Hugging Face's logo

optimum-internal-testing
/
neuron-testing-cache

Model card Files
xet
 Community
tengomucho HF Staff commited on
Commit
b854869
·
verified ·
1 Parent(s): 7525a88

Synchronizing local compiler cache.

Browse files
Files changed (5) hide show
  1. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev2/qwen3/Qwen/Qwen3-0.6B/c4069e9e5ecd272fe629.json +87 -0
  2. neuronxcc-2.21.33363.0+82129205/MODULE_4c7cbe55e26f80b4ca5e+a32116a7/model.hlo_module.pb +1 -1
  3. neuronxcc-2.21.33363.0+82129205/MODULE_4c7cbe55e26f80b4ca5e+a32116a7/model.neff +1 -1
  4. neuronxcc-2.21.33363.0+82129205/MODULE_68cb8e9f3c2305481f3a+ac10809c/model.neff +1 -1
  5. neuronxcc-2.21.33363.0+82129205/MODULE_68cb8e9f3c2305481f3a+ac10809c/wrapped_neff.hlo +1 -1
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev2/qwen3/Qwen/Qwen3-0.6B/c4069e9e5ecd272fe629.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": 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": 4,
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": true,
66
+ "optimum_neuron_version": "0.4.5.dev2",
67
+ "output_logits": false,
68
+ "pp_degree": 1,
69
+ "sequence_length": 4096,
70
+ "speculation_length": 0,
71
+ "start_rank_id": 0,
72
+ "target": "trn2",
73
+ "torch_dtype": "bfloat16",
74
+ "tp_degree": 4
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_4c7cbe55e26f80b4ca5e+a32116a7/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e64677a84206184e9f71e52a51b1446ab23eafc189002ddb06d00107ad695aef
3
  size 2114263
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:65ace86c54e1b508276b79c88d3bdd18e9555be4ae83ae564fdfe3bf9187ffe4
3
  size 2114263
neuronxcc-2.21.33363.0+82129205/MODULE_4c7cbe55e26f80b4ca5e+a32116a7/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ca1e0684d149f28a2434f50f0fd4eb30d492f42848c6c748bdd51db759a7fa4a
3
  size 15719424
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bb2a1b54911d1eb250b3338e9110052d2627f7ccafa519366787174f97bae94e
3
  size 15719424
neuronxcc-2.21.33363.0+82129205/MODULE_68cb8e9f3c2305481f3a+ac10809c/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c8ff73a482b4a9b7d196c0ce6c3eddaf511d05140b4419b62c464c5b76877f06
3
  size 2970624
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c95f820ddd3d798c1c2ed974e858011a9f764a892131618a3b92339406295def
3
  size 2970624
neuronxcc-2.21.33363.0+82129205/MODULE_68cb8e9f3c2305481f3a+ac10809c/wrapped_neff.hlo CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5fccf83bc43bfc007277d4e6744b65505504160c4d8e0c68a2c8ba297ea6ab46
3
  size 3107400
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7cb2be1eb196bff8c250a6a3cce6e418dd7728f5b499708164826ba78011a94e
3
  size 3107400