Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
Community
648
dacorvo HF Staff commited on
Commit
42958bf
·
verified ·
1 Parent(s): 46a0189

Synchronizing local compiler cache.

Browse files
Files changed (11) hide show
  1. .gitattributes +3 -0
  2. neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.1.dev0/qwen3/Qwen/Qwen3-14B/263e7366132fef3f64d3.json +98 -0
  3. neuronxcc-2.21.18209.0+043b1bf7/MODULE_d0d617e5ebcb6d427b47+24129607/compile_flags.json +1 -0
  4. neuronxcc-2.21.18209.0+043b1bf7/MODULE_d0d617e5ebcb6d427b47+24129607/model.done +0 -0
  5. neuronxcc-2.21.18209.0+043b1bf7/MODULE_d0d617e5ebcb6d427b47+24129607/model.hlo_module.pb +3 -0
  6. neuronxcc-2.21.18209.0+043b1bf7/MODULE_d0d617e5ebcb6d427b47+24129607/model.neff +3 -0
  7. neuronxcc-2.21.18209.0+043b1bf7/MODULE_f7213036558ed85aff51+a02c3a36/compile_flags.json +1 -0
  8. neuronxcc-2.21.18209.0+043b1bf7/MODULE_f7213036558ed85aff51+a02c3a36/model.done +0 -0
  9. neuronxcc-2.21.18209.0+043b1bf7/MODULE_f7213036558ed85aff51+a02c3a36/model.hlo_module.pb +3 -0
  10. neuronxcc-2.21.18209.0+043b1bf7/MODULE_f7213036558ed85aff51+a02c3a36/model.neff +3 -0
  11. neuronxcc-2.21.18209.0+043b1bf7/MODULE_f7213036558ed85aff51+a02c3a36/wrapped_neff.hlo +3 -0
.gitattributes CHANGED
@@ -12403,3 +12403,6 @@ neuronxcc-2.21.18209.0+043b1bf7/MODULE_648617d2920c9c514e3d+24129607/model.neff
12403
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_56d13d2351925c63c07f+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
12404
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_d2bd5330d1a4e4dd0288+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
12405
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_d2bd5330d1a4e4dd0288+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
12403
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_56d13d2351925c63c07f+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
12404
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_d2bd5330d1a4e4dd0288+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
12405
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_d2bd5330d1a4e4dd0288+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
12406
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_d0d617e5ebcb6d427b47+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
12407
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_f7213036558ed85aff51+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
12408
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_f7213036558ed85aff51+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.1.dev0/qwen3/Qwen/Qwen3-14B/263e7366132fef3f64d3.json ADDED
@@ -0,0 +1,98 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "ep_degree": 1,
68
+ "fused_qkv": true,
69
+ "glu_mlp": true,
70
+ "local_ranks_size": 8,
71
+ "max_batch_size": 16,
72
+ "max_context_length": 4096,
73
+ "max_topk": 256,
74
+ "n_active_tokens": 4096,
75
+ "neuronxcc_version": "2.21.18209.0+043b1bf7",
76
+ "on_device_sampling": true,
77
+ "optimum_neuron_version": "0.4.1.dev0",
78
+ "output_logits": false,
79
+ "pp_degree": 1,
80
+ "sequence_length": 4096,
81
+ "speculation_length": 0,
82
+ "start_rank_id": 0,
83
+ "target": "trn1",
84
+ "torch_dtype": "bfloat16",
85
+ "tp_degree": 8
86
+ },
87
+ "num_attention_heads": 40,
88
+ "num_hidden_layers": 40,
89
+ "num_key_value_heads": 8,
90
+ "rms_norm_eps": 1e-06,
91
+ "rope_scaling": null,
92
+ "rope_theta": 1000000,
93
+ "sliding_window": null,
94
+ "tie_word_embeddings": false,
95
+ "use_cache": true,
96
+ "use_sliding_window": false,
97
+ "vocab_size": 151936
98
+ }
neuronxcc-2.21.18209.0+043b1bf7/MODULE_d0d617e5ebcb6d427b47+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-2.21.18209.0+043b1bf7/MODULE_d0d617e5ebcb6d427b47+24129607/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_d0d617e5ebcb6d427b47+24129607/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6e436775752d4959f36aa2b34c18d8d9887749b1446630b4b9d08b2f06f005fb
3
+ size 1266407
neuronxcc-2.21.18209.0+043b1bf7/MODULE_d0d617e5ebcb6d427b47+24129607/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:88027c3c939139cf678b6809ab57f7b33eba02123d02678e65f4b034320e373f
3
+ size 3421184
neuronxcc-2.21.18209.0+043b1bf7/MODULE_f7213036558ed85aff51+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-2.21.18209.0+043b1bf7/MODULE_f7213036558ed85aff51+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_f7213036558ed85aff51+a02c3a36/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_f7213036558ed85aff51+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:00ffab1c7c2fee8e8907f5acad497c5855b14daa9c11a18b2da7e22b25374d01
3
+ size 8346624
neuronxcc-2.21.18209.0+043b1bf7/MODULE_f7213036558ed85aff51+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4e655c2369be95902aaf63e6e0859e485ec67e9fb1bd782368930649598c5aa4
3
+ size 8541279