Hugging Face's logo

optimum-internal-testing
/
neuron-testing-cache

Model card Files
xet
Community
dacorvo HF Staff commited on
Commit
ccc9890
·
verified ·
1 Parent(s): 0e4ed53

Synchronizing local compiler cache.

Browse files
Files changed (5) hide show
  1. neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.2.dev1/qwen2/Qwen/Qwen2.5-0.5B/262017b899fd3b5b56c8.json +82 -0
  2. neuronxcc-2.21.18209.0+043b1bf7/MODULE_9d35248a93d4142a3cf1+24129607/model.hlo_module.pb +1 -1
  3. neuronxcc-2.21.18209.0+043b1bf7/MODULE_9d35248a93d4142a3cf1+24129607/model.neff +1 -1
  4. neuronxcc-2.21.18209.0+043b1bf7/MODULE_c2b3afa51c57d431a332+a02c3a36/model.neff +1 -1
  5. neuronxcc-2.21.18209.0+043b1bf7/MODULE_c2b3afa51c57d431a332+a02c3a36/wrapped_neff.hlo +1 -1
neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.2.dev1/qwen2/Qwen/Qwen2.5-0.5B/262017b899fd3b5b56c8.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": "0.4.2.dev1",
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_9d35248a93d4142a3cf1+24129607/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:770f59a15ab233bea70cb16360c196b23b356f67051422593116053095e3db38
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 sha256:58017c0e14cfb2aa2617054665041a111f19a8658e3856ce7f80963501abfa68
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_c2b3afa51c57d431a332+a02c3a36/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4fb43f4ea0d29f851cd0c0d3575ab21545fe6863d03f53111b381301e3eee665
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 sha256:2cfc00602dde4b251f7ee5e382c599874555cf1857694aab39c3a65c89e133d4
3
  size 2082478
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6af3d77381b3fde23da8966bdbf8f343010c5869ecb7fef585166f1178547f33
3
  size 2082478