Hugging Face's logo

optimum-internal-testing
/
neuron-testing-cache

Model card Files
xet
Community
dacorvo HF Staff commited on
Commit
57c7d47
·
verified ·
1 Parent(s): 39302e4

Synchronizing local compiler cache.

Browse files
Files changed (11) hide show
  1. .gitattributes +1 -0
  2. neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.5.dev1/60feecaa0c4c075e2f3e46a3f55d9a273f0ddd75a0ecf64e4ae27352e0819506/ee71309c308c0bf312e7.json +87 -0
  3. neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.5.dev1/qwen3/Qwen/Qwen3-Embedding-0.6B/ee71309c308c0bf312e7.json +87 -0
  4. neuronxcc-2.21.18209.0+043b1bf7/MODULE_9d35248a93d4142a3cf1+24129607/model.hlo_module.pb +1 -1
  5. neuronxcc-2.21.18209.0+043b1bf7/MODULE_9d35248a93d4142a3cf1+24129607/model.neff +1 -1
  6. neuronxcc-2.21.18209.0+043b1bf7/MODULE_a9ea3bcc615b10517bb2+fb4cc044/compile_flags.json +1 -0
  7. neuronxcc-2.21.18209.0+043b1bf7/MODULE_a9ea3bcc615b10517bb2+fb4cc044/model.done +0 -0
  8. neuronxcc-2.21.18209.0+043b1bf7/MODULE_a9ea3bcc615b10517bb2+fb4cc044/model.hlo_module.pb +3 -0
  9. neuronxcc-2.21.18209.0+043b1bf7/MODULE_a9ea3bcc615b10517bb2+fb4cc044/model.neff +3 -0
  10. neuronxcc-2.21.18209.0+043b1bf7/MODULE_c2b3afa51c57d431a332+a02c3a36/model.neff +1 -1
  11. neuronxcc-2.21.18209.0+043b1bf7/MODULE_c2b3afa51c57d431a332+a02c3a36/wrapped_neff.hlo +1 -1
.gitattributes CHANGED
@@ -6295,3 +6295,4 @@ neuronxcc-2.21.33363.0+82129205/MODULE_065f10c75a3a003852db+fb4cc044/model.neff
6295
  neuronxcc-2.21.33363.0+82129205/MODULE_1aace3454acab4662925+fb4cc044/model.neff filter=lfs diff=lfs merge=lfs -text
6296
  neuronxcc-2.21.33363.0+82129205/MODULE_a17ecef0a02d21e4d5a8+fb4cc044/model.neff filter=lfs diff=lfs merge=lfs -text
6297
  neuronxcc-2.21.33363.0+82129205/MODULE_cb86ed7fc724b06726b3+fb4cc044/model.neff filter=lfs diff=lfs merge=lfs -text
 
 
6295
  neuronxcc-2.21.33363.0+82129205/MODULE_1aace3454acab4662925+fb4cc044/model.neff filter=lfs diff=lfs merge=lfs -text
6296
  neuronxcc-2.21.33363.0+82129205/MODULE_a17ecef0a02d21e4d5a8+fb4cc044/model.neff filter=lfs diff=lfs merge=lfs -text
6297
  neuronxcc-2.21.33363.0+82129205/MODULE_cb86ed7fc724b06726b3+fb4cc044/model.neff filter=lfs diff=lfs merge=lfs -text
6298
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_a9ea3bcc615b10517bb2+fb4cc044/model.neff filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.5.dev1/60feecaa0c4c075e2f3e46a3f55d9a273f0ddd75a0ecf64e4ae27352e0819506/ee71309c308c0bf312e7.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen3-Embedding-0.6B",
4
+ "_task": "feature-extraction",
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": 32768,
47
+ "max_window_layers": 28,
48
+ "model_type": "qwen3",
49
+ "neuron": {
50
+ "_serialized_key": "NxDNeuronConfig",
51
+ "batch_size": 6,
52
+ "capacity_factor": null,
53
+ "checkpoint_id": "Qwen/Qwen3-Embedding-0.6B",
54
+ "checkpoint_revision": "c54f2e6e80b2d7b7de06f51cec4959f6b3e03418",
55
+ "continuous_batching": false,
56
+ "ep_degree": 1,
57
+ "fused_qkv": true,
58
+ "glu_mlp": true,
59
+ "local_ranks_size": 2,
60
+ "max_batch_size": 6,
61
+ "max_context_length": 8192,
62
+ "max_topk": 256,
63
+ "n_active_tokens": 8192,
64
+ "neuronxcc_version": "2.21.18209.0+043b1bf7",
65
+ "on_device_sampling": false,
66
+ "optimum_neuron_version": "0.4.5.dev1",
67
+ "output_logits": false,
68
+ "pp_degree": 1,
69
+ "sequence_length": 8192,
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": 151669
87
+ }
neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.5.dev1/qwen3/Qwen/Qwen3-Embedding-0.6B/ee71309c308c0bf312e7.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen3-Embedding-0.6B",
4
+ "_task": "feature-extraction",
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": 32768,
47
+ "max_window_layers": 28,
48
+ "model_type": "qwen3",
49
+ "neuron": {
50
+ "_serialized_key": "NxDNeuronConfig",
51
+ "batch_size": 6,
52
+ "capacity_factor": null,
53
+ "checkpoint_id": "Qwen/Qwen3-Embedding-0.6B",
54
+ "checkpoint_revision": "c54f2e6e80b2d7b7de06f51cec4959f6b3e03418",
55
+ "continuous_batching": false,
56
+ "ep_degree": 1,
57
+ "fused_qkv": true,
58
+ "glu_mlp": true,
59
+ "local_ranks_size": 2,
60
+ "max_batch_size": 6,
61
+ "max_context_length": 8192,
62
+ "max_topk": 256,
63
+ "n_active_tokens": 8192,
64
+ "neuronxcc_version": "2.21.18209.0+043b1bf7",
65
+ "on_device_sampling": false,
66
+ "optimum_neuron_version": "0.4.5.dev1",
67
+ "output_logits": false,
68
+ "pp_degree": 1,
69
+ "sequence_length": 8192,
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": 151669
87
+ }
neuronxcc-2.21.18209.0+043b1bf7/MODULE_9d35248a93d4142a3cf1+24129607/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9177517d99e0d55cefa841fe3c498143f9af0ee266e41e8165daa096bcf53048
3
  size 739558
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8c0311f0ae811c00f59cb813f4ee06008d0d18e5bdf0534ecf04e14c6eadcad8
3
  size 739558
neuronxcc-2.21.18209.0+043b1bf7/MODULE_9d35248a93d4142a3cf1+24129607/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:30586b03aa5f62a9b1b59eaf6ea52626b879dfaf40d5cea7a3bfd3c381e9bdac
3
  size 26133504
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e80290bb0de9fba2d791707189cfe81ec5896082ad2fc95838ce7c687a2f6ed4
3
  size 26133504
neuronxcc-2.21.18209.0+043b1bf7/MODULE_a9ea3bcc615b10517bb2+fb4cc044/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/encoding/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.21.18209.0+043b1bf7/MODULE_a9ea3bcc615b10517bb2+fb4cc044/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_a9ea3bcc615b10517bb2+fb4cc044/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ccd8dd07c208eb87030d40b6fe53666dbd98b0ad6fa3cf5fa46bd1e5d023563f
3
+ size 662166
neuronxcc-2.21.18209.0+043b1bf7/MODULE_a9ea3bcc615b10517bb2+fb4cc044/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3cd272e40e65fb42ee0c8247f443e277f005acb0ff4230bdc0d2f6c5dfdbe901
3
+ size 27689984
neuronxcc-2.21.18209.0+043b1bf7/MODULE_c2b3afa51c57d431a332+a02c3a36/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f3670e861c13c471660d8ade19eb8885d819345b12e54e2db228badaed202c5e
3
  size 1926144
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c2835fba37931b9c58c33d25543bd60d85d61d996e61ac3b6e78e5b4c95bfb85
3
  size 1926144
neuronxcc-2.21.18209.0+043b1bf7/MODULE_c2b3afa51c57d431a332+a02c3a36/wrapped_neff.hlo CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4ed49de1d6d5dc05c91b2f314dca6872f1d5b85ea476e2ee986e3764f8fb1424
3
  size 2082478
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6af3d77381b3fde23da8966bdbf8f343010c5869ecb7fef585166f1178547f33
3
  size 2082478