Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
 Community
695
dacorvo HF Staff commited on
Commit
b32f422
·
verified ·
1 Parent(s): 6f89505

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/Qwen2.5-7B-Instruct/1ccce1ab4d92a2d3df10.json +85 -0
  3. neuronxcc-2.21.18209.0+043b1bf7/MODULE_974e66f3382c9122a794+ed72d204/compile_flags.json +1 -0
  4. neuronxcc-2.21.18209.0+043b1bf7/MODULE_974e66f3382c9122a794+ed72d204/model.done +0 -0
  5. neuronxcc-2.21.18209.0+043b1bf7/MODULE_974e66f3382c9122a794+ed72d204/model.hlo_module.pb +3 -0
  6. neuronxcc-2.21.18209.0+043b1bf7/MODULE_974e66f3382c9122a794+ed72d204/model.neff +3 -0
  7. neuronxcc-2.21.18209.0+043b1bf7/MODULE_bf9fce2c422fe3f1956e+a9d440f5/compile_flags.json +1 -0
  8. neuronxcc-2.21.18209.0+043b1bf7/MODULE_bf9fce2c422fe3f1956e+a9d440f5/model.done +0 -0
  9. neuronxcc-2.21.18209.0+043b1bf7/MODULE_bf9fce2c422fe3f1956e+a9d440f5/model.hlo_module.pb +3 -0
  10. neuronxcc-2.21.18209.0+043b1bf7/MODULE_bf9fce2c422fe3f1956e+a9d440f5/model.neff +3 -0
  11. neuronxcc-2.21.18209.0+043b1bf7/MODULE_bf9fce2c422fe3f1956e+a9d440f5/wrapped_neff.hlo +3 -0
.gitattributes CHANGED
@@ -11524,3 +11524,6 @@ neuronxcc-2.21.18209.0+043b1bf7/MODULE_85fe6d3a7c65fbb1ab27+a9d440f5/wrapped_nef
11524
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_ac3fab9b5d0ae00ad513+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11525
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_ac3fab9b5d0ae00ad513+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
11526
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_ebee364bbfc2e280f573+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
 
 
 
 
11524
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_ac3fab9b5d0ae00ad513+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11525
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_ac3fab9b5d0ae00ad513+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
11526
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_ebee364bbfc2e280f573+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11527
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_974e66f3382c9122a794+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11528
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_bf9fce2c422fe3f1956e+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11529
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_bf9fce2c422fe3f1956e+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/Qwen2.5-7B-Instruct/1ccce1ab4d92a2d3df10.json ADDED
@@ -0,0 +1,85 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen2.5-7B-Instruct",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Qwen2ForCausalLM"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 3584,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 18944,
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
+ ],
43
+ "max_position_embeddings": 32768,
44
+ "max_window_layers": 28,
45
+ "model_type": "qwen2",
46
+ "neuron": {
47
+ "_serialized_key": "NxDNeuronConfig",
48
+ "batch_size": 32,
49
+ "capacity_factor": null,
50
+ "checkpoint_id": "Qwen/Qwen2.5-7B-Instruct",
51
+ "checkpoint_revision": "a09a35458c702b33eeacc393d103063234e8bc28",
52
+ "continuous_batching": true,
53
+ "enable_bucketing": false,
54
+ "ep_degree": 1,
55
+ "fused_qkv": false,
56
+ "glu_mlp": true,
57
+ "local_ranks_size": 8,
58
+ "max_batch_size": 32,
59
+ "max_context_length": 4096,
60
+ "max_topk": 256,
61
+ "n_active_tokens": 4096,
62
+ "neuronxcc_version": "2.21.18209.0+043b1bf7",
63
+ "on_device_sampling": true,
64
+ "optimum_neuron_version": "0.4.0.dev0",
65
+ "output_logits": false,
66
+ "pp_degree": 1,
67
+ "sequence_length": 4096,
68
+ "speculation_length": 0,
69
+ "start_rank_id": 0,
70
+ "target": "trn1",
71
+ "torch_dtype": "bfloat16",
72
+ "tp_degree": 8
73
+ },
74
+ "num_attention_heads": 28,
75
+ "num_hidden_layers": 28,
76
+ "num_key_value_heads": 4,
77
+ "rms_norm_eps": 1e-06,
78
+ "rope_scaling": null,
79
+ "rope_theta": 1000000.0,
80
+ "sliding_window": null,
81
+ "tie_word_embeddings": false,
82
+ "use_cache": true,
83
+ "use_sliding_window": false,
84
+ "vocab_size": 152064
85
+ }
neuronxcc-2.21.18209.0+043b1bf7/MODULE_974e66f3382c9122a794+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_974e66f3382c9122a794+ed72d204/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_974e66f3382c9122a794+ed72d204/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d8f714f77c179a7d6127116bef68c1898a7fced01942dfd161190e019cae7394
3
+ size 911685
neuronxcc-2.21.18209.0+043b1bf7/MODULE_974e66f3382c9122a794+ed72d204/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e123e86d18afabbd92ced88482b32bfc2f36c4515a127eaeddc178406e26106c
3
+ size 41114624
neuronxcc-2.21.18209.0+043b1bf7/MODULE_bf9fce2c422fe3f1956e+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_bf9fce2c422fe3f1956e+a9d440f5/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_bf9fce2c422fe3f1956e+a9d440f5/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:46eeb438d2e41f10f76fc07c0d4ce3c289bd77b727745204d5279e4cea3355a2
3
+ size 735407
neuronxcc-2.21.18209.0+043b1bf7/MODULE_bf9fce2c422fe3f1956e+a9d440f5/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0821442e246093abac327054c443150035846cd0cebcbeb8d7506698439e1c44
3
+ size 6728704
neuronxcc-2.21.18209.0+043b1bf7/MODULE_bf9fce2c422fe3f1956e+a9d440f5/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:edad49ba704dd43baa31e5f3038a7d711254e14e6e85f4f92d89a4c6cbb34177
3
+ size 6912039