Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
 Community
695
dacorvo HF Staff commited on
Commit
4558ae2
·
verified ·
1 Parent(s): 6d4dc89

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/1dcd1ff99fdf3a8faa1b.json +85 -0
  3. neuronxcc-2.21.18209.0+043b1bf7/MODULE_1cf816134f4b0d45bef0+a9d440f5/compile_flags.json +1 -0
  4. neuronxcc-2.21.18209.0+043b1bf7/MODULE_1cf816134f4b0d45bef0+a9d440f5/model.done +0 -0
  5. neuronxcc-2.21.18209.0+043b1bf7/MODULE_1cf816134f4b0d45bef0+a9d440f5/model.hlo_module.pb +3 -0
  6. neuronxcc-2.21.18209.0+043b1bf7/MODULE_1cf816134f4b0d45bef0+a9d440f5/model.neff +3 -0
  7. neuronxcc-2.21.18209.0+043b1bf7/MODULE_1cf816134f4b0d45bef0+a9d440f5/wrapped_neff.hlo +3 -0
  8. neuronxcc-2.21.18209.0+043b1bf7/MODULE_e6cc1a3818b0ebc99756+ed72d204/compile_flags.json +1 -0
  9. neuronxcc-2.21.18209.0+043b1bf7/MODULE_e6cc1a3818b0ebc99756+ed72d204/model.done +0 -0
  10. neuronxcc-2.21.18209.0+043b1bf7/MODULE_e6cc1a3818b0ebc99756+ed72d204/model.hlo_module.pb +3 -0
  11. neuronxcc-2.21.18209.0+043b1bf7/MODULE_e6cc1a3818b0ebc99756+ed72d204/model.neff +3 -0
.gitattributes CHANGED
@@ -11482,3 +11482,6 @@ neuronxcc-2.21.18209.0+043b1bf7/MODULE_ae73fd7ace385f9e4051+ed72d204/model.neff
11482
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_5b7172cc10110a470457+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11483
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_9361fb1ac241d0d047f1+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11484
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_9361fb1ac241d0d047f1+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
11482
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_5b7172cc10110a470457+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11483
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_9361fb1ac241d0d047f1+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11484
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_9361fb1ac241d0d047f1+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
11485
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_1cf816134f4b0d45bef0+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11486
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_1cf816134f4b0d45bef0+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
11487
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_e6cc1a3818b0ebc99756+ed72d204/model.neff 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/1dcd1ff99fdf3a8faa1b.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": 4,
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": 2,
58
+ "max_batch_size": 4,
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": false,
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": 2
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_1cf816134f4b0d45bef0+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_1cf816134f4b0d45bef0+a9d440f5/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_1cf816134f4b0d45bef0+a9d440f5/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fa57d9fbdbba39ee5bc4ce9e5564ad09819fd9aeaf8e1a6cf11f4f97d098055b
3
+ size 703176
neuronxcc-2.21.18209.0+043b1bf7/MODULE_1cf816134f4b0d45bef0+a9d440f5/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0649a70f62f4445507158bd452b8b41a5ea5a7d017e48356ec95665517b54833
3
+ size 9923584
neuronxcc-2.21.18209.0+043b1bf7/MODULE_1cf816134f4b0d45bef0+a9d440f5/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cd8b32ad28531cb11f722e26fd0ee0e3a36a0754470ccc3a630bd1c3962be8bc
3
+ size 10132378
neuronxcc-2.21.18209.0+043b1bf7/MODULE_e6cc1a3818b0ebc99756+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_e6cc1a3818b0ebc99756+ed72d204/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_e6cc1a3818b0ebc99756+ed72d204/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1f051c3c6e00e0fb51e8044936edb4fdfea2925e88f756cae547d0969c43c4bb
3
+ size 889560
neuronxcc-2.21.18209.0+043b1bf7/MODULE_e6cc1a3818b0ebc99756+ed72d204/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:22f8d65665f4ee104ea18a4c583ceb744ed2750df5cc4a6f62f81c5099328c9e
3
+ size 6503424