Hugging Face's logo

optimum-internal-testing
/
neuron-testing-cache

Model card Files
xet
Community
optimum-internal-testing-user commited on
Commit
4516a78
·
verified ·
1 Parent(s): 06a3da9

Synchronizing local compiler cache.

Browse files
Files changed (6) hide show
  1. neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.3.dev0/qwen2/Qwen/Qwen2.5-0.5B/a00506c78d9c3ead4d9d.json +82 -0
  2. neuronxcc-2.21.18209.0+043b1bf7/MODULE_b3cfe28275e1a02a015c+24129607/model.hlo_module.pb +1 -1
  3. neuronxcc-2.21.18209.0+043b1bf7/MODULE_b3cfe28275e1a02a015c+24129607/model.neff +1 -1
  4. neuronxcc-2.21.18209.0+043b1bf7/MODULE_d4df6648fd68fe444e68+a02c3a36/model.hlo_module.pb +1 -1
  5. neuronxcc-2.21.18209.0+043b1bf7/MODULE_d4df6648fd68fe444e68+a02c3a36/model.neff +1 -1
  6. neuronxcc-2.21.18209.0+043b1bf7/MODULE_d4df6648fd68fe444e68+a02c3a36/wrapped_neff.hlo +1 -1
neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.3.dev0/qwen2/Qwen/Qwen2.5-0.5B/a00506c78d9c3ead4d9d.json ADDED
@@ -0,0 +1,82 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen2.5-0.5B",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Qwen2ForCausalLM"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "dtype": "bfloat16",
10
+ "hidden_act": "silu",
11
+ "hidden_size": 896,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 4864,
14
+ "layer_types": [
15
+ "full_attention",
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
+ ],
40
+ "max_position_embeddings": 32768,
41
+ "max_window_layers": 24,
42
+ "model_type": "qwen2",
43
+ "neuron": {
44
+ "_serialized_key": "NxDNeuronConfig",
45
+ "batch_size": 4,
46
+ "capacity_factor": null,
47
+ "checkpoint_id": "Qwen/Qwen2.5-0.5B",
48
+ "checkpoint_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
49
+ "continuous_batching": true,
50
+ "ep_degree": 1,
51
+ "fused_qkv": false,
52
+ "glu_mlp": true,
53
+ "local_ranks_size": 2,
54
+ "max_batch_size": 4,
55
+ "max_context_length": 4096,
56
+ "max_topk": 256,
57
+ "n_active_tokens": 4096,
58
+ "neuronxcc_version": "2.21.18209.0+043b1bf7",
59
+ "on_device_sampling": false,
60
+ "optimum_neuron_version": "0.4.3.dev0",
61
+ "output_logits": false,
62
+ "pp_degree": 1,
63
+ "sequence_length": 4096,
64
+ "speculation_length": 0,
65
+ "start_rank_id": 0,
66
+ "target": "trn1",
67
+ "torch_dtype": "bfloat16",
68
+ "tp_degree": 2
69
+ },
70
+ "num_attention_heads": 14,
71
+ "num_hidden_layers": 24,
72
+ "num_key_value_heads": 2,
73
+ "rms_norm_eps": 1e-06,
74
+ "rope_scaling": null,
75
+ "rope_theta": 1000000.0,
76
+ "sliding_window": null,
77
+ "tie_word_embeddings": true,
78
+ "use_cache": true,
79
+ "use_mrope": false,
80
+ "use_sliding_window": false,
81
+ "vocab_size": 151936
82
+ }
neuronxcc-2.21.18209.0+043b1bf7/MODULE_b3cfe28275e1a02a015c+24129607/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:649ff83e0cbee13abda84e02c1a114e8b60d780456161e42f68c99c57744691c
3
  size 739558
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7d98bb4abcef46bc4dfcb539ece861af724c555279e78c4ba6ef27c16f25cecc
3
  size 739558
neuronxcc-2.21.18209.0+043b1bf7/MODULE_b3cfe28275e1a02a015c+24129607/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:75c45e80d4542e2aa466a08fb7deb473d23b4391a4ab8932e8822151bdb7d57f
3
  size 26133504
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a8f11dc4f21ab9be25f2f8e6fa30afe961d957c06d88794401230105ff3bfe06
3
  size 26133504
neuronxcc-2.21.18209.0+043b1bf7/MODULE_d4df6648fd68fe444e68+a02c3a36/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2c2920e5fde77a80d2d66c9ba5addf46e53d032f533c7097cba34e31243f84ed
3
  size 588406
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4b570926ce0fcfd782ef61d818285f77f7df61321cfef5c4f1085b51b81d2a1a
3
  size 588406
neuronxcc-2.21.18209.0+043b1bf7/MODULE_d4df6648fd68fe444e68+a02c3a36/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5ffb9d13327b496986f509229731d4a197e4eeee86cc389475a14f02ca71bd8c
3
  size 1926144
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0b7e1df7c167307eebb0cb890f0069cfca3b9d6856d75d02599d69d071525bc7
3
  size 1926144
neuronxcc-2.21.18209.0+043b1bf7/MODULE_d4df6648fd68fe444e68+a02c3a36/wrapped_neff.hlo CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e41e60345218d405d424a0dd964051bb5e0db8d7d224d1a904de33fa8f56679d
3
  size 2082478
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5891964fa1d2e4d1f4dcf53122d8a8fe7f5d25bfe1b908e9330abcdebb982270
3
  size 2082478