Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
Community
648
dacorvo HF Staff commited on
Commit
dc3b9b5
·
verified ·
1 Parent(s): 6a37f47

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/Qwen3-32B/457278e020e93446d04e.json +122 -0
  3. neuronxcc-2.21.18209.0+043b1bf7/MODULE_bc66b7b1609635e0e41d+a02c3a36/compile_flags.json +1 -0
  4. neuronxcc-2.21.18209.0+043b1bf7/MODULE_bc66b7b1609635e0e41d+a02c3a36/model.done +0 -0
  5. neuronxcc-2.21.18209.0+043b1bf7/MODULE_bc66b7b1609635e0e41d+a02c3a36/model.hlo_module.pb +3 -0
  6. neuronxcc-2.21.18209.0+043b1bf7/MODULE_bc66b7b1609635e0e41d+a02c3a36/model.neff +3 -0
  7. neuronxcc-2.21.18209.0+043b1bf7/MODULE_bc66b7b1609635e0e41d+a02c3a36/wrapped_neff.hlo +3 -0
  8. neuronxcc-2.21.18209.0+043b1bf7/MODULE_e4bb408a036fb990e90a+24129607/compile_flags.json +1 -0
  9. neuronxcc-2.21.18209.0+043b1bf7/MODULE_e4bb408a036fb990e90a+24129607/model.done +0 -0
  10. neuronxcc-2.21.18209.0+043b1bf7/MODULE_e4bb408a036fb990e90a+24129607/model.hlo_module.pb +3 -0
  11. neuronxcc-2.21.18209.0+043b1bf7/MODULE_e4bb408a036fb990e90a+24129607/model.neff +3 -0
.gitattributes CHANGED
@@ -12427,3 +12427,6 @@ neuronxcc-2.21.18209.0+043b1bf7/MODULE_c0d8c793345bb142d034+24129607/model.neff
12427
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_54197861b0e068020a8d+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
12428
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_7c1393a02d9cc9ac0a99+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
12429
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_7c1393a02d9cc9ac0a99+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
12427
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_54197861b0e068020a8d+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
12428
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_7c1393a02d9cc9ac0a99+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
12429
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_7c1393a02d9cc9ac0a99+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
12430
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_bc66b7b1609635e0e41d+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
12431
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_bc66b7b1609635e0e41d+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
12432
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_e4bb408a036fb990e90a+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.1.dev0/qwen3/Qwen/Qwen3-32B/457278e020e93446d04e.json ADDED
@@ -0,0 +1,122 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "head_dim": 128,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 5120,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 25600,
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
+ "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
+ ],
81
+ "max_position_embeddings": 40960,
82
+ "max_window_layers": 64,
83
+ "model_type": "qwen3",
84
+ "neuron": {
85
+ "_serialized_key": "NxDNeuronConfig",
86
+ "batch_size": 16,
87
+ "capacity_factor": null,
88
+ "checkpoint_id": "Qwen/Qwen3-32B",
89
+ "checkpoint_revision": "9216db5781bf21249d130ec9da846c4624c16137",
90
+ "continuous_batching": true,
91
+ "ep_degree": 1,
92
+ "fused_qkv": true,
93
+ "glu_mlp": true,
94
+ "local_ranks_size": 8,
95
+ "max_batch_size": 16,
96
+ "max_context_length": 4096,
97
+ "max_topk": 256,
98
+ "n_active_tokens": 4096,
99
+ "neuronxcc_version": "2.21.18209.0+043b1bf7",
100
+ "on_device_sampling": true,
101
+ "optimum_neuron_version": "0.4.1.dev0",
102
+ "output_logits": false,
103
+ "pp_degree": 1,
104
+ "sequence_length": 4096,
105
+ "speculation_length": 0,
106
+ "start_rank_id": 0,
107
+ "target": "trn1",
108
+ "torch_dtype": "bfloat16",
109
+ "tp_degree": 8
110
+ },
111
+ "num_attention_heads": 64,
112
+ "num_hidden_layers": 64,
113
+ "num_key_value_heads": 8,
114
+ "rms_norm_eps": 1e-06,
115
+ "rope_scaling": null,
116
+ "rope_theta": 1000000,
117
+ "sliding_window": null,
118
+ "tie_word_embeddings": false,
119
+ "use_cache": true,
120
+ "use_sliding_window": false,
121
+ "vocab_size": 151936
122
+ }
neuronxcc-2.21.18209.0+043b1bf7/MODULE_bc66b7b1609635e0e41d+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_bc66b7b1609635e0e41d+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_bc66b7b1609635e0e41d+a02c3a36/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fb6017f03fbc9fde8570350117f1766c1a1c2b3f04bc1def092a323bd218c300
3
+ size 1643429
neuronxcc-2.21.18209.0+043b1bf7/MODULE_bc66b7b1609635e0e41d+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:98b76f192ffb1daec6da7d83063e0264b47120cd937ec2a0778a094ec5cd1680
3
+ size 15156224
neuronxcc-2.21.18209.0+043b1bf7/MODULE_bc66b7b1609635e0e41d+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a84828505e89670f9f5ec673a57dac92443668a898d560cdbb885061867c28f9
3
+ size 15467421
neuronxcc-2.21.18209.0+043b1bf7/MODULE_e4bb408a036fb990e90a+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_e4bb408a036fb990e90a+24129607/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_e4bb408a036fb990e90a+24129607/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:108f3f1d5ef064a004667ae2d3f69652cca061fa4fb1b07e14d537ac0ee22004
3
+ size 2009056
neuronxcc-2.21.18209.0+043b1bf7/MODULE_e4bb408a036fb990e90a+24129607/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d68db65bef8a0b3f62bdfccf178e4872f9b6c3a855c1c88f0ff9749ee3ffd66f
3
+ size 4199424