Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
Community
653
dacorvo HF Staff commited on
Commit
e227b67
·
verified ·
1 Parent(s): 9461bd6

Synchronizing local compiler cache.

Browse files
Files changed (11) hide show
  1. .gitattributes +3 -0
  2. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev2/qwen3/Qwen/Qwen3-0.6B/af09f014db5aef044c35.json +87 -0
  3. neuronxcc-2.21.33363.0+82129205/MODULE_82d8f27b6a22b72e8871+24129607/compile_flags.json +1 -0
  4. neuronxcc-2.21.33363.0+82129205/MODULE_82d8f27b6a22b72e8871+24129607/model.done +0 -0
  5. neuronxcc-2.21.33363.0+82129205/MODULE_82d8f27b6a22b72e8871+24129607/model.hlo_module.pb +3 -0
  6. neuronxcc-2.21.33363.0+82129205/MODULE_82d8f27b6a22b72e8871+24129607/model.neff +3 -0
  7. neuronxcc-2.21.33363.0+82129205/MODULE_f84a4331ebb0ee31a2c2+a02c3a36/compile_flags.json +1 -0
  8. neuronxcc-2.21.33363.0+82129205/MODULE_f84a4331ebb0ee31a2c2+a02c3a36/model.done +0 -0
  9. neuronxcc-2.21.33363.0+82129205/MODULE_f84a4331ebb0ee31a2c2+a02c3a36/model.hlo_module.pb +3 -0
  10. neuronxcc-2.21.33363.0+82129205/MODULE_f84a4331ebb0ee31a2c2+a02c3a36/model.neff +3 -0
  11. neuronxcc-2.21.33363.0+82129205/MODULE_f84a4331ebb0ee31a2c2+a02c3a36/wrapped_neff.hlo +3 -0
.gitattributes CHANGED
@@ -15292,3 +15292,6 @@ neuronxcc-2.21.33363.0+82129205/MODULE_b6029c55c0c07bbb5b23+a02c3a36/wrapped_nef
15292
  neuronxcc-2.21.33363.0+82129205/MODULE_677af38c5f56c2ad4368+283df001/model.neff filter=lfs diff=lfs merge=lfs -text
15293
  neuronxcc-2.21.33363.0+82129205/MODULE_8432c515789df57ec824+2dde74c7/model.neff filter=lfs diff=lfs merge=lfs -text
15294
  neuronxcc-2.21.33363.0+82129205/MODULE_8432c515789df57ec824+2dde74c7/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
15292
  neuronxcc-2.21.33363.0+82129205/MODULE_677af38c5f56c2ad4368+283df001/model.neff filter=lfs diff=lfs merge=lfs -text
15293
  neuronxcc-2.21.33363.0+82129205/MODULE_8432c515789df57ec824+2dde74c7/model.neff filter=lfs diff=lfs merge=lfs -text
15294
  neuronxcc-2.21.33363.0+82129205/MODULE_8432c515789df57ec824+2dde74c7/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
15295
+ neuronxcc-2.21.33363.0+82129205/MODULE_82d8f27b6a22b72e8871+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
15296
+ neuronxcc-2.21.33363.0+82129205/MODULE_f84a4331ebb0ee31a2c2+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
15297
+ neuronxcc-2.21.33363.0+82129205/MODULE_f84a4331ebb0ee31a2c2+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev2/qwen3/Qwen/Qwen3-0.6B/af09f014db5aef044c35.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "dtype": "bfloat16",
11
+ "head_dim": 128,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 1024,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 3072,
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
+ ],
46
+ "max_position_embeddings": 40960,
47
+ "max_window_layers": 28,
48
+ "model_type": "qwen3",
49
+ "neuron": {
50
+ "_serialized_key": "NxDNeuronConfig",
51
+ "batch_size": 1,
52
+ "capacity_factor": null,
53
+ "checkpoint_id": "Qwen/Qwen3-0.6B",
54
+ "checkpoint_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
55
+ "continuous_batching": false,
56
+ "ep_degree": 1,
57
+ "fused_qkv": true,
58
+ "glu_mlp": true,
59
+ "local_ranks_size": 2,
60
+ "max_batch_size": 1,
61
+ "max_context_length": 4096,
62
+ "max_topk": 256,
63
+ "n_active_tokens": 4096,
64
+ "neuronxcc_version": "2.21.33363.0+82129205",
65
+ "on_device_sampling": true,
66
+ "optimum_neuron_version": "0.4.4.dev2",
67
+ "output_logits": false,
68
+ "pp_degree": 1,
69
+ "sequence_length": 4096,
70
+ "speculation_length": 0,
71
+ "start_rank_id": 0,
72
+ "target": "trn1",
73
+ "torch_dtype": "bfloat16",
74
+ "tp_degree": 2
75
+ },
76
+ "num_attention_heads": 16,
77
+ "num_hidden_layers": 28,
78
+ "num_key_value_heads": 8,
79
+ "rms_norm_eps": 1e-06,
80
+ "rope_scaling": null,
81
+ "rope_theta": 1000000,
82
+ "sliding_window": null,
83
+ "tie_word_embeddings": true,
84
+ "use_cache": true,
85
+ "use_sliding_window": false,
86
+ "vocab_size": 151936
87
+ }
neuronxcc-2.21.33363.0+82129205/MODULE_82d8f27b6a22b72e8871+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_82d8f27b6a22b72e8871+24129607/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_82d8f27b6a22b72e8871+24129607/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:853f6f3e40e4fe2ad039b7fb10118e5aeabe0c7d064b0e851d26c37a2f3e65f1
3
+ size 843918
neuronxcc-2.21.33363.0+82129205/MODULE_82d8f27b6a22b72e8871+24129607/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b021273027a141f3523c1822b611989e9a036b61562f94be5abb81f425826a84
3
+ size 35288064
neuronxcc-2.21.33363.0+82129205/MODULE_f84a4331ebb0ee31a2c2+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_f84a4331ebb0ee31a2c2+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_f84a4331ebb0ee31a2c2+a02c3a36/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1dd1d859f2ae22e9227e06bff542d143a215abb98684b909e0bec87f33efd817
3
+ size 749861
neuronxcc-2.21.33363.0+82129205/MODULE_f84a4331ebb0ee31a2c2+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d68a48a85fa333f0be85d224b6ecab2c311dbd221ec11e9409c84c9fc1035706
3
+ size 2008064
neuronxcc-2.21.33363.0+82129205/MODULE_f84a4331ebb0ee31a2c2+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:566cf236dcb76853fe5dde39ab1eefb92ef05f31c9d534c1e7cf306a749a4408
3
+ size 2145229