Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
 Community
709
dacorvo HF Staff commited on
Commit
f523ac1
·
verified ·
1 Parent(s): 74771bb

Synchronizing local compiler cache.

Browse files
Files changed (12) hide show
  1. .gitattributes +3 -0
  2. neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/1f0ace940b2be4a6ff910ece9c5246b9189f14b7a2696a82afebbf71ff3339a3/80ace577c8a3335bd6b9.json +124 -0
  3. neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/qwen3/Qwen/Qwen3-32B/80ace577c8a3335bd6b9.json +124 -0
  4. neuronxcc-2.26.6360.0+6f180f47/MODULE_df4a508f7287d86ba784+84337dd9/compile_flags.json +1 -0
  5. neuronxcc-2.26.6360.0+6f180f47/MODULE_df4a508f7287d86ba784+84337dd9/model.done +0 -0
  6. neuronxcc-2.26.6360.0+6f180f47/MODULE_df4a508f7287d86ba784+84337dd9/model.hlo_module.pb +3 -0
  7. neuronxcc-2.26.6360.0+6f180f47/MODULE_df4a508f7287d86ba784+84337dd9/model.neff +3 -0
  8. neuronxcc-2.26.6360.0+6f180f47/MODULE_df4a508f7287d86ba784+84337dd9/wrapped_neff.hlo +3 -0
  9. neuronxcc-2.26.6360.0+6f180f47/MODULE_efb768f296a8d23dcb5b+12dd9119/compile_flags.json +1 -0
  10. neuronxcc-2.26.6360.0+6f180f47/MODULE_efb768f296a8d23dcb5b+12dd9119/model.done +0 -0
  11. neuronxcc-2.26.6360.0+6f180f47/MODULE_efb768f296a8d23dcb5b+12dd9119/model.hlo_module.pb +3 -0
  12. neuronxcc-2.26.6360.0+6f180f47/MODULE_efb768f296a8d23dcb5b+12dd9119/model.neff +3 -0
.gitattributes CHANGED
@@ -17954,3 +17954,6 @@ neuronxcc-2.26.6360.0+6f180f47/MODULE_d98f5bd7620fb6537805+84337dd9/wrapped_neff
17954
  neuronxcc-2.26.6360.0+6f180f47/MODULE_a160415949d6715b2f20+2a395fce/model.neff filter=lfs diff=lfs merge=lfs -text
17955
  neuronxcc-2.26.6360.0+6f180f47/MODULE_f0714b23325a2fa8caf5+dbd8466c/model.neff filter=lfs diff=lfs merge=lfs -text
17956
  neuronxcc-2.26.6360.0+6f180f47/MODULE_f0714b23325a2fa8caf5+dbd8466c/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
17954
  neuronxcc-2.26.6360.0+6f180f47/MODULE_a160415949d6715b2f20+2a395fce/model.neff filter=lfs diff=lfs merge=lfs -text
17955
  neuronxcc-2.26.6360.0+6f180f47/MODULE_f0714b23325a2fa8caf5+dbd8466c/model.neff filter=lfs diff=lfs merge=lfs -text
17956
  neuronxcc-2.26.6360.0+6f180f47/MODULE_f0714b23325a2fa8caf5+dbd8466c/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
17957
+ neuronxcc-2.26.6360.0+6f180f47/MODULE_df4a508f7287d86ba784+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
17958
+ neuronxcc-2.26.6360.0+6f180f47/MODULE_df4a508f7287d86ba784+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
17959
+ neuronxcc-2.26.6360.0+6f180f47/MODULE_efb768f296a8d23dcb5b+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/1f0ace940b2be4a6ff910ece9c5246b9189f14b7a2696a82afebbf71ff3339a3/80ace577c8a3335bd6b9.json ADDED
@@ -0,0 +1,124 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "dtype": "bfloat16",
11
+ "head_dim": 128,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 5120,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 25600,
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
+ "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
+ "full_attention"
81
+ ],
82
+ "max_position_embeddings": 40960,
83
+ "max_window_layers": 64,
84
+ "model_type": "qwen3",
85
+ "neuron": {
86
+ "_serialized_key": "NxDNeuronConfig",
87
+ "batch_size": 1,
88
+ "capacity_factor": null,
89
+ "checkpoint_id": "Qwen/Qwen3-32B",
90
+ "checkpoint_revision": "9216db5781bf21249d130ec9da846c4624c16137",
91
+ "continuous_batching": false,
92
+ "ep_degree": 1,
93
+ "fused_qkv": true,
94
+ "glu_mlp": true,
95
+ "local_ranks_size": 8,
96
+ "max_batch_size": 1,
97
+ "max_context_length": 4096,
98
+ "max_topk": 256,
99
+ "n_active_tokens": 4096,
100
+ "neuronxcc_version": "2.26.6360.0+6f180f47",
101
+ "on_device_sampling": true,
102
+ "optimum_neuron_version": "0.4.7.dev0",
103
+ "output_logits": false,
104
+ "pp_degree": 1,
105
+ "prefill_chunk_size": 1024,
106
+ "sequence_length": 4096,
107
+ "speculation_length": 0,
108
+ "start_rank_id": 0,
109
+ "target": "trn1",
110
+ "torch_dtype": "bfloat16",
111
+ "tp_degree": 8
112
+ },
113
+ "num_attention_heads": 64,
114
+ "num_hidden_layers": 64,
115
+ "num_key_value_heads": 8,
116
+ "rms_norm_eps": 1e-06,
117
+ "rope_scaling": null,
118
+ "rope_theta": 1000000,
119
+ "sliding_window": null,
120
+ "tie_word_embeddings": false,
121
+ "use_cache": true,
122
+ "use_sliding_window": false,
123
+ "vocab_size": 151936
124
+ }
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/qwen3/Qwen/Qwen3-32B/80ace577c8a3335bd6b9.json ADDED
@@ -0,0 +1,124 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "dtype": "bfloat16",
11
+ "head_dim": 128,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 5120,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 25600,
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
+ "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
+ "full_attention"
81
+ ],
82
+ "max_position_embeddings": 40960,
83
+ "max_window_layers": 64,
84
+ "model_type": "qwen3",
85
+ "neuron": {
86
+ "_serialized_key": "NxDNeuronConfig",
87
+ "batch_size": 1,
88
+ "capacity_factor": null,
89
+ "checkpoint_id": "Qwen/Qwen3-32B",
90
+ "checkpoint_revision": "9216db5781bf21249d130ec9da846c4624c16137",
91
+ "continuous_batching": false,
92
+ "ep_degree": 1,
93
+ "fused_qkv": true,
94
+ "glu_mlp": true,
95
+ "local_ranks_size": 8,
96
+ "max_batch_size": 1,
97
+ "max_context_length": 4096,
98
+ "max_topk": 256,
99
+ "n_active_tokens": 4096,
100
+ "neuronxcc_version": "2.26.6360.0+6f180f47",
101
+ "on_device_sampling": true,
102
+ "optimum_neuron_version": "0.4.7.dev0",
103
+ "output_logits": false,
104
+ "pp_degree": 1,
105
+ "prefill_chunk_size": 1024,
106
+ "sequence_length": 4096,
107
+ "speculation_length": 0,
108
+ "start_rank_id": 0,
109
+ "target": "trn1",
110
+ "torch_dtype": "bfloat16",
111
+ "tp_degree": 8
112
+ },
113
+ "num_attention_heads": 64,
114
+ "num_hidden_layers": 64,
115
+ "num_key_value_heads": 8,
116
+ "rms_norm_eps": 1e-06,
117
+ "rope_scaling": null,
118
+ "rope_theta": 1000000,
119
+ "sliding_window": null,
120
+ "tie_word_embeddings": false,
121
+ "use_cache": true,
122
+ "use_sliding_window": false,
123
+ "vocab_size": 151936
124
+ }
neuronxcc-2.26.6360.0+6f180f47/MODULE_df4a508f7287d86ba784+84337dd9/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", "--verbose=35", "--logfile=/tmp/nxd_model/token_generation/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.26.6360.0+6f180f47/MODULE_df4a508f7287d86ba784+84337dd9/model.done ADDED
File without changes
neuronxcc-2.26.6360.0+6f180f47/MODULE_df4a508f7287d86ba784+84337dd9/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:920a421e532ae5edb6c32ec247e66b73849c168b449c4e0c8dacd8aca5859376
3
+ size 6513872
neuronxcc-2.26.6360.0+6f180f47/MODULE_df4a508f7287d86ba784+84337dd9/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2ff52c866790f93647fbc53c2598c808ae9b6271488a8db3e15a2b92a36b4774
3
+ size 9339904
neuronxcc-2.26.6360.0+6f180f47/MODULE_df4a508f7287d86ba784+84337dd9/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:744ce17c3c71b8b9b32c94dd671527e49b3e3d8cc49c78f9362268c1b090eb2e
3
+ size 9650962
neuronxcc-2.26.6360.0+6f180f47/MODULE_efb768f296a8d23dcb5b+12dd9119/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", "--verbose=35", "--logfile=/tmp/nxd_model/chunked_prefill/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.26.6360.0+6f180f47/MODULE_efb768f296a8d23dcb5b+12dd9119/model.done ADDED
File without changes
neuronxcc-2.26.6360.0+6f180f47/MODULE_efb768f296a8d23dcb5b+12dd9119/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2d684878fab35f49500aeebb6784232af08c23d2d0fef13be86c5dcc6ccbc9f1
3
+ size 8631620
neuronxcc-2.26.6360.0+6f180f47/MODULE_efb768f296a8d23dcb5b+12dd9119/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4f59be90a4c4b0bd6b9a59786cba491b9f792545795ab9b5b59d728464f9c693
3
+ size 3298304