Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
Community
640
dacorvo HF Staff commited on
Commit
e10d2ce
·
verified ·
1 Parent(s): 75d2024

Synchronizing local compiler cache.

Browse files
Files changed (7) hide show
  1. .gitattributes +1 -0
  2. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev1/af58eb15d8e02338dc2f2e880e9c6ec803a98278914b3606acdcc252e7e18429/84e04c656de22f121b56.json +95 -0
  3. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev1/qwen3/Qwen/Qwen3-Embedding-8B/84e04c656de22f121b56.json +95 -0
  4. neuronxcc-2.21.33363.0+82129205/MODULE_a2f92cc081ac96561d06+fb4cc044/compile_flags.json +1 -0
  5. neuronxcc-2.21.33363.0+82129205/MODULE_a2f92cc081ac96561d06+fb4cc044/model.done +0 -0
  6. neuronxcc-2.21.33363.0+82129205/MODULE_a2f92cc081ac96561d06+fb4cc044/model.hlo_module.pb +3 -0
  7. neuronxcc-2.21.33363.0+82129205/MODULE_a2f92cc081ac96561d06+fb4cc044/model.neff +3 -0
.gitattributes CHANGED
@@ -15655,3 +15655,4 @@ neuronxcc-2.21.33363.0+82129205/MODULE_cbc2f48eb5e8ed6670cf+24e0fc5a/model.neff
15655
  neuronxcc-2.21.33363.0+82129205/MODULE_cbc2f48eb5e8ed6670cf+24e0fc5a/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
15656
  neuronxcc-2.21.33363.0+82129205/MODULE_ce5d1578a50df07b8d6e+8981f7e5/model.neff filter=lfs diff=lfs merge=lfs -text
15657
  neuronxcc-2.21.33363.0+82129205/MODULE_ce5d1578a50df07b8d6e+8981f7e5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
15655
  neuronxcc-2.21.33363.0+82129205/MODULE_cbc2f48eb5e8ed6670cf+24e0fc5a/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
15656
  neuronxcc-2.21.33363.0+82129205/MODULE_ce5d1578a50df07b8d6e+8981f7e5/model.neff filter=lfs diff=lfs merge=lfs -text
15657
  neuronxcc-2.21.33363.0+82129205/MODULE_ce5d1578a50df07b8d6e+8981f7e5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
15658
+ neuronxcc-2.21.33363.0+82129205/MODULE_a2f92cc081ac96561d06+fb4cc044/model.neff filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev1/af58eb15d8e02338dc2f2e880e9c6ec803a98278914b3606acdcc252e7e18429/84e04c656de22f121b56.json ADDED
@@ -0,0 +1,95 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen3-Embedding-8B",
4
+ "_task": "feature-extraction",
5
+ "architectures": [
6
+ "Qwen3ForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "dtype": "bfloat16",
11
+ "head_dim": 128,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 4096,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 12288,
16
+ "layer_types": [
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
+ ],
54
+ "max_position_embeddings": 40960,
55
+ "max_window_layers": 36,
56
+ "model_type": "qwen3",
57
+ "neuron": {
58
+ "_serialized_key": "NxDNeuronConfig",
59
+ "batch_size": 1,
60
+ "capacity_factor": null,
61
+ "checkpoint_id": "Qwen/Qwen3-Embedding-8B",
62
+ "checkpoint_revision": "1d8ad4ca9b3dd8059ad90a75d4983776a23d44af",
63
+ "continuous_batching": false,
64
+ "ep_degree": 1,
65
+ "fused_qkv": true,
66
+ "glu_mlp": true,
67
+ "local_ranks_size": 2,
68
+ "max_batch_size": 1,
69
+ "max_context_length": 16384,
70
+ "max_topk": 256,
71
+ "n_active_tokens": 16384,
72
+ "neuronxcc_version": "2.21.33363.0+82129205",
73
+ "on_device_sampling": false,
74
+ "optimum_neuron_version": "0.4.5.dev1",
75
+ "output_logits": false,
76
+ "pp_degree": 1,
77
+ "sequence_length": 16384,
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": 151665
95
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev1/qwen3/Qwen/Qwen3-Embedding-8B/84e04c656de22f121b56.json ADDED
@@ -0,0 +1,95 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen3-Embedding-8B",
4
+ "_task": "feature-extraction",
5
+ "architectures": [
6
+ "Qwen3ForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "dtype": "bfloat16",
11
+ "head_dim": 128,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 4096,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 12288,
16
+ "layer_types": [
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
+ ],
54
+ "max_position_embeddings": 40960,
55
+ "max_window_layers": 36,
56
+ "model_type": "qwen3",
57
+ "neuron": {
58
+ "_serialized_key": "NxDNeuronConfig",
59
+ "batch_size": 1,
60
+ "capacity_factor": null,
61
+ "checkpoint_id": "Qwen/Qwen3-Embedding-8B",
62
+ "checkpoint_revision": "1d8ad4ca9b3dd8059ad90a75d4983776a23d44af",
63
+ "continuous_batching": false,
64
+ "ep_degree": 1,
65
+ "fused_qkv": true,
66
+ "glu_mlp": true,
67
+ "local_ranks_size": 2,
68
+ "max_batch_size": 1,
69
+ "max_context_length": 16384,
70
+ "max_topk": 256,
71
+ "n_active_tokens": 16384,
72
+ "neuronxcc_version": "2.21.33363.0+82129205",
73
+ "on_device_sampling": false,
74
+ "optimum_neuron_version": "0.4.5.dev1",
75
+ "output_logits": false,
76
+ "pp_degree": 1,
77
+ "sequence_length": 16384,
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": 151665
95
+ }
neuronxcc-2.21.33363.0+82129205/MODULE_a2f92cc081ac96561d06+fb4cc044/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/encoding/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.21.33363.0+82129205/MODULE_a2f92cc081ac96561d06+fb4cc044/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_a2f92cc081ac96561d06+fb4cc044/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c7ed34618c50dc80082ee33f137a59406abb4d2813188909370832d1da8f6f46
3
+ size 857100
neuronxcc-2.21.33363.0+82129205/MODULE_a2f92cc081ac96561d06+fb4cc044/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:928e8600b4b9fd183b8217df12c67fd03cac1eed0630abc49b0ac44bf7b0f3e7
3
+ size 38267904