Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
Community
614
dacorvo HF Staff commited on
Commit
a394be6
·
verified ·
1 Parent(s): f74fc7c

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/qwen2/Qwen/Qwen2.5-32B-Instruct/2ff1e6b3684b3ae567e1.json +121 -0
  3. neuronxcc-2.21.18209.0+043b1bf7/MODULE_5113f089d70916d7538a+a9d440f5/compile_flags.json +1 -0
  4. neuronxcc-2.21.18209.0+043b1bf7/MODULE_5113f089d70916d7538a+a9d440f5/model.done +0 -0
  5. neuronxcc-2.21.18209.0+043b1bf7/MODULE_5113f089d70916d7538a+a9d440f5/model.hlo_module.pb +3 -0
  6. neuronxcc-2.21.18209.0+043b1bf7/MODULE_5113f089d70916d7538a+a9d440f5/model.neff +3 -0
  7. neuronxcc-2.21.18209.0+043b1bf7/MODULE_5113f089d70916d7538a+a9d440f5/wrapped_neff.hlo +3 -0
  8. neuronxcc-2.21.18209.0+043b1bf7/MODULE_e5e197a9b78873f93025+ed72d204/compile_flags.json +1 -0
  9. neuronxcc-2.21.18209.0+043b1bf7/MODULE_e5e197a9b78873f93025+ed72d204/model.done +0 -0
  10. neuronxcc-2.21.18209.0+043b1bf7/MODULE_e5e197a9b78873f93025+ed72d204/model.hlo_module.pb +3 -0
  11. neuronxcc-2.21.18209.0+043b1bf7/MODULE_e5e197a9b78873f93025+ed72d204/model.neff +3 -0
.gitattributes CHANGED
@@ -11665,3 +11665,6 @@ neuronxcc-2.21.18209.0+043b1bf7/MODULE_e245222df47bf489b9a8+ed72d204/model.neff
11665
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_6e0707468c3d5a25922c+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11666
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_79a608e9f32962123ffd+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11667
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_79a608e9f32962123ffd+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
11665
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_6e0707468c3d5a25922c+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11666
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_79a608e9f32962123ffd+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11667
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_79a608e9f32962123ffd+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
11668
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_5113f089d70916d7538a+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11669
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_5113f089d70916d7538a+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
11670
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_e5e197a9b78873f93025+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.0.dev0/qwen2/Qwen/Qwen2.5-32B-Instruct/2ff1e6b3684b3ae567e1.json ADDED
@@ -0,0 +1,121 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen2.5-32B-Instruct",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Qwen2ForCausalLM"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 5120,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 27648,
13
+ "layer_types": [
14
+ "full_attention",
15
+ "full_attention",
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
+ ],
79
+ "max_position_embeddings": 32768,
80
+ "max_window_layers": 70,
81
+ "model_type": "qwen2",
82
+ "neuron": {
83
+ "_serialized_key": "NxDNeuronConfig",
84
+ "batch_size": 8,
85
+ "capacity_factor": null,
86
+ "checkpoint_id": "Qwen/Qwen2.5-32B-Instruct",
87
+ "checkpoint_revision": "5ede1c97bbab6ce5cda5812749b4c0bdf79b18dd",
88
+ "continuous_batching": true,
89
+ "enable_bucketing": false,
90
+ "ep_degree": 1,
91
+ "fused_qkv": false,
92
+ "glu_mlp": true,
93
+ "local_ranks_size": 8,
94
+ "max_batch_size": 8,
95
+ "max_context_length": 4096,
96
+ "max_topk": 256,
97
+ "n_active_tokens": 4096,
98
+ "neuronxcc_version": "2.21.18209.0+043b1bf7",
99
+ "on_device_sampling": true,
100
+ "optimum_neuron_version": "0.4.0.dev0",
101
+ "output_logits": false,
102
+ "pp_degree": 1,
103
+ "sequence_length": 4096,
104
+ "speculation_length": 0,
105
+ "start_rank_id": 0,
106
+ "target": "trn1",
107
+ "torch_dtype": "bfloat16",
108
+ "tp_degree": 8
109
+ },
110
+ "num_attention_heads": 40,
111
+ "num_hidden_layers": 64,
112
+ "num_key_value_heads": 8,
113
+ "rms_norm_eps": 1e-06,
114
+ "rope_scaling": null,
115
+ "rope_theta": 1000000.0,
116
+ "sliding_window": null,
117
+ "tie_word_embeddings": false,
118
+ "use_cache": true,
119
+ "use_sliding_window": false,
120
+ "vocab_size": 152064
121
+ }
neuronxcc-2.21.18209.0+043b1bf7/MODULE_5113f089d70916d7538a+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_5113f089d70916d7538a+a9d440f5/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_5113f089d70916d7538a+a9d440f5/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b8c7f017046bfa4e743e9841205ca24138237f400971df807fde89f82f64527f
3
+ size 1642677
neuronxcc-2.21.18209.0+043b1bf7/MODULE_5113f089d70916d7538a+a9d440f5/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:51f03817cbeb79fcc59adce3d1dca9dc71994c2d465088799d02c72396a6de4d
3
+ size 13487104
neuronxcc-2.21.18209.0+043b1bf7/MODULE_5113f089d70916d7538a+a9d440f5/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a151583e4ff626b3b2790233c3854ae59da5a31d13a9269ff5e6ac64fa24b8d1
3
+ size 13909623
neuronxcc-2.21.18209.0+043b1bf7/MODULE_e5e197a9b78873f93025+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_e5e197a9b78873f93025+ed72d204/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_e5e197a9b78873f93025+ed72d204/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7341a602b92a814c26bb7242129b289d1e0e7f7d806e99013a07e780b2c87cf1
3
+ size 2039689
neuronxcc-2.21.18209.0+043b1bf7/MODULE_e5e197a9b78873f93025+ed72d204/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9d89766e62812230cef19d4dd8ae31ec1f69477412efe686ccbe6cb813abffc6
3
+ size 3779584