Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
Community
651
dacorvo HF Staff commited on
Commit
1642de3
·
verified ·
1 Parent(s): ca06c16

Synchronizing local compiler cache.

Browse files
Files changed (11) hide show
  1. .gitattributes +3 -0
  2. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev2/qwen3/Qwen/Qwen3-14B/fb68f71c3969ee742867.json +99 -0
  3. neuronxcc-2.21.33363.0+82129205/MODULE_7300f5516bc6f2872a3a+24129607/compile_flags.json +1 -0
  4. neuronxcc-2.21.33363.0+82129205/MODULE_7300f5516bc6f2872a3a+24129607/model.done +0 -0
  5. neuronxcc-2.21.33363.0+82129205/MODULE_7300f5516bc6f2872a3a+24129607/model.hlo_module.pb +3 -0
  6. neuronxcc-2.21.33363.0+82129205/MODULE_7300f5516bc6f2872a3a+24129607/model.neff +3 -0
  7. neuronxcc-2.21.33363.0+82129205/MODULE_a692be11252819c07b67+a02c3a36/compile_flags.json +1 -0
  8. neuronxcc-2.21.33363.0+82129205/MODULE_a692be11252819c07b67+a02c3a36/model.done +0 -0
  9. neuronxcc-2.21.33363.0+82129205/MODULE_a692be11252819c07b67+a02c3a36/model.hlo_module.pb +3 -0
  10. neuronxcc-2.21.33363.0+82129205/MODULE_a692be11252819c07b67+a02c3a36/model.neff +3 -0
  11. neuronxcc-2.21.33363.0+82129205/MODULE_a692be11252819c07b67+a02c3a36/wrapped_neff.hlo +3 -0
.gitattributes CHANGED
@@ -15363,3 +15363,6 @@ neuronxcc-2.21.33363.0+82129205/MODULE_9bf3d53005f3f0becb28+a02c3a36/wrapped_nef
15363
  neuronxcc-2.21.33363.0+82129205/MODULE_acd5a804304ba1625d45+a32116a7/model.neff filter=lfs diff=lfs merge=lfs -text
15364
  neuronxcc-2.21.33363.0+82129205/MODULE_f2e00d42303afd7207f3+ac10809c/model.neff filter=lfs diff=lfs merge=lfs -text
15365
  neuronxcc-2.21.33363.0+82129205/MODULE_f2e00d42303afd7207f3+ac10809c/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
15363
  neuronxcc-2.21.33363.0+82129205/MODULE_acd5a804304ba1625d45+a32116a7/model.neff filter=lfs diff=lfs merge=lfs -text
15364
  neuronxcc-2.21.33363.0+82129205/MODULE_f2e00d42303afd7207f3+ac10809c/model.neff filter=lfs diff=lfs merge=lfs -text
15365
  neuronxcc-2.21.33363.0+82129205/MODULE_f2e00d42303afd7207f3+ac10809c/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
15366
+ neuronxcc-2.21.33363.0+82129205/MODULE_7300f5516bc6f2872a3a+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
15367
+ neuronxcc-2.21.33363.0+82129205/MODULE_a692be11252819c07b67+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
15368
+ neuronxcc-2.21.33363.0+82129205/MODULE_a692be11252819c07b67+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev2/qwen3/Qwen/Qwen3-14B/fb68f71c3969ee742867.json ADDED
@@ -0,0 +1,99 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen3-14B",
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": 17408,
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
+ ],
58
+ "max_position_embeddings": 40960,
59
+ "max_window_layers": 40,
60
+ "model_type": "qwen3",
61
+ "neuron": {
62
+ "_serialized_key": "NxDNeuronConfig",
63
+ "batch_size": 16,
64
+ "capacity_factor": null,
65
+ "checkpoint_id": "Qwen/Qwen3-14B",
66
+ "checkpoint_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
67
+ "continuous_batching": true,
68
+ "ep_degree": 1,
69
+ "fused_qkv": true,
70
+ "glu_mlp": true,
71
+ "local_ranks_size": 8,
72
+ "max_batch_size": 16,
73
+ "max_context_length": 4096,
74
+ "max_topk": 256,
75
+ "n_active_tokens": 4096,
76
+ "neuronxcc_version": "2.21.33363.0+82129205",
77
+ "on_device_sampling": true,
78
+ "optimum_neuron_version": "0.4.4.dev2",
79
+ "output_logits": false,
80
+ "pp_degree": 1,
81
+ "sequence_length": 4096,
82
+ "speculation_length": 0,
83
+ "start_rank_id": 0,
84
+ "target": "trn1",
85
+ "torch_dtype": "bfloat16",
86
+ "tp_degree": 8
87
+ },
88
+ "num_attention_heads": 40,
89
+ "num_hidden_layers": 40,
90
+ "num_key_value_heads": 8,
91
+ "rms_norm_eps": 1e-06,
92
+ "rope_scaling": null,
93
+ "rope_theta": 1000000,
94
+ "sliding_window": null,
95
+ "tie_word_embeddings": false,
96
+ "use_cache": true,
97
+ "use_sliding_window": false,
98
+ "vocab_size": 151936
99
+ }
neuronxcc-2.21.33363.0+82129205/MODULE_7300f5516bc6f2872a3a+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_7300f5516bc6f2872a3a+24129607/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_7300f5516bc6f2872a3a+24129607/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d194f2f407522e4a9411fba06431df57208c440360e2ae013b1974b59b0c0dce
3
+ size 1266407
neuronxcc-2.21.33363.0+82129205/MODULE_7300f5516bc6f2872a3a+24129607/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4fcf08cd763140ef215d470b97208e1cb5eed61c3a3d73daf1d9f0b82cd74993
3
+ size 3421184
neuronxcc-2.21.33363.0+82129205/MODULE_a692be11252819c07b67+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_a692be11252819c07b67+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_a692be11252819c07b67+a02c3a36/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c82d714f9dedeef648e3163a4e2e00db5220f36cb753c1d8012364be8fe43278
3
+ size 1041347
neuronxcc-2.21.33363.0+82129205/MODULE_a692be11252819c07b67+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ea4cacef5bfe2b603733095489080ae9fe83194d3e67903c67de6bd00930c69f
3
+ size 8326144
neuronxcc-2.21.33363.0+82129205/MODULE_a692be11252819c07b67+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:31728270157b95ad7f92e47529cbeb27f50b48938e9c5344add5c2675508f474
3
+ size 8520656