Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
Community
603
jburtoft commited on
Commit
4ea822c
·
verified ·
1 Parent(s): 17940c5

Synchronizing local compiler cache.

Browse files
Files changed (19) hide show
  1. .gitattributes +3 -0
  2. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.1/qwen3/Qwen/Qwen3-0.6B/7fbf409eea4fd3fa0cfc.json +86 -0
  3. neuronxcc-2.21.33363.0+82129205/MODULE_15786686745870740384+3cc9a3cb/compile_flags.json +1 -0
  4. neuronxcc-2.21.33363.0+82129205/MODULE_15786686745870740384+3cc9a3cb/model.done +0 -0
  5. neuronxcc-2.21.33363.0+82129205/MODULE_15786686745870740384+3cc9a3cb/model.hlo_module.pb +3 -0
  6. neuronxcc-2.21.33363.0+82129205/MODULE_15786686745870740384+3cc9a3cb/model.neff +0 -0
  7. neuronxcc-2.21.33363.0+82129205/MODULE_15804984955746879243+3cc9a3cb/compile_flags.json +1 -0
  8. neuronxcc-2.21.33363.0+82129205/MODULE_15804984955746879243+3cc9a3cb/model.done +0 -0
  9. neuronxcc-2.21.33363.0+82129205/MODULE_15804984955746879243+3cc9a3cb/model.hlo_module.pb +3 -0
  10. neuronxcc-2.21.33363.0+82129205/MODULE_15804984955746879243+3cc9a3cb/model.neff +0 -0
  11. neuronxcc-2.21.33363.0+82129205/MODULE_7ea07b6bc9f4a05c97f0+24129607/compile_flags.json +1 -0
  12. neuronxcc-2.21.33363.0+82129205/MODULE_7ea07b6bc9f4a05c97f0+24129607/model.done +0 -0
  13. neuronxcc-2.21.33363.0+82129205/MODULE_7ea07b6bc9f4a05c97f0+24129607/model.hlo_module.pb +3 -0
  14. neuronxcc-2.21.33363.0+82129205/MODULE_7ea07b6bc9f4a05c97f0+24129607/model.neff +3 -0
  15. neuronxcc-2.21.33363.0+82129205/MODULE_b6ab468f70deaeb5f407+a02c3a36/compile_flags.json +1 -0
  16. neuronxcc-2.21.33363.0+82129205/MODULE_b6ab468f70deaeb5f407+a02c3a36/model.done +0 -0
  17. neuronxcc-2.21.33363.0+82129205/MODULE_b6ab468f70deaeb5f407+a02c3a36/model.hlo_module.pb +3 -0
  18. neuronxcc-2.21.33363.0+82129205/MODULE_b6ab468f70deaeb5f407+a02c3a36/model.neff +3 -0
  19. neuronxcc-2.21.33363.0+82129205/MODULE_b6ab468f70deaeb5f407+a02c3a36/wrapped_neff.hlo +3 -0
.gitattributes CHANGED
@@ -12510,3 +12510,6 @@ neuronxcc-2.21.18209.0+043b1bf7/MODULE_8623340492324789537+bad9cf09/model.neff f
12510
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_87301aa5994aa3ec3488+ac10809c/model.neff filter=lfs diff=lfs merge=lfs -text
12511
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_87301aa5994aa3ec3488+ac10809c/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
12512
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_d520c8be8d8ce950a101+a32116a7/model.neff filter=lfs diff=lfs merge=lfs -text
 
 
 
 
12510
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_87301aa5994aa3ec3488+ac10809c/model.neff filter=lfs diff=lfs merge=lfs -text
12511
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_87301aa5994aa3ec3488+ac10809c/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
12512
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_d520c8be8d8ce950a101+a32116a7/model.neff filter=lfs diff=lfs merge=lfs -text
12513
+ neuronxcc-2.21.33363.0+82129205/MODULE_7ea07b6bc9f4a05c97f0+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
12514
+ neuronxcc-2.21.33363.0+82129205/MODULE_b6ab468f70deaeb5f407+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
12515
+ neuronxcc-2.21.33363.0+82129205/MODULE_b6ab468f70deaeb5f407+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.1/qwen3/Qwen/Qwen3-0.6B/7fbf409eea4fd3fa0cfc.json ADDED
@@ -0,0 +1,86 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen3-0.6B",
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": 1024,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 3072,
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
+ ],
45
+ "max_position_embeddings": 40960,
46
+ "max_window_layers": 28,
47
+ "model_type": "qwen3",
48
+ "neuron": {
49
+ "_serialized_key": "NxDNeuronConfig",
50
+ "batch_size": 4,
51
+ "capacity_factor": null,
52
+ "checkpoint_id": "Qwen/Qwen3-0.6B",
53
+ "checkpoint_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
54
+ "continuous_batching": true,
55
+ "ep_degree": 1,
56
+ "fused_qkv": true,
57
+ "glu_mlp": true,
58
+ "local_ranks_size": 1,
59
+ "max_batch_size": 4,
60
+ "max_context_length": 512,
61
+ "max_topk": 256,
62
+ "n_active_tokens": 512,
63
+ "neuronxcc_version": "2.21.33363.0+82129205",
64
+ "on_device_sampling": true,
65
+ "optimum_neuron_version": "0.4.1",
66
+ "output_logits": false,
67
+ "pp_degree": 1,
68
+ "sequence_length": 512,
69
+ "speculation_length": 0,
70
+ "start_rank_id": 0,
71
+ "target": "trn1",
72
+ "torch_dtype": "bfloat16",
73
+ "tp_degree": 1
74
+ },
75
+ "num_attention_heads": 16,
76
+ "num_hidden_layers": 28,
77
+ "num_key_value_heads": 8,
78
+ "rms_norm_eps": 1e-06,
79
+ "rope_scaling": null,
80
+ "rope_theta": 1000000,
81
+ "sliding_window": null,
82
+ "tie_word_embeddings": true,
83
+ "use_cache": true,
84
+ "use_sliding_window": false,
85
+ "vocab_size": 151936
86
+ }
neuronxcc-2.21.33363.0+82129205/MODULE_15786686745870740384+3cc9a3cb/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn1", "--model-type=transformer"]
neuronxcc-2.21.33363.0+82129205/MODULE_15786686745870740384+3cc9a3cb/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_15786686745870740384+3cc9a3cb/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d5415e070fef2b0f82ab8dc83a1b7235ee56907766147007335809b59d549572
3
+ size 1810
neuronxcc-2.21.33363.0+82129205/MODULE_15786686745870740384+3cc9a3cb/model.neff ADDED
Binary file (31.7 kB). View file
 
neuronxcc-2.21.33363.0+82129205/MODULE_15804984955746879243+3cc9a3cb/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn1", "--model-type=transformer"]
neuronxcc-2.21.33363.0+82129205/MODULE_15804984955746879243+3cc9a3cb/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_15804984955746879243+3cc9a3cb/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ed0b10113b065f9981de47571651f2ef421e5b6c355d83a034a5803d07ea61ff
3
+ size 346
neuronxcc-2.21.33363.0+82129205/MODULE_15804984955746879243+3cc9a3cb/model.neff ADDED
Binary file (11.3 kB). View file
 
neuronxcc-2.21.33363.0+82129205/MODULE_7ea07b6bc9f4a05c97f0+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-2.21.33363.0+82129205/MODULE_7ea07b6bc9f4a05c97f0+24129607/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_7ea07b6bc9f4a05c97f0+24129607/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7c1741d4b49235895bda890c574a9db0af00cef1c86c5bb1d9cfa908abae99ae
3
+ size 684042
neuronxcc-2.21.33363.0+82129205/MODULE_7ea07b6bc9f4a05c97f0+24129607/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:50c41b10419f2a2fde7d9a69bdfcbc00d3cecf2cf784967a905d7176d821e8bd
3
+ size 3748864
neuronxcc-2.21.33363.0+82129205/MODULE_b6ab468f70deaeb5f407+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-2.21.33363.0+82129205/MODULE_b6ab468f70deaeb5f407+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_b6ab468f70deaeb5f407+a02c3a36/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ed55c765c80773e8662c242cf086af79521a55a779f76f7a58ed6dc2bb78951a
3
+ size 694668
neuronxcc-2.21.33363.0+82129205/MODULE_b6ab468f70deaeb5f407+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9327e3b58f781201118c19e0d4ca2d325a410aa856992deedeeca0c8f62ca911
3
+ size 2868224
neuronxcc-2.21.33363.0+82129205/MODULE_b6ab468f70deaeb5f407+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6246708a909c92d0f77aabacd9d63cb58feaf47790767c2fb83536cad28e6dc2
3
+ size 3006146