Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
Community
629
dacorvo HF Staff commited on
Commit
dc3486b
·
verified ·
1 Parent(s): d127233

Synchronizing local compiler cache.

Browse files
Files changed (11) hide show
  1. .gitattributes +3 -0
  2. neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.3.1.dev7/qwen2/Qwen/Qwen2.5-0.5B/010085191d9c76431e4a.json +83 -0
  3. neuronxcc-2.21.18209.0+043b1bf7/MODULE_500ae99927947dab9c67+ed72d204/compile_flags.json +1 -0
  4. neuronxcc-2.21.18209.0+043b1bf7/MODULE_500ae99927947dab9c67+ed72d204/model.done +0 -0
  5. neuronxcc-2.21.18209.0+043b1bf7/MODULE_500ae99927947dab9c67+ed72d204/model.hlo_module.pb +3 -0
  6. neuronxcc-2.21.18209.0+043b1bf7/MODULE_500ae99927947dab9c67+ed72d204/model.neff +3 -0
  7. neuronxcc-2.21.18209.0+043b1bf7/MODULE_852c8a18de93f098d898+a9d440f5/compile_flags.json +1 -0
  8. neuronxcc-2.21.18209.0+043b1bf7/MODULE_852c8a18de93f098d898+a9d440f5/model.done +0 -0
  9. neuronxcc-2.21.18209.0+043b1bf7/MODULE_852c8a18de93f098d898+a9d440f5/model.hlo_module.pb +3 -0
  10. neuronxcc-2.21.18209.0+043b1bf7/MODULE_852c8a18de93f098d898+a9d440f5/model.neff +3 -0
  11. neuronxcc-2.21.18209.0+043b1bf7/MODULE_852c8a18de93f098d898+a9d440f5/wrapped_neff.hlo +3 -0
.gitattributes CHANGED
@@ -11386,3 +11386,6 @@ neuronxcc-2.21.18209.0+043b1bf7/MODULE_6504d4f3837aa33fa288+a9d440f5/wrapped_nef
11386
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_01ea5f705e6416f82f71+cd3419b6/model.neff filter=lfs diff=lfs merge=lfs -text
11387
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_01ea5f705e6416f82f71+cd3419b6/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
11388
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_e1edd716cab925764994+253d6470/model.neff filter=lfs diff=lfs merge=lfs -text
 
 
 
 
11386
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_01ea5f705e6416f82f71+cd3419b6/model.neff filter=lfs diff=lfs merge=lfs -text
11387
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_01ea5f705e6416f82f71+cd3419b6/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
11388
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_e1edd716cab925764994+253d6470/model.neff filter=lfs diff=lfs merge=lfs -text
11389
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_500ae99927947dab9c67+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11390
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_852c8a18de93f098d898+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11391
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_852c8a18de93f098d898+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.3.1.dev7/qwen2/Qwen/Qwen2.5-0.5B/010085191d9c76431e4a.json ADDED
@@ -0,0 +1,83 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "hidden_act": "silu",
10
+ "hidden_size": 896,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 4864,
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
+ ],
39
+ "max_position_embeddings": 32768,
40
+ "max_window_layers": 24,
41
+ "model_type": "qwen2",
42
+ "neuron": {
43
+ "_serialized_key": "NxDNeuronConfig",
44
+ "batch_size": 4,
45
+ "capacity_factor": null,
46
+ "checkpoint_id": "Qwen/Qwen2.5-0.5B",
47
+ "checkpoint_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
48
+ "continuous_batching": true,
49
+ "enable_bucketing": false,
50
+ "ep_degree": 1,
51
+ "fused_qkv": false,
52
+ "glu_mlp": true,
53
+ "local_ranks_size": 2,
54
+ "logical_nc_config": 1,
55
+ "max_batch_size": 4,
56
+ "max_context_length": 4096,
57
+ "max_topk": 256,
58
+ "n_active_tokens": 4096,
59
+ "neuronxcc_version": "2.21.18209.0+043b1bf7",
60
+ "on_device_sampling": false,
61
+ "optimum_neuron_version": "0.3.1.dev7",
62
+ "output_logits": false,
63
+ "pp_degree": 1,
64
+ "sequence_length": 4096,
65
+ "speculation_length": 0,
66
+ "start_rank_id": 0,
67
+ "target": null,
68
+ "torch_dtype": "bfloat16",
69
+ "tp_degree": 2
70
+ },
71
+ "num_attention_heads": 14,
72
+ "num_hidden_layers": 24,
73
+ "num_key_value_heads": 2,
74
+ "rms_norm_eps": 1e-06,
75
+ "rope_scaling": null,
76
+ "rope_theta": 1000000.0,
77
+ "sliding_window": null,
78
+ "tie_word_embeddings": true,
79
+ "use_cache": true,
80
+ "use_mrope": false,
81
+ "use_sliding_window": false,
82
+ "vocab_size": 151936
83
+ }
neuronxcc-2.21.18209.0+043b1bf7/MODULE_500ae99927947dab9c67+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_500ae99927947dab9c67+ed72d204/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_500ae99927947dab9c67+ed72d204/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9c97c8d96a50c1c323dfccff030cabaeb8225e7eb2f46b218de146f917b65e9a
3
+ size 739558
neuronxcc-2.21.18209.0+043b1bf7/MODULE_500ae99927947dab9c67+ed72d204/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4535a7584df6f2784cb3dc380caa23cbed1b1e8ac51239a37c3e9847e34aec26
3
+ size 26133504
neuronxcc-2.21.18209.0+043b1bf7/MODULE_852c8a18de93f098d898+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_852c8a18de93f098d898+a9d440f5/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_852c8a18de93f098d898+a9d440f5/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2c2920e5fde77a80d2d66c9ba5addf46e53d032f533c7097cba34e31243f84ed
3
+ size 588406
neuronxcc-2.21.18209.0+043b1bf7/MODULE_852c8a18de93f098d898+a9d440f5/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dfea45f987b901ba015b14a51d15bfe239798b1f92b947e5e7796c304317913c
3
+ size 1926144
neuronxcc-2.21.18209.0+043b1bf7/MODULE_852c8a18de93f098d898+a9d440f5/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:644b14c4fd20b466a4f127aefc9ee0f13696a3a8702bb349de00533b74e8b6cd
3
+ size 2082478