Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
Community
601
dacorvo HF Staff commited on
Commit
17c3142
·
verified ·
1 Parent(s): abc978d

Synchronizing local compiler cache.

Browse files
Files changed (11) hide show
  1. .gitattributes +3 -0
  2. neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.0.dev0/qwen2/Qwen/Qwen2.5-14B/44f8062bf00e038347ee.json +105 -0
  3. neuronxcc-2.21.18209.0+043b1bf7/MODULE_715e7fb7f68904bdb82d+ed72d204/compile_flags.json +1 -0
  4. neuronxcc-2.21.18209.0+043b1bf7/MODULE_715e7fb7f68904bdb82d+ed72d204/model.done +0 -0
  5. neuronxcc-2.21.18209.0+043b1bf7/MODULE_715e7fb7f68904bdb82d+ed72d204/model.hlo_module.pb +3 -0
  6. neuronxcc-2.21.18209.0+043b1bf7/MODULE_715e7fb7f68904bdb82d+ed72d204/model.neff +3 -0
  7. neuronxcc-2.21.18209.0+043b1bf7/MODULE_f2cde388cf9f4159501b+a9d440f5/compile_flags.json +1 -0
  8. neuronxcc-2.21.18209.0+043b1bf7/MODULE_f2cde388cf9f4159501b+a9d440f5/model.done +0 -0
  9. neuronxcc-2.21.18209.0+043b1bf7/MODULE_f2cde388cf9f4159501b+a9d440f5/model.hlo_module.pb +3 -0
  10. neuronxcc-2.21.18209.0+043b1bf7/MODULE_f2cde388cf9f4159501b+a9d440f5/model.neff +3 -0
  11. neuronxcc-2.21.18209.0+043b1bf7/MODULE_f2cde388cf9f4159501b+a9d440f5/wrapped_neff.hlo +3 -0
.gitattributes CHANGED
@@ -11680,3 +11680,6 @@ neuronxcc-2.21.18209.0+043b1bf7/MODULE_3ae065f5a935cccdd04d+a9d440f5/wrapped_nef
11680
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_bfe91ee4b01adcd697b9+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11681
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_fe833f6355feb2441f1b+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11682
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_fe833f6355feb2441f1b+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
11680
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_bfe91ee4b01adcd697b9+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11681
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_fe833f6355feb2441f1b+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11682
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_fe833f6355feb2441f1b+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
11683
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_715e7fb7f68904bdb82d+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11684
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_f2cde388cf9f4159501b+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11685
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_f2cde388cf9f4159501b+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.0.dev0/qwen2/Qwen/Qwen2.5-14B/44f8062bf00e038347ee.json ADDED
@@ -0,0 +1,105 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen2.5-14B",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Qwen2ForCausalLM"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 5120,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 13824,
13
+ "layer_types": [
14
+ "full_attention",
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
+ "full_attention",
40
+ "full_attention",
41
+ "full_attention",
42
+ "full_attention",
43
+ "full_attention",
44
+ "full_attention",
45
+ "full_attention",
46
+ "full_attention",
47
+ "full_attention",
48
+ "full_attention",
49
+ "full_attention",
50
+ "full_attention",
51
+ "full_attention",
52
+ "full_attention",
53
+ "full_attention",
54
+ "full_attention",
55
+ "full_attention",
56
+ "full_attention",
57
+ "full_attention",
58
+ "full_attention",
59
+ "full_attention",
60
+ "full_attention",
61
+ "full_attention"
62
+ ],
63
+ "max_position_embeddings": 131072,
64
+ "max_window_layers": 48,
65
+ "model_type": "qwen2",
66
+ "neuron": {
67
+ "_serialized_key": "NxDNeuronConfig",
68
+ "batch_size": 1,
69
+ "capacity_factor": null,
70
+ "checkpoint_id": "Qwen/Qwen2.5-14B",
71
+ "checkpoint_revision": "97e1e76335b7017d8f67c08a19d103c0504298c9",
72
+ "continuous_batching": false,
73
+ "enable_bucketing": false,
74
+ "ep_degree": 1,
75
+ "fused_qkv": false,
76
+ "glu_mlp": true,
77
+ "local_ranks_size": 8,
78
+ "max_batch_size": 1,
79
+ "max_context_length": 4096,
80
+ "max_topk": 256,
81
+ "n_active_tokens": 4096,
82
+ "neuronxcc_version": "2.21.18209.0+043b1bf7",
83
+ "on_device_sampling": true,
84
+ "optimum_neuron_version": "0.4.0.dev0",
85
+ "output_logits": false,
86
+ "pp_degree": 1,
87
+ "sequence_length": 4096,
88
+ "speculation_length": 0,
89
+ "start_rank_id": 0,
90
+ "target": "trn1",
91
+ "torch_dtype": "bfloat16",
92
+ "tp_degree": 8
93
+ },
94
+ "num_attention_heads": 40,
95
+ "num_hidden_layers": 48,
96
+ "num_key_value_heads": 8,
97
+ "rms_norm_eps": 1e-05,
98
+ "rope_scaling": null,
99
+ "rope_theta": 1000000.0,
100
+ "sliding_window": null,
101
+ "tie_word_embeddings": false,
102
+ "use_cache": true,
103
+ "use_sliding_window": false,
104
+ "vocab_size": 152064
105
+ }
neuronxcc-2.21.18209.0+043b1bf7/MODULE_715e7fb7f68904bdb82d+ed72d204/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_model/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.21.18209.0+043b1bf7/MODULE_715e7fb7f68904bdb82d+ed72d204/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_715e7fb7f68904bdb82d+ed72d204/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a4ed3de2d9b8a58ae98f61bc5b7a7aba9befb95f00631ae58e5422ce21f5ce56
3
+ size 1426478
neuronxcc-2.21.18209.0+043b1bf7/MODULE_715e7fb7f68904bdb82d+ed72d204/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:66ccd7450ac2059151763c893dc2b6dbd81a7959473146831e00fa3509172279
3
+ size 3124224
neuronxcc-2.21.18209.0+043b1bf7/MODULE_f2cde388cf9f4159501b+a9d440f5/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_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.21.18209.0+043b1bf7/MODULE_f2cde388cf9f4159501b+a9d440f5/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_f2cde388cf9f4159501b+a9d440f5/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cae54528756108bfa6aae8abe44e09a61866b075ed0c703e3848aa0907642de7
3
+ size 1233606
neuronxcc-2.21.18209.0+043b1bf7/MODULE_f2cde388cf9f4159501b+a9d440f5/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1b28b98147fc86c02dd7694cd97976d7fd5f98cd218aa51220db00d8c794b82a
3
+ size 5264384
neuronxcc-2.21.18209.0+043b1bf7/MODULE_f2cde388cf9f4159501b+a9d440f5/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6e6cc162c25002240b37be99976f828c96144cb1ce0d268441ff31185b197b4d
3
+ size 5580525