Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
 Community
684
dacorvo HF Staff commited on
Commit
f9c3873
·
verified ·
1 Parent(s): dca0c68

Synchronizing local compiler cache.

Browse files
Files changed (11) hide show
  1. .gitattributes +3 -0
  2. neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.3.1.dev7/qwen3/Qwen/Qwen3-14B/1ef81a6ee423ed405202.json +100 -0
  3. neuronxcc-2.21.18209.0+043b1bf7/MODULE_7804b84207cd92077782+ed72d204/compile_flags.json +1 -0
  4. neuronxcc-2.21.18209.0+043b1bf7/MODULE_7804b84207cd92077782+ed72d204/model.done +0 -0
  5. neuronxcc-2.21.18209.0+043b1bf7/MODULE_7804b84207cd92077782+ed72d204/model.hlo_module.pb +3 -0
  6. neuronxcc-2.21.18209.0+043b1bf7/MODULE_7804b84207cd92077782+ed72d204/model.neff +3 -0
  7. neuronxcc-2.21.18209.0+043b1bf7/MODULE_99934e6b0a01a0fff37a+a9d440f5/compile_flags.json +1 -0
  8. neuronxcc-2.21.18209.0+043b1bf7/MODULE_99934e6b0a01a0fff37a+a9d440f5/model.done +0 -0
  9. neuronxcc-2.21.18209.0+043b1bf7/MODULE_99934e6b0a01a0fff37a+a9d440f5/model.hlo_module.pb +3 -0
  10. neuronxcc-2.21.18209.0+043b1bf7/MODULE_99934e6b0a01a0fff37a+a9d440f5/model.neff +3 -0
  11. neuronxcc-2.21.18209.0+043b1bf7/MODULE_99934e6b0a01a0fff37a+a9d440f5/wrapped_neff.hlo +3 -0
.gitattributes CHANGED
@@ -11368,3 +11368,6 @@ neuronxcc-2.21.18209.0+043b1bf7/MODULE_e113d55f1f361e1644f3+ed72d204/model.neff
11368
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_71777dc7066a131328e9+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11369
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_71777dc7066a131328e9+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
11370
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_d7a6d942281cbe0d0222+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
 
 
 
 
11368
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_71777dc7066a131328e9+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11369
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_71777dc7066a131328e9+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
11370
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_d7a6d942281cbe0d0222+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11371
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_7804b84207cd92077782+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11372
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_99934e6b0a01a0fff37a+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11373
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_99934e6b0a01a0fff37a+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.3.1.dev7/qwen3/Qwen/Qwen3-14B/1ef81a6ee423ed405202.json ADDED
@@ -0,0 +1,100 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "head_dim": 128,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 5120,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 17408,
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
+ ],
57
+ "max_position_embeddings": 40960,
58
+ "max_window_layers": 40,
59
+ "model_type": "qwen3",
60
+ "neuron": {
61
+ "_serialized_key": "NxDNeuronConfig",
62
+ "batch_size": 16,
63
+ "capacity_factor": null,
64
+ "checkpoint_id": "Qwen/Qwen3-14B",
65
+ "checkpoint_revision": "40c069824f4251a91eefaf281ebe4c544efd3e18",
66
+ "continuous_batching": true,
67
+ "enable_bucketing": false,
68
+ "ep_degree": 1,
69
+ "fused_qkv": true,
70
+ "glu_mlp": true,
71
+ "local_ranks_size": 8,
72
+ "logical_nc_config": 1,
73
+ "max_batch_size": 16,
74
+ "max_context_length": 4096,
75
+ "max_topk": 256,
76
+ "n_active_tokens": 4096,
77
+ "neuronxcc_version": "2.21.18209.0+043b1bf7",
78
+ "on_device_sampling": true,
79
+ "optimum_neuron_version": "0.3.1.dev7",
80
+ "output_logits": false,
81
+ "pp_degree": 1,
82
+ "sequence_length": 4096,
83
+ "speculation_length": 0,
84
+ "start_rank_id": 0,
85
+ "target": null,
86
+ "torch_dtype": "bfloat16",
87
+ "tp_degree": 8
88
+ },
89
+ "num_attention_heads": 40,
90
+ "num_hidden_layers": 40,
91
+ "num_key_value_heads": 8,
92
+ "rms_norm_eps": 1e-06,
93
+ "rope_scaling": null,
94
+ "rope_theta": 1000000,
95
+ "sliding_window": null,
96
+ "tie_word_embeddings": false,
97
+ "use_cache": true,
98
+ "use_sliding_window": false,
99
+ "vocab_size": 151936
100
+ }
neuronxcc-2.21.18209.0+043b1bf7/MODULE_7804b84207cd92077782+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_7804b84207cd92077782+ed72d204/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_7804b84207cd92077782+ed72d204/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8d3b47fafbce4aca4036677b68338f8f8e049e65d843d10e87b80f6c06a528f5
3
+ size 1266407
neuronxcc-2.21.18209.0+043b1bf7/MODULE_7804b84207cd92077782+ed72d204/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b83b22ffb50cbc34db89fa830d1e0b777fe867816f83a070ad7ca096d544e1a4
3
+ size 3421184
neuronxcc-2.21.18209.0+043b1bf7/MODULE_99934e6b0a01a0fff37a+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_99934e6b0a01a0fff37a+a9d440f5/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_99934e6b0a01a0fff37a+a9d440f5/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:65e9dacb4c94ba650652ee3b413cca9499f2d16a6a9adc93ef11c354762039f7
3
+ size 1032793
neuronxcc-2.21.18209.0+043b1bf7/MODULE_99934e6b0a01a0fff37a+a9d440f5/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1ff9144bfb1a83d9ee6d1c06cf5353ba2cb8231130b57f03d5ef06f73df56883
3
+ size 8346624
neuronxcc-2.21.18209.0+043b1bf7/MODULE_99934e6b0a01a0fff37a+a9d440f5/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2cb2eab473eb4fc770369ef2212f54c101a8d1e31a4f94a375bb7f85d36f8e9c
3
+ size 8541279