Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
 Community
679
dacorvo HF Staff commited on
Commit
c3af7a1
·
verified ·
1 Parent(s): 548d308

Synchronizing local compiler cache.

Browse files
Files changed (11) hide show
  1. .gitattributes +3 -0
  2. neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.1.dev0/qwen3/Qwen/Qwen3-0.6B/af16941c15e5dc76a5aa.json +86 -0
  3. neuronxcc-2.21.18209.0+043b1bf7/MODULE_8414d37eab328896acb7+a02c3a36/compile_flags.json +1 -0
  4. neuronxcc-2.21.18209.0+043b1bf7/MODULE_8414d37eab328896acb7+a02c3a36/model.done +0 -0
  5. neuronxcc-2.21.18209.0+043b1bf7/MODULE_8414d37eab328896acb7+a02c3a36/model.hlo_module.pb +3 -0
  6. neuronxcc-2.21.18209.0+043b1bf7/MODULE_8414d37eab328896acb7+a02c3a36/model.neff +3 -0
  7. neuronxcc-2.21.18209.0+043b1bf7/MODULE_8414d37eab328896acb7+a02c3a36/wrapped_neff.hlo +3 -0
  8. neuronxcc-2.21.18209.0+043b1bf7/MODULE_c0808787f3555a6627cc+24129607/compile_flags.json +1 -0
  9. neuronxcc-2.21.18209.0+043b1bf7/MODULE_c0808787f3555a6627cc+24129607/model.done +0 -0
  10. neuronxcc-2.21.18209.0+043b1bf7/MODULE_c0808787f3555a6627cc+24129607/model.hlo_module.pb +3 -0
  11. neuronxcc-2.21.18209.0+043b1bf7/MODULE_c0808787f3555a6627cc+24129607/model.neff +3 -0
.gitattributes CHANGED
@@ -12343,3 +12343,6 @@ neuronxcc-2.21.18209.0+043b1bf7/MODULE_c3e59a1b9c9219843839+24129607/model.neff
12343
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_b3823a1049a26ba7c274+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
12344
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_e595e217d65e663a19f1+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
12345
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_e595e217d65e663a19f1+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
12343
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_b3823a1049a26ba7c274+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
12344
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_e595e217d65e663a19f1+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
12345
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_e595e217d65e663a19f1+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
12346
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_8414d37eab328896acb7+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
12347
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_8414d37eab328896acb7+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
12348
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_c0808787f3555a6627cc+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.1.dev0/qwen3/Qwen/Qwen3-0.6B/af16941c15e5dc76a5aa.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": 2,
59
+ "max_batch_size": 4,
60
+ "max_context_length": 4096,
61
+ "max_topk": 256,
62
+ "n_active_tokens": 4096,
63
+ "neuronxcc_version": "2.21.18209.0+043b1bf7",
64
+ "on_device_sampling": false,
65
+ "optimum_neuron_version": "0.4.1.dev0",
66
+ "output_logits": false,
67
+ "pp_degree": 1,
68
+ "sequence_length": 4096,
69
+ "speculation_length": 0,
70
+ "start_rank_id": 0,
71
+ "target": "trn1",
72
+ "torch_dtype": "bfloat16",
73
+ "tp_degree": 2
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-2.21.18209.0+043b1bf7/MODULE_8414d37eab328896acb7+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-2.21.18209.0+043b1bf7/MODULE_8414d37eab328896acb7+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_8414d37eab328896acb7+a02c3a36/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_8414d37eab328896acb7+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bc43d12e254c5d9e9836dc62b2f42286ba3656c63872358979aa60cba90c1747
3
+ size 3503104
neuronxcc-2.21.18209.0+043b1bf7/MODULE_8414d37eab328896acb7+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:41ebc70c9a9557a0017beb75a9466659eb410462f8a73871d6de7ea35aeb3621
3
+ size 3640300
neuronxcc-2.21.18209.0+043b1bf7/MODULE_c0808787f3555a6627cc+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-2.21.18209.0+043b1bf7/MODULE_c0808787f3555a6627cc+24129607/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_c0808787f3555a6627cc+24129607/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3e46b218e15d50391d6a48b6055d6ae1ef98f6018b4982e4745fa5db87d302a3
3
+ size 865460
neuronxcc-2.21.18209.0+043b1bf7/MODULE_c0808787f3555a6627cc+24129607/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd50a72e0cc90ba8dcda3b370b92e0be78de67ce1a95587be24ab0a9ec0922f6
3
+ size 36148224