Hugging Face's logo

optimum-internal-testing
/
neuron-testing-cache

Model card Files
xet
 Community
optimum-internal-testing-user commited on
Commit
767c7c8
·
verified ·
1 Parent(s): d021c73

Synchronizing local compiler cache.

Browse files
Files changed (11) hide show
  1. .gitattributes +3 -0
  2. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev2/qwen2/Qwen/Qwen2.5-0.5B/f1a4bb21823df0476631.json +82 -0
  3. neuronxcc-2.21.33363.0+82129205/MODULE_5426f921e14368bcf904+24129607/compile_flags.json +1 -0
  4. neuronxcc-2.21.33363.0+82129205/MODULE_5426f921e14368bcf904+24129607/model.done +0 -0
  5. neuronxcc-2.21.33363.0+82129205/MODULE_5426f921e14368bcf904+24129607/model.hlo_module.pb +3 -0
  6. neuronxcc-2.21.33363.0+82129205/MODULE_5426f921e14368bcf904+24129607/model.neff +3 -0
  7. neuronxcc-2.21.33363.0+82129205/MODULE_5ec8c8cf0f5bf89c9d6b+a02c3a36/compile_flags.json +1 -0
  8. neuronxcc-2.21.33363.0+82129205/MODULE_5ec8c8cf0f5bf89c9d6b+a02c3a36/model.done +0 -0
  9. neuronxcc-2.21.33363.0+82129205/MODULE_5ec8c8cf0f5bf89c9d6b+a02c3a36/model.hlo_module.pb +3 -0
  10. neuronxcc-2.21.33363.0+82129205/MODULE_5ec8c8cf0f5bf89c9d6b+a02c3a36/model.neff +3 -0
  11. neuronxcc-2.21.33363.0+82129205/MODULE_5ec8c8cf0f5bf89c9d6b+a02c3a36/wrapped_neff.hlo +3 -0
.gitattributes CHANGED
@@ -6068,3 +6068,6 @@ neuronxcc-2.22.12471.0+b4a00d10/MODULE_0be862564335826e1b17+84337dd9/wrapped_nef
6068
  neuronxcc-2.22.12471.0+b4a00d10/MODULE_27cd864628c5de289499+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
6069
  neuronxcc-2.22.12471.0+b4a00d10/MODULE_65249bfcb708326b6f91+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
6070
  neuronxcc-2.22.12471.0+b4a00d10/MODULE_65249bfcb708326b6f91+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
6068
  neuronxcc-2.22.12471.0+b4a00d10/MODULE_27cd864628c5de289499+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
6069
  neuronxcc-2.22.12471.0+b4a00d10/MODULE_65249bfcb708326b6f91+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
6070
  neuronxcc-2.22.12471.0+b4a00d10/MODULE_65249bfcb708326b6f91+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
6071
+ neuronxcc-2.21.33363.0+82129205/MODULE_5426f921e14368bcf904+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
6072
+ neuronxcc-2.21.33363.0+82129205/MODULE_5ec8c8cf0f5bf89c9d6b+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
6073
+ neuronxcc-2.21.33363.0+82129205/MODULE_5ec8c8cf0f5bf89c9d6b+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev2/qwen2/Qwen/Qwen2.5-0.5B/f1a4bb21823df0476631.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": "2.21.33363.0+82129205",
59
+ "on_device_sampling": false,
60
+ "optimum_neuron_version": "0.4.4.dev2",
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-2.21.33363.0+82129205/MODULE_5426f921e14368bcf904+24129607/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/context_encoding/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.21.33363.0+82129205/MODULE_5426f921e14368bcf904+24129607/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_5426f921e14368bcf904+24129607/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2143e28e1d0b82695d02278df3d2442182f6e68a4fd9fd84588b6ee3a4ad4f47
3
+ size 739558
neuronxcc-2.21.33363.0+82129205/MODULE_5426f921e14368bcf904+24129607/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1b413cede89c12b746351746eca8d15edb0957e6056e2962a77eac01c47585cb
3
+ size 26133504
neuronxcc-2.21.33363.0+82129205/MODULE_5ec8c8cf0f5bf89c9d6b+a02c3a36/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/token_generation/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.21.33363.0+82129205/MODULE_5ec8c8cf0f5bf89c9d6b+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_5ec8c8cf0f5bf89c9d6b+a02c3a36/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:00199479d642f8e41b2d2d86073cc56342f8db6f1f5b0802ab804a154a158ed1
3
+ size 596952
neuronxcc-2.21.33363.0+82129205/MODULE_5ec8c8cf0f5bf89c9d6b+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7a2d51dbe91269fb6a6ccf842ec026c043b5cfbc28290a846f888fb033fb31be
3
+ size 1936384
neuronxcc-2.21.33363.0+82129205/MODULE_5ec8c8cf0f5bf89c9d6b+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e7fff4843d478f5275137ffc3309b2291909f7f5ef2aae82cd2765d10071ce3c
3
+ size 2092575