Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
Community
596
dacorvo HF Staff commited on
Commit
cb22b79
·
verified ·
1 Parent(s): c999d66

Synchronizing local compiler cache.

Browse files
Files changed (11) hide show
  1. .gitattributes +3 -0
  2. neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.0.dev0/qwen3/Qwen/Qwen3-8B/ccad97ea7623f57f7e2d.json +95 -0
  3. neuronxcc-2.21.18209.0+043b1bf7/MODULE_012e553087466b43c8a6+a9d440f5/compile_flags.json +1 -0
  4. neuronxcc-2.21.18209.0+043b1bf7/MODULE_012e553087466b43c8a6+a9d440f5/model.done +0 -0
  5. neuronxcc-2.21.18209.0+043b1bf7/MODULE_012e553087466b43c8a6+a9d440f5/model.hlo_module.pb +3 -0
  6. neuronxcc-2.21.18209.0+043b1bf7/MODULE_012e553087466b43c8a6+a9d440f5/model.neff +3 -0
  7. neuronxcc-2.21.18209.0+043b1bf7/MODULE_012e553087466b43c8a6+a9d440f5/wrapped_neff.hlo +3 -0
  8. neuronxcc-2.21.18209.0+043b1bf7/MODULE_597a28f3c38358793ef7+ed72d204/compile_flags.json +1 -0
  9. neuronxcc-2.21.18209.0+043b1bf7/MODULE_597a28f3c38358793ef7+ed72d204/model.done +0 -0
  10. neuronxcc-2.21.18209.0+043b1bf7/MODULE_597a28f3c38358793ef7+ed72d204/model.hlo_module.pb +3 -0
  11. neuronxcc-2.21.18209.0+043b1bf7/MODULE_597a28f3c38358793ef7+ed72d204/model.neff +3 -0
.gitattributes CHANGED
@@ -11506,3 +11506,6 @@ neuronxcc-2.21.18209.0+043b1bf7/MODULE_686e3b40310d36e8a2bd+a9d440f5/wrapped_nef
11506
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_958c72c2b832e393cb20+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11507
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_efc5eabcaf310ae46e01+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11508
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_efc5eabcaf310ae46e01+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
11506
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_958c72c2b832e393cb20+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11507
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_efc5eabcaf310ae46e01+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11508
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_efc5eabcaf310ae46e01+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
11509
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_012e553087466b43c8a6+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11510
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_012e553087466b43c8a6+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
11511
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_597a28f3c38358793ef7+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.0.dev0/qwen3/Qwen/Qwen3-8B/ccad97ea7623f57f7e2d.json ADDED
@@ -0,0 +1,95 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen3-8B",
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": 4096,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 12288,
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
+ ],
53
+ "max_position_embeddings": 40960,
54
+ "max_window_layers": 36,
55
+ "model_type": "qwen3",
56
+ "neuron": {
57
+ "_serialized_key": "NxDNeuronConfig",
58
+ "batch_size": 8,
59
+ "capacity_factor": null,
60
+ "checkpoint_id": "Qwen/Qwen3-8B",
61
+ "checkpoint_revision": "b968826d9c46dd6066d109eabc6255188de91218",
62
+ "continuous_batching": true,
63
+ "enable_bucketing": false,
64
+ "ep_degree": 1,
65
+ "fused_qkv": true,
66
+ "glu_mlp": true,
67
+ "local_ranks_size": 2,
68
+ "max_batch_size": 8,
69
+ "max_context_length": 4096,
70
+ "max_topk": 256,
71
+ "n_active_tokens": 4096,
72
+ "neuronxcc_version": "2.21.18209.0+043b1bf7",
73
+ "on_device_sampling": false,
74
+ "optimum_neuron_version": "0.4.0.dev0",
75
+ "output_logits": false,
76
+ "pp_degree": 1,
77
+ "sequence_length": 4096,
78
+ "speculation_length": 0,
79
+ "start_rank_id": 0,
80
+ "target": "trn1",
81
+ "torch_dtype": "bfloat16",
82
+ "tp_degree": 2
83
+ },
84
+ "num_attention_heads": 32,
85
+ "num_hidden_layers": 36,
86
+ "num_key_value_heads": 8,
87
+ "rms_norm_eps": 1e-06,
88
+ "rope_scaling": null,
89
+ "rope_theta": 1000000,
90
+ "sliding_window": null,
91
+ "tie_word_embeddings": false,
92
+ "use_cache": true,
93
+ "use_sliding_window": false,
94
+ "vocab_size": 151936
95
+ }
neuronxcc-2.21.18209.0+043b1bf7/MODULE_012e553087466b43c8a6+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_012e553087466b43c8a6+a9d440f5/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_012e553087466b43c8a6+a9d440f5/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf70645d7046d072e5ae37760ecbb9ada5c566026789e025fa4248fded35ed98
3
+ size 905395
neuronxcc-2.21.18209.0+043b1bf7/MODULE_012e553087466b43c8a6+a9d440f5/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:22366dd5d35047795a90efff1218bf6ef12a77e236e22862c7938101651cfd00
3
+ size 12821504
neuronxcc-2.21.18209.0+043b1bf7/MODULE_012e553087466b43c8a6+a9d440f5/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:089a41cedaf7748e478428b4c2f0f8d937d8cfb653df0b532430af20f82fbe19
3
+ size 12996604
neuronxcc-2.21.18209.0+043b1bf7/MODULE_597a28f3c38358793ef7+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_597a28f3c38358793ef7+ed72d204/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_597a28f3c38358793ef7+ed72d204/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2d58056f45e6e4b999189b0239e908fd4e44dc47fa0a34e320252f852b6b9d40
3
+ size 1110837
neuronxcc-2.21.18209.0+043b1bf7/MODULE_597a28f3c38358793ef7+ed72d204/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c2b3289b9cd2709c08a23446316c0832c533181a55ed8912337c6d7d11f317ae
3
+ size 6810624