Hugging Face's logo

optimum-internal-testing
/
neuron-testing-cache

Model card Files
xet
Community
optimum-internal-testing-user commited on
Commit
23e3613
·
verified ·
1 Parent(s): 1e77066

Synchronizing local compiler cache.

Browse files
Files changed (11) hide show
  1. .gitattributes +3 -0
  2. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.3.dev2/qwen2/Qwen/Qwen2.5-0.5B/c2fd06a569eed2bbbeb4.json +82 -0
  3. neuronxcc-2.21.33363.0+82129205/MODULE_b3cfe28275e1a02a015c+24129607/compile_flags.json +1 -0
  4. neuronxcc-2.21.33363.0+82129205/MODULE_b3cfe28275e1a02a015c+24129607/model.done +0 -0
  5. neuronxcc-2.21.33363.0+82129205/MODULE_b3cfe28275e1a02a015c+24129607/model.hlo_module.pb +3 -0
  6. neuronxcc-2.21.33363.0+82129205/MODULE_b3cfe28275e1a02a015c+24129607/model.neff +3 -0
  7. neuronxcc-2.21.33363.0+82129205/MODULE_d4df6648fd68fe444e68+a02c3a36/compile_flags.json +1 -0
  8. neuronxcc-2.21.33363.0+82129205/MODULE_d4df6648fd68fe444e68+a02c3a36/model.done +0 -0
  9. neuronxcc-2.21.33363.0+82129205/MODULE_d4df6648fd68fe444e68+a02c3a36/model.hlo_module.pb +3 -0
  10. neuronxcc-2.21.33363.0+82129205/MODULE_d4df6648fd68fe444e68+a02c3a36/model.neff +3 -0
  11. neuronxcc-2.21.33363.0+82129205/MODULE_d4df6648fd68fe444e68+a02c3a36/wrapped_neff.hlo +3 -0
.gitattributes CHANGED
@@ -5566,3 +5566,6 @@ neuronxcc-2.21.18209.0+043b1bf7/MODULE_f9c594cf85ceddc7f3ad+24129607/model.neff
5566
  neuronxcc-2.21.33363.0+82129205/MODULE_3d83db23b8fc661cf397+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
5567
  neuronxcc-2.21.33363.0+82129205/MODULE_86841313265618bb997a+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
5568
  neuronxcc-2.21.33363.0+82129205/MODULE_86841313265618bb997a+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
5566
  neuronxcc-2.21.33363.0+82129205/MODULE_3d83db23b8fc661cf397+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
5567
  neuronxcc-2.21.33363.0+82129205/MODULE_86841313265618bb997a+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
5568
  neuronxcc-2.21.33363.0+82129205/MODULE_86841313265618bb997a+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
5569
+ neuronxcc-2.21.33363.0+82129205/MODULE_b3cfe28275e1a02a015c+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
5570
+ neuronxcc-2.21.33363.0+82129205/MODULE_d4df6648fd68fe444e68+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
5571
+ neuronxcc-2.21.33363.0+82129205/MODULE_d4df6648fd68fe444e68+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.3.dev2/qwen2/Qwen/Qwen2.5-0.5B/c2fd06a569eed2bbbeb4.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": "0.4.3.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_b3cfe28275e1a02a015c+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_b3cfe28275e1a02a015c+24129607/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_b3cfe28275e1a02a015c+24129607/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7d98bb4abcef46bc4dfcb539ece861af724c555279e78c4ba6ef27c16f25cecc
3
+ size 739558
neuronxcc-2.21.33363.0+82129205/MODULE_b3cfe28275e1a02a015c+24129607/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c908c04ad9e0825f85a3c53f966861b5e2d2a75e81c9c78c8d2db47354bcaed2
3
+ size 26133504
neuronxcc-2.21.33363.0+82129205/MODULE_d4df6648fd68fe444e68+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_d4df6648fd68fe444e68+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_d4df6648fd68fe444e68+a02c3a36/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4b570926ce0fcfd782ef61d818285f77f7df61321cfef5c4f1085b51b81d2a1a
3
+ size 588406
neuronxcc-2.21.33363.0+82129205/MODULE_d4df6648fd68fe444e68+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7b3a714bb0263c8460dd7078d9019fac3f5aeae822722e3f62825b29f4efdf42
3
+ size 1926144
neuronxcc-2.21.33363.0+82129205/MODULE_d4df6648fd68fe444e68+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:828b8c98fb041ed9aec3bac4ed5c08160c014335d4511bd8906d611c58297ca4
3
+ size 2082478