Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
 Community
709
dacorvo HF Staff commited on
Commit
5a7865f
·
verified ·
1 Parent(s): dc4ced8

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.dev0/490ffc6ca91a35b216d2dbdb4ab8960c5c8b886ec8c2aceab930553e16b0eb58/4bd73c8f3cb73fc842f4.json +96 -0
  3. neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/qwen3/Qwen/Qwen3-4B/4bd73c8f3cb73fc842f4.json +96 -0
  4. neuronxcc-2.26.6360.0+6f180f47/MODULE_7be0c41bfa68130d2a26+84337dd9/compile_flags.json +1 -0
  5. neuronxcc-2.26.6360.0+6f180f47/MODULE_7be0c41bfa68130d2a26+84337dd9/model.done +0 -0
  6. neuronxcc-2.26.6360.0+6f180f47/MODULE_7be0c41bfa68130d2a26+84337dd9/model.hlo_module.pb +3 -0
  7. neuronxcc-2.26.6360.0+6f180f47/MODULE_7be0c41bfa68130d2a26+84337dd9/model.neff +3 -0
  8. neuronxcc-2.26.6360.0+6f180f47/MODULE_7be0c41bfa68130d2a26+84337dd9/wrapped_neff.hlo +3 -0
  9. neuronxcc-2.26.6360.0+6f180f47/MODULE_df5ac55d00b940623a44+12dd9119/compile_flags.json +1 -0
  10. neuronxcc-2.26.6360.0+6f180f47/MODULE_df5ac55d00b940623a44+12dd9119/model.done +0 -0
  11. neuronxcc-2.26.6360.0+6f180f47/MODULE_df5ac55d00b940623a44+12dd9119/model.hlo_module.pb +3 -0
  12. neuronxcc-2.26.6360.0+6f180f47/MODULE_df5ac55d00b940623a44+12dd9119/model.neff +3 -0
.gitattributes CHANGED
@@ -17906,3 +17906,6 @@ neuronxcc-2.26.6360.0+6f180f47/MODULE_b44accf8ee64393d2e97+84337dd9/wrapped_neff
17906
  neuronxcc-2.26.6360.0+6f180f47/MODULE_530ad83100cc3dd42356+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
17907
  neuronxcc-2.26.6360.0+6f180f47/MODULE_530ad83100cc3dd42356+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
17908
  neuronxcc-2.26.6360.0+6f180f47/MODULE_5b87c52d288ff0e3b45b+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
 
 
 
 
17906
  neuronxcc-2.26.6360.0+6f180f47/MODULE_530ad83100cc3dd42356+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
17907
  neuronxcc-2.26.6360.0+6f180f47/MODULE_530ad83100cc3dd42356+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
17908
  neuronxcc-2.26.6360.0+6f180f47/MODULE_5b87c52d288ff0e3b45b+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
17909
+ neuronxcc-2.26.6360.0+6f180f47/MODULE_7be0c41bfa68130d2a26+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
17910
+ neuronxcc-2.26.6360.0+6f180f47/MODULE_7be0c41bfa68130d2a26+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
17911
+ neuronxcc-2.26.6360.0+6f180f47/MODULE_df5ac55d00b940623a44+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/490ffc6ca91a35b216d2dbdb4ab8960c5c8b886ec8c2aceab930553e16b0eb58/4bd73c8f3cb73fc842f4.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen3-4B",
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": 2560,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 9728,
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/Qwen3-4B",
62
+ "checkpoint_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
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": 4096,
70
+ "max_topk": 256,
71
+ "n_active_tokens": 4096,
72
+ "neuronxcc_version": "2.26.6360.0+6f180f47",
73
+ "on_device_sampling": true,
74
+ "optimum_neuron_version": "0.4.7.dev0",
75
+ "output_logits": false,
76
+ "pp_degree": 1,
77
+ "prefill_chunk_size": 1024,
78
+ "sequence_length": 4096,
79
+ "speculation_length": 0,
80
+ "start_rank_id": 0,
81
+ "target": "trn1",
82
+ "torch_dtype": "bfloat16",
83
+ "tp_degree": 2
84
+ },
85
+ "num_attention_heads": 32,
86
+ "num_hidden_layers": 36,
87
+ "num_key_value_heads": 8,
88
+ "rms_norm_eps": 1e-06,
89
+ "rope_scaling": null,
90
+ "rope_theta": 1000000,
91
+ "sliding_window": null,
92
+ "tie_word_embeddings": true,
93
+ "use_cache": true,
94
+ "use_sliding_window": false,
95
+ "vocab_size": 151936
96
+ }
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/qwen3/Qwen/Qwen3-4B/4bd73c8f3cb73fc842f4.json ADDED
@@ -0,0 +1,96 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen3-4B",
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": 2560,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 9728,
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/Qwen3-4B",
62
+ "checkpoint_revision": "1cfa9a7208912126459214e8b04321603b3df60c",
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": 4096,
70
+ "max_topk": 256,
71
+ "n_active_tokens": 4096,
72
+ "neuronxcc_version": "2.26.6360.0+6f180f47",
73
+ "on_device_sampling": true,
74
+ "optimum_neuron_version": "0.4.7.dev0",
75
+ "output_logits": false,
76
+ "pp_degree": 1,
77
+ "prefill_chunk_size": 1024,
78
+ "sequence_length": 4096,
79
+ "speculation_length": 0,
80
+ "start_rank_id": 0,
81
+ "target": "trn1",
82
+ "torch_dtype": "bfloat16",
83
+ "tp_degree": 2
84
+ },
85
+ "num_attention_heads": 32,
86
+ "num_hidden_layers": 36,
87
+ "num_key_value_heads": 8,
88
+ "rms_norm_eps": 1e-06,
89
+ "rope_scaling": null,
90
+ "rope_theta": 1000000,
91
+ "sliding_window": null,
92
+ "tie_word_embeddings": true,
93
+ "use_cache": true,
94
+ "use_sliding_window": false,
95
+ "vocab_size": 151936
96
+ }
neuronxcc-2.26.6360.0+6f180f47/MODULE_7be0c41bfa68130d2a26+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_7be0c41bfa68130d2a26+84337dd9/model.done ADDED
File without changes
neuronxcc-2.26.6360.0+6f180f47/MODULE_7be0c41bfa68130d2a26+84337dd9/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bb4045142a5cf045e97fcd66f1e1a97fcb864259178cba90fa149f6a55e13dde
3
+ size 3751896
neuronxcc-2.26.6360.0+6f180f47/MODULE_7be0c41bfa68130d2a26+84337dd9/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2e6220523589c4dadcb1dfd5c7fe41c76894c801fb6b08e7ad6dc0a7819fcbbe
3
+ size 6677504
neuronxcc-2.26.6360.0+6f180f47/MODULE_7be0c41bfa68130d2a26+84337dd9/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:741538bbed0de2f0eddc80e3fd2ad6a7bd4e4d9e96216cbf47ed5697297f056a
3
+ size 6882288
neuronxcc-2.26.6360.0+6f180f47/MODULE_df5ac55d00b940623a44+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_df5ac55d00b940623a44+12dd9119/model.done ADDED
File without changes
neuronxcc-2.26.6360.0+6f180f47/MODULE_df5ac55d00b940623a44+12dd9119/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e3f89cbc575cad799413aafd10747f9a78a0dee89cb62db10e2e569d5c1fad65
3
+ size 4983080
neuronxcc-2.26.6360.0+6f180f47/MODULE_df5ac55d00b940623a44+12dd9119/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:47234b96ba107fdd08396f95e22ee257729d47cbb47726a390800b27398363d0
3
+ size 4721664