Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
Community
611
dacorvo HF Staff commited on
Commit
8a475f1
·
verified ·
1 Parent(s): 1095503

Synchronizing local compiler cache.

Browse files
Files changed (11) hide show
  1. .gitattributes +3 -0
  2. neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.0.dev0/qwen3/Qwen/Qwen3-0.6B/b50d66486c2cb4406cab.json +87 -0
  3. neuronxcc-2.21.18209.0+043b1bf7/MODULE_19285e519732111e1cab+ed72d204/compile_flags.json +1 -0
  4. neuronxcc-2.21.18209.0+043b1bf7/MODULE_19285e519732111e1cab+ed72d204/model.done +0 -0
  5. neuronxcc-2.21.18209.0+043b1bf7/MODULE_19285e519732111e1cab+ed72d204/model.hlo_module.pb +3 -0
  6. neuronxcc-2.21.18209.0+043b1bf7/MODULE_19285e519732111e1cab+ed72d204/model.neff +3 -0
  7. neuronxcc-2.21.18209.0+043b1bf7/MODULE_75b1e776b16b9953c241+a9d440f5/compile_flags.json +1 -0
  8. neuronxcc-2.21.18209.0+043b1bf7/MODULE_75b1e776b16b9953c241+a9d440f5/model.done +0 -0
  9. neuronxcc-2.21.18209.0+043b1bf7/MODULE_75b1e776b16b9953c241+a9d440f5/model.hlo_module.pb +3 -0
  10. neuronxcc-2.21.18209.0+043b1bf7/MODULE_75b1e776b16b9953c241+a9d440f5/model.neff +3 -0
  11. neuronxcc-2.21.18209.0+043b1bf7/MODULE_75b1e776b16b9953c241+a9d440f5/wrapped_neff.hlo +3 -0
.gitattributes CHANGED
@@ -11572,3 +11572,6 @@ neuronxcc-2.21.18209.0+043b1bf7/MODULE_88724518ccebd3fa6470+ed72d204/model.neff
11572
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_c243c517d60e8dd22317+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11573
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_ea0b08427c20ef00b9e3+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11574
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_ea0b08427c20ef00b9e3+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
11572
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_c243c517d60e8dd22317+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11573
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_ea0b08427c20ef00b9e3+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11574
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_ea0b08427c20ef00b9e3+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
11575
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_19285e519732111e1cab+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11576
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_75b1e776b16b9953c241+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11577
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_75b1e776b16b9953c241+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.0.dev0/qwen3/Qwen/Qwen3-0.6B/b50d66486c2cb4406cab.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
+ "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
+ "enable_bucketing": false,
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": "2.21.18209.0+043b1bf7",
65
+ "on_device_sampling": false,
66
+ "optimum_neuron_version": "0.4.0.dev0",
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-2.21.18209.0+043b1bf7/MODULE_19285e519732111e1cab+ed72d204/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_model/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.21.18209.0+043b1bf7/MODULE_19285e519732111e1cab+ed72d204/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_19285e519732111e1cab+ed72d204/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:deee3234b6eb3e653ddf7a217e0d91225a6a75a693448a7ebcaa5e7130301d96
3
+ size 865460
neuronxcc-2.21.18209.0+043b1bf7/MODULE_19285e519732111e1cab+ed72d204/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:68b5ed3bae6ebe80a4fc01b7a3fece50baed00204d3bf5da169d8e80ba8460ef
3
+ size 36148224
neuronxcc-2.21.18209.0+043b1bf7/MODULE_75b1e776b16b9953c241+a9d440f5/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_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.21.18209.0+043b1bf7/MODULE_75b1e776b16b9953c241+a9d440f5/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_75b1e776b16b9953c241+a9d440f5/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:828b718ac0d544f14a9a683c76b442c49520573027ba2e8347fdd5e79531704b
3
+ size 702870
neuronxcc-2.21.18209.0+043b1bf7/MODULE_75b1e776b16b9953c241+a9d440f5/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f6559e5a0e9e7abbb467c8f487684efdd5133668477d11307cdb0cda351cc0cf
3
+ size 3503104
neuronxcc-2.21.18209.0+043b1bf7/MODULE_75b1e776b16b9953c241+a9d440f5/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bf581891e99ea008b569c7eb60005a98943301047c70a98b1d246619753e4538
3
+ size 3640300