Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
Community
630
dacorvo HF Staff commited on
Commit
4968af9
·
verified ·
1 Parent(s): af50d1e

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.dev1/qwen3/Qwen/Qwen3-0.6B/f708785aa70c5f42e849.json +88 -0
  3. neuronxcc-2.21.33363.0+82129205/MODULE_1944f672afeaa6480c61+24129607/compile_flags.json +1 -0
  4. neuronxcc-2.21.33363.0+82129205/MODULE_1944f672afeaa6480c61+24129607/model.done +0 -0
  5. neuronxcc-2.21.33363.0+82129205/MODULE_1944f672afeaa6480c61+24129607/model.hlo_module.pb +3 -0
  6. neuronxcc-2.21.33363.0+82129205/MODULE_1944f672afeaa6480c61+24129607/model.neff +3 -0
  7. neuronxcc-2.21.33363.0+82129205/MODULE_b9b261eb8d39678d6323+a02c3a36/compile_flags.json +1 -0
  8. neuronxcc-2.21.33363.0+82129205/MODULE_b9b261eb8d39678d6323+a02c3a36/model.done +0 -0
  9. neuronxcc-2.21.33363.0+82129205/MODULE_b9b261eb8d39678d6323+a02c3a36/model.hlo_module.pb +3 -0
  10. neuronxcc-2.21.33363.0+82129205/MODULE_b9b261eb8d39678d6323+a02c3a36/model.neff +3 -0
  11. neuronxcc-2.21.33363.0+82129205/MODULE_b9b261eb8d39678d6323+a02c3a36/wrapped_neff.hlo +3 -0
.gitattributes CHANGED
@@ -14619,3 +14619,6 @@ neuronxcc-2.21.33363.0+82129205/MODULE_95798093ff9a2909405f+c66944d2/model.neff
14619
  neuronxcc-2.21.33363.0+82129205/MODULE_95798093ff9a2909405f+c66944d2/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
14620
  neuronxcc-2.21.33363.0+82129205/MODULE_b49d0bf2a78d20a40bef+77bd7d82/model.neff filter=lfs diff=lfs merge=lfs -text
14621
  neuronxcc-2.21.33363.0+82129205/MODULE_b49d0bf2a78d20a40bef+77bd7d82/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
14619
  neuronxcc-2.21.33363.0+82129205/MODULE_95798093ff9a2909405f+c66944d2/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
14620
  neuronxcc-2.21.33363.0+82129205/MODULE_b49d0bf2a78d20a40bef+77bd7d82/model.neff filter=lfs diff=lfs merge=lfs -text
14621
  neuronxcc-2.21.33363.0+82129205/MODULE_b49d0bf2a78d20a40bef+77bd7d82/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
14622
+ neuronxcc-2.21.33363.0+82129205/MODULE_1944f672afeaa6480c61+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
14623
+ neuronxcc-2.21.33363.0+82129205/MODULE_b9b261eb8d39678d6323+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
14624
+ neuronxcc-2.21.33363.0+82129205/MODULE_b9b261eb8d39678d6323+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev1/qwen3/Qwen/Qwen3-0.6B/f708785aa70c5f42e849.json ADDED
@@ -0,0 +1,88 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": 4,
52
+ "capacity_factor": null,
53
+ "checkpoint_id": "Qwen/Qwen3-0.6B",
54
+ "checkpoint_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
55
+ "continuous_batching": true,
56
+ "ep_degree": 1,
57
+ "fused_qkv": true,
58
+ "glu_mlp": true,
59
+ "local_ranks_size": 2,
60
+ "max_batch_size": 4,
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": false,
66
+ "optimum_neuron_version": "0.4.4.dev1",
67
+ "output_logits": false,
68
+ "pp_degree": 1,
69
+ "sequence_length": 4096,
70
+ "sequence_parallel_enabled": false,
71
+ "speculation_length": 0,
72
+ "start_rank_id": 0,
73
+ "target": "trn1",
74
+ "torch_dtype": "bfloat16",
75
+ "tp_degree": 2
76
+ },
77
+ "num_attention_heads": 16,
78
+ "num_hidden_layers": 28,
79
+ "num_key_value_heads": 8,
80
+ "rms_norm_eps": 1e-06,
81
+ "rope_scaling": null,
82
+ "rope_theta": 1000000,
83
+ "sliding_window": null,
84
+ "tie_word_embeddings": true,
85
+ "use_cache": true,
86
+ "use_sliding_window": false,
87
+ "vocab_size": 151936
88
+ }
neuronxcc-2.21.33363.0+82129205/MODULE_1944f672afeaa6480c61+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_1944f672afeaa6480c61+24129607/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_1944f672afeaa6480c61+24129607/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b77f0fd54de2b3f0e4d80ee388867e6ab1f396c5fd6c14be411f63d198649118
3
+ size 865460
neuronxcc-2.21.33363.0+82129205/MODULE_1944f672afeaa6480c61+24129607/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c63a961d38d6e2faccfb3057e2a3a8ac96875fbd1ade15ba62f09333cfa1e1f0
3
+ size 36148224
neuronxcc-2.21.33363.0+82129205/MODULE_b9b261eb8d39678d6323+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_b9b261eb8d39678d6323+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_b9b261eb8d39678d6323+a02c3a36/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a009da1b0f420f4556b03630a8156f63f7d7fd03c8a41d7a3beb0139fb04f6fe
3
+ size 711424
neuronxcc-2.21.33363.0+82129205/MODULE_b9b261eb8d39678d6323+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:51733771a690420ffa86450e03c4581d1ae1cbd5c1ad207bd5b6521e3c6fa30c
3
+ size 3503104
neuronxcc-2.21.33363.0+82129205/MODULE_b9b261eb8d39678d6323+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a945dbca161bd9f96c79046d5127698c1f0f5aca4abaecd22f1388b6274a01f2
3
+ size 3640161