Hugging Face's logo

optimum-internal-testing
/
neuron-testing-cache

Model card Files
xet
 Community
optimum-internal-testing-user commited on
Commit
36913b0
·
verified ·
1 Parent(s): 407b7da

Synchronizing local compiler cache.

Browse files
Files changed (17) hide show
  1. .gitattributes +3 -0
  2. neuronxcc-2.22.12471.0+b4a00d10/0_REGISTRY/0.4.4.dev2/qwen2/Qwen/Qwen2.5-0.5B/c616b83f7e861571a5a4.json +82 -0
  3. neuronxcc-2.22.12471.0+b4a00d10/MODULE_1167ac875f15bc907a0e+84337dd9/model.neff +1 -1
  4. neuronxcc-2.22.12471.0+b4a00d10/MODULE_1167ac875f15bc907a0e+84337dd9/wrapped_neff.hlo +1 -1
  5. neuronxcc-2.22.12471.0+b4a00d10/MODULE_324c3bad86c071a7947d+84337dd9/compile_flags.json +1 -0
  6. neuronxcc-2.22.12471.0+b4a00d10/MODULE_324c3bad86c071a7947d+84337dd9/model.done +0 -0
  7. neuronxcc-2.22.12471.0+b4a00d10/MODULE_324c3bad86c071a7947d+84337dd9/model.hlo_module.pb +3 -0
  8. neuronxcc-2.22.12471.0+b4a00d10/MODULE_324c3bad86c071a7947d+84337dd9/model.neff +3 -0
  9. neuronxcc-2.22.12471.0+b4a00d10/MODULE_324c3bad86c071a7947d+84337dd9/wrapped_neff.hlo +3 -0
  10. neuronxcc-2.22.12471.0+b4a00d10/MODULE_5274831e83842758b831+519f203d/model.neff +1 -1
  11. neuronxcc-2.22.12471.0+b4a00d10/MODULE_52ac114efcae1dd03863+519f203d/model.neff +1 -1
  12. neuronxcc-2.22.12471.0+b4a00d10/MODULE_a968bbe6dc74ca5087a2+84337dd9/model.neff +1 -1
  13. neuronxcc-2.22.12471.0+b4a00d10/MODULE_a968bbe6dc74ca5087a2+84337dd9/wrapped_neff.hlo +1 -1
  14. neuronxcc-2.22.12471.0+b4a00d10/MODULE_dd85f09172e533f22911+519f203d/compile_flags.json +1 -0
  15. neuronxcc-2.22.12471.0+b4a00d10/MODULE_dd85f09172e533f22911+519f203d/model.done +0 -0
  16. neuronxcc-2.22.12471.0+b4a00d10/MODULE_dd85f09172e533f22911+519f203d/model.hlo_module.pb +3 -0
  17. neuronxcc-2.22.12471.0+b4a00d10/MODULE_dd85f09172e533f22911+519f203d/model.neff +3 -0
.gitattributes CHANGED
@@ -5907,3 +5907,6 @@ neuronxcc-2.22.12471.0+b4a00d10/MODULE_a968bbe6dc74ca5087a2+84337dd9/wrapped_nef
5907
  neuronxcc-2.22.12471.0+b4a00d10/MODULE_1167ac875f15bc907a0e+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
5908
  neuronxcc-2.22.12471.0+b4a00d10/MODULE_1167ac875f15bc907a0e+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
5909
  neuronxcc-2.22.12471.0+b4a00d10/MODULE_52ac114efcae1dd03863+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
 
 
 
 
5907
  neuronxcc-2.22.12471.0+b4a00d10/MODULE_1167ac875f15bc907a0e+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
5908
  neuronxcc-2.22.12471.0+b4a00d10/MODULE_1167ac875f15bc907a0e+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
5909
  neuronxcc-2.22.12471.0+b4a00d10/MODULE_52ac114efcae1dd03863+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
5910
+ neuronxcc-2.22.12471.0+b4a00d10/MODULE_324c3bad86c071a7947d+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
5911
+ neuronxcc-2.22.12471.0+b4a00d10/MODULE_324c3bad86c071a7947d+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
5912
+ neuronxcc-2.22.12471.0+b4a00d10/MODULE_dd85f09172e533f22911+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.22.12471.0+b4a00d10/0_REGISTRY/0.4.4.dev2/qwen2/Qwen/Qwen2.5-0.5B/c616b83f7e861571a5a4.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": "2.22.12471.0+b4a00d10",
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-2.22.12471.0+b4a00d10/MODULE_1167ac875f15bc907a0e+84337dd9/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b021dece4c0ebfe8bfd373232e5b96b0b92f22133890b7fc114eec6f45a15c9d
3
  size 2601984
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:707a75c503e5f4a3bc4de05d829a435db86683439fe141b64bb08ab0f04929b6
3
  size 2601984
neuronxcc-2.22.12471.0+b4a00d10/MODULE_1167ac875f15bc907a0e+84337dd9/wrapped_neff.hlo CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:28dc4f36cd0e1f2a503ac596174f0155af110d4add9eaf37512bbf97d351a88b
3
  size 2676071
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:15b3d014612428a6113178275b5cb03acf6ee7b7fcb2b139095373db5716e0be
3
  size 2676071
neuronxcc-2.22.12471.0+b4a00d10/MODULE_324c3bad86c071a7947d+84337dd9/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", "--verbose=35", "--logfile=/tmp/nxd_model/token_generation/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.22.12471.0+b4a00d10/MODULE_324c3bad86c071a7947d+84337dd9/model.done ADDED
File without changes
neuronxcc-2.22.12471.0+b4a00d10/MODULE_324c3bad86c071a7947d+84337dd9/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:909c7a7f5dc014f0707f0cb4e377504b2b814c709227e4a7d98af57d8f69afbb
3
+ size 2565149
neuronxcc-2.22.12471.0+b4a00d10/MODULE_324c3bad86c071a7947d+84337dd9/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd1b8f1a355819a105587059a9146dca4d24be3a8ea4f3acca30a5e0f6457130
3
+ size 2151424
neuronxcc-2.22.12471.0+b4a00d10/MODULE_324c3bad86c071a7947d+84337dd9/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fceeefcd56ec48b740237629874473a60023ca61f84b684b37fbd629be33aba9
3
+ size 2307618
neuronxcc-2.22.12471.0+b4a00d10/MODULE_5274831e83842758b831+519f203d/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:76537704781b11b3d496666cae05a28ea0940403996348dd002a2f36cc45ae54
3
  size 4219904
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5df396b699dc54f4e90f1bf7132b6f2bb5dbcc31630cb56c922ffa7badde4416
3
  size 4219904
neuronxcc-2.22.12471.0+b4a00d10/MODULE_52ac114efcae1dd03863+519f203d/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:42b1a5967aac9138f8e168614077cbd251a6a15cbc9b9cde419ba22415abb539
3
  size 10650624
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bddc86ae908c4fedff86b2f2bb0c3953935d14005bcab08c672eca3f7a76a376
3
  size 10650624
neuronxcc-2.22.12471.0+b4a00d10/MODULE_a968bbe6dc74ca5087a2+84337dd9/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:586b43d7d2f2030a1608cbb68bf9160687432aa2fa26d4a2f656f0ef1005c7b7
3
  size 3257344
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5c87e8fa425f3a6eee0b67b1dc4657a91a4598686e5b0ad9a1b1b8813f61dc24
3
  size 3257344
neuronxcc-2.22.12471.0+b4a00d10/MODULE_a968bbe6dc74ca5087a2+84337dd9/wrapped_neff.hlo CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:425acfb48f46b6e0b3bb71a5de707dfa9221fb2c90637062fc0d5d34f61fb835
3
  size 3331655
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e41012fda7d4998c101efde3a213f17492ad42975b34fb1dbbbac8b7038f6ffd
3
  size 3331655
neuronxcc-2.22.12471.0+b4a00d10/MODULE_dd85f09172e533f22911+519f203d/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", "--verbose=35", "--logfile=/tmp/nxd_model/context_encoding/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.22.12471.0+b4a00d10/MODULE_dd85f09172e533f22911+519f203d/model.done ADDED
File without changes
neuronxcc-2.22.12471.0+b4a00d10/MODULE_dd85f09172e533f22911+519f203d/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cc56fd1aaead9133afdfd10eec74ccd78720ef773bd42c5878996ac113065836
3
+ size 2364614
neuronxcc-2.22.12471.0+b4a00d10/MODULE_dd85f09172e533f22911+519f203d/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f65a1d691df768eb8d2aff5709ea9ea8a447d3192f6562721aa937cd2302b7f8
3
+ size 28017664