Hugging Face's logo

optimum-internal-testing
/
neuron-testing-cache

Model card Files
xet
 Community
optimum-internal-testing-user commited on
Commit
63c2686
·
verified ·
1 Parent(s): c3918b8

Synchronizing local compiler cache.

Browse files
Files changed (6) hide show
  1. neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/qwen3/Qwen/Qwen3-0.6B/3611243e26d473ff0bf7.json +90 -0
  2. neuronxcc-2.26.6360.0+6f180f47/MODULE_b8f3fcec7654dae3b7e7+84337dd9/model.hlo_module.pb +1 -1
  3. neuronxcc-2.26.6360.0+6f180f47/MODULE_b8f3fcec7654dae3b7e7+84337dd9/model.neff +1 -1
  4. neuronxcc-2.26.6360.0+6f180f47/MODULE_b8f3fcec7654dae3b7e7+84337dd9/wrapped_neff.hlo +1 -1
  5. neuronxcc-2.26.6360.0+6f180f47/MODULE_ffe3f8a441d27920d88f+519f203d/model.hlo_module.pb +1 -1
  6. neuronxcc-2.26.6360.0+6f180f47/MODULE_ffe3f8a441d27920d88f+519f203d/model.neff +1 -1
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/qwen3/Qwen/Qwen3-0.6B/3611243e26d473ff0bf7.json ADDED
@@ -0,0 +1,90 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": 1024,
62
+ "max_topk": 256,
63
+ "n_active_tokens": 1024,
64
+ "neuronxcc_version": "2.26.6360.0+6f180f47",
65
+ "on_device_sampling": false,
66
+ "optimum_neuron_version": "0.4.7.dev0",
67
+ "output_logits": false,
68
+ "pp_degree": 1,
69
+ "prefill_chunk_size": 0,
70
+ "sequence_length": 1024,
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_parameters": {
82
+ "rope_theta": 1000000,
83
+ "rope_type": "default"
84
+ },
85
+ "sliding_window": null,
86
+ "tie_word_embeddings": true,
87
+ "use_cache": true,
88
+ "use_sliding_window": false,
89
+ "vocab_size": 151936
90
+ }
neuronxcc-2.26.6360.0+6f180f47/MODULE_b8f3fcec7654dae3b7e7+84337dd9/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b5f27c51449d099e0b040ef6037190931a30e86698d5c376db4d83bbe8cb9a29
3
  size 690974
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a9b15f6ab0274a007738faabd275eca15001bd2d88e3c2be321216405847c618
3
  size 690974
neuronxcc-2.26.6360.0+6f180f47/MODULE_b8f3fcec7654dae3b7e7+84337dd9/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6a0bd5f84abc8a911ed9ed924532a7de44fa47916b4f4a92d5ca8d356e3b6bf8
3
  size 2489344
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e055b439e14d8993756a43a705ba82cbb777667fcf39cb13f193d402df6f7787
3
  size 2489344
neuronxcc-2.26.6360.0+6f180f47/MODULE_b8f3fcec7654dae3b7e7+84337dd9/wrapped_neff.hlo CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0c66897b6458760f0294c3550f5a10b8480170fe3c845273e24c66651d5f633e
3
  size 2649489
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:58c25edf86cf101116541f9c9e9a6c97a629c7cce950c069967bb9219a9cf229
3
  size 2649489
neuronxcc-2.26.6360.0+6f180f47/MODULE_ffe3f8a441d27920d88f+519f203d/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e1b885135c3cec0e58716c65fe89d0a57d385dc6f399af8701dd8f8e1d11930b
3
  size 681009
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9f6be3e5b042866b1d905f2dc6f4946a9e2e0ab8a2d94214438b0c62421db5c6
3
  size 681009
neuronxcc-2.26.6360.0+6f180f47/MODULE_ffe3f8a441d27920d88f+519f203d/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:fd7b97e07d42a04d46ca0ef3b35e1e96f4a57fdc55fd606b212729254b01776e
3
  size 2438144
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:626f174d0283277b48eb2a585e0e0f82b5e46d90cbcd63bc2e383c6efb928cc0
3
  size 2438144