Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
 Community
695
dacorvo HF Staff commited on
Commit
7f893f9
·
verified ·
1 Parent(s): 8699720

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/smollm3/HuggingFaceTB/SmolLM3-3B/385fd862dfa03371766d.json +135 -0
  3. neuronxcc-2.21.18209.0+043b1bf7/MODULE_1889a7eeb014562de4da+ed72d204/compile_flags.json +1 -0
  4. neuronxcc-2.21.18209.0+043b1bf7/MODULE_1889a7eeb014562de4da+ed72d204/model.done +0 -0
  5. neuronxcc-2.21.18209.0+043b1bf7/MODULE_1889a7eeb014562de4da+ed72d204/model.hlo_module.pb +3 -0
  6. neuronxcc-2.21.18209.0+043b1bf7/MODULE_1889a7eeb014562de4da+ed72d204/model.neff +3 -0
  7. neuronxcc-2.21.18209.0+043b1bf7/MODULE_f46f3d7bae2e8595d580+a9d440f5/compile_flags.json +1 -0
  8. neuronxcc-2.21.18209.0+043b1bf7/MODULE_f46f3d7bae2e8595d580+a9d440f5/model.done +0 -0
  9. neuronxcc-2.21.18209.0+043b1bf7/MODULE_f46f3d7bae2e8595d580+a9d440f5/model.hlo_module.pb +3 -0
  10. neuronxcc-2.21.18209.0+043b1bf7/MODULE_f46f3d7bae2e8595d580+a9d440f5/model.neff +3 -0
  11. neuronxcc-2.21.18209.0+043b1bf7/MODULE_f46f3d7bae2e8595d580+a9d440f5/wrapped_neff.hlo +3 -0
.gitattributes CHANGED
@@ -11290,3 +11290,6 @@ neuronxcc-2.21.18209.0+043b1bf7/MODULE_870d96484095eb416802+a9d440f5/wrapped_nef
11290
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_746f8865916db9b26d04+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11291
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_d6cd8a3725dbdacc2b09+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11292
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_d6cd8a3725dbdacc2b09+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
11290
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_746f8865916db9b26d04+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11291
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_d6cd8a3725dbdacc2b09+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11292
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_d6cd8a3725dbdacc2b09+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
11293
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_1889a7eeb014562de4da+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11294
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_f46f3d7bae2e8595d580+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11295
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_f46f3d7bae2e8595d580+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.3.1.dev7/smollm3/HuggingFaceTB/SmolLM3-3B/385fd862dfa03371766d.json ADDED
@@ -0,0 +1,135 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "HuggingFaceTB/SmolLM3-3B",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "SmolLM3ForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 2048,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 11008,
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
+ "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
+ ],
52
+ "max_position_embeddings": 65536,
53
+ "max_window_layers": 28,
54
+ "mlp_bias": false,
55
+ "model_type": "smollm3",
56
+ "neuron": {
57
+ "_serialized_key": "NxDNeuronConfig",
58
+ "batch_size": 1,
59
+ "capacity_factor": null,
60
+ "checkpoint_id": "HuggingFaceTB/SmolLM3-3B",
61
+ "checkpoint_revision": "a07cc9a04f16550a088caea529712d1d335b0ac1",
62
+ "continuous_batching": false,
63
+ "enable_bucketing": false,
64
+ "ep_degree": 1,
65
+ "fused_qkv": true,
66
+ "glu_mlp": true,
67
+ "local_ranks_size": 2,
68
+ "logical_nc_config": 1,
69
+ "max_batch_size": 1,
70
+ "max_context_length": 4096,
71
+ "max_topk": 256,
72
+ "n_active_tokens": 4096,
73
+ "neuronxcc_version": "2.21.18209.0+043b1bf7",
74
+ "on_device_sampling": true,
75
+ "optimum_neuron_version": "0.3.1.dev7",
76
+ "output_logits": false,
77
+ "pp_degree": 1,
78
+ "sequence_length": 4096,
79
+ "speculation_length": 0,
80
+ "start_rank_id": 0,
81
+ "target": null,
82
+ "torch_dtype": "bfloat16",
83
+ "tp_degree": 2
84
+ },
85
+ "no_rope_layer_interval": 4,
86
+ "no_rope_layers": [
87
+ 1,
88
+ 1,
89
+ 1,
90
+ 0,
91
+ 1,
92
+ 1,
93
+ 1,
94
+ 0,
95
+ 1,
96
+ 1,
97
+ 1,
98
+ 0,
99
+ 1,
100
+ 1,
101
+ 1,
102
+ 0,
103
+ 1,
104
+ 1,
105
+ 1,
106
+ 0,
107
+ 1,
108
+ 1,
109
+ 1,
110
+ 0,
111
+ 1,
112
+ 1,
113
+ 1,
114
+ 0,
115
+ 1,
116
+ 1,
117
+ 1,
118
+ 0,
119
+ 1,
120
+ 1,
121
+ 1,
122
+ 0
123
+ ],
124
+ "num_attention_heads": 16,
125
+ "num_hidden_layers": 36,
126
+ "num_key_value_heads": 4,
127
+ "pretraining_tp": 2,
128
+ "rms_norm_eps": 1e-06,
129
+ "rope_scaling": null,
130
+ "rope_theta": 5000000.0,
131
+ "sliding_window": null,
132
+ "use_cache": false,
133
+ "use_sliding_window": false,
134
+ "vocab_size": 128256
135
+ }
neuronxcc-2.21.18209.0+043b1bf7/MODULE_1889a7eeb014562de4da+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_1889a7eeb014562de4da+ed72d204/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_1889a7eeb014562de4da+ed72d204/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:686f06874d649211026bdb82acb2ad4589550e52ff9031a7c2386e1d5bc1709a
3
+ size 982549
neuronxcc-2.21.18209.0+043b1bf7/MODULE_1889a7eeb014562de4da+ed72d204/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e7ae5b56c5fbd3c3cbc5bb21c052a5b8f7df043b72192973dd0db769bed5301c
3
+ size 9411584
neuronxcc-2.21.18209.0+043b1bf7/MODULE_f46f3d7bae2e8595d580+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_f46f3d7bae2e8595d580+a9d440f5/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_f46f3d7bae2e8595d580+a9d440f5/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:12ed68c10715600519e0818bdcdc7c076be520eac162690dd9f0d0403c6f0244
3
+ size 863008
neuronxcc-2.21.18209.0+043b1bf7/MODULE_f46f3d7bae2e8595d580+a9d440f5/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1a95496a492f8885b18caf3b12a11c425da500e111a8d6d313f4710cb7204760
3
+ size 5356544
neuronxcc-2.21.18209.0+043b1bf7/MODULE_f46f3d7bae2e8595d580+a9d440f5/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e34a76e238813ae943cbdff550f0f80305e8e26edf1ed9234c8fa6e3c4e43f42
3
+ size 5523211