MODULE_d9eacd4b781267f13a69+a02c3a36/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/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.21.33363.0+82129205/MODULE_d9eacd4b781267f13a69+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_d9eacd4b781267f13a69+a02c3a36/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:023572e16837e0b6bb9edfc89cf7e567a8d0c97605c0d2e9e63848d6e4fa85b7
3
+ size 1643429
neuronxcc-2.21.33363.0+82129205/MODULE_d9eacd4b781267f13a69+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:30bc6da6e206377b208420671439a7c53b64573013ec5525f06579ad22d22a95
3
+ size 15156224
neuronxcc-2.21.33363.0+82129205/MODULE_d9eacd4b781267f13a69+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:15ad5b5743b2bcadd9edff1d5ecbabe15c2329e65b088ae561afad4a778418dd
3
+ size 15467421