Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
Community
610
dacorvo HF Staff commited on
Commit
dea607e
·
verified ·
1 Parent(s): db84f27

Synchronizing local compiler cache.

Browse files
Files changed (11) hide show
  1. .gitattributes +3 -0
  2. neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/smollm3/HuggingFaceTB/SmolLM3-3B/765088ddf9ced28aa9a1.json +149 -0
  3. neuronxcc-2.19.8089.0+8ab9f450/MODULE_5d8b3bbf6ba518b8162f+ed72d204/compile_flags.json +1 -0
  4. neuronxcc-2.19.8089.0+8ab9f450/MODULE_5d8b3bbf6ba518b8162f+ed72d204/model.done +0 -0
  5. neuronxcc-2.19.8089.0+8ab9f450/MODULE_5d8b3bbf6ba518b8162f+ed72d204/model.hlo_module.pb +3 -0
  6. neuronxcc-2.19.8089.0+8ab9f450/MODULE_5d8b3bbf6ba518b8162f+ed72d204/model.neff +3 -0
  7. neuronxcc-2.19.8089.0+8ab9f450/MODULE_aa860ba18e95034cf093+a9d440f5/compile_flags.json +1 -0
  8. neuronxcc-2.19.8089.0+8ab9f450/MODULE_aa860ba18e95034cf093+a9d440f5/model.done +0 -0
  9. neuronxcc-2.19.8089.0+8ab9f450/MODULE_aa860ba18e95034cf093+a9d440f5/model.hlo_module.pb +3 -0
  10. neuronxcc-2.19.8089.0+8ab9f450/MODULE_aa860ba18e95034cf093+a9d440f5/model.neff +3 -0
  11. neuronxcc-2.19.8089.0+8ab9f450/MODULE_aa860ba18e95034cf093+a9d440f5/wrapped_neff.hlo +3 -0
.gitattributes CHANGED
@@ -10232,3 +10232,6 @@ neuronxcc-2.19.8089.0+8ab9f450/MODULE_e6d572d4fefccab468fd+a9d440f5/model.neff f
10232
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_e6d572d4fefccab468fd+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
10233
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_eaab44bef2023a6646ef+431f5505/model.neff filter=lfs diff=lfs merge=lfs -text
10234
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_f9ed5e3ab30730f18597+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
 
 
 
 
10232
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_e6d572d4fefccab468fd+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
10233
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_eaab44bef2023a6646ef+431f5505/model.neff filter=lfs diff=lfs merge=lfs -text
10234
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_f9ed5e3ab30730f18597+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
10235
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_5d8b3bbf6ba518b8162f+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
10236
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_aa860ba18e95034cf093+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
10237
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_aa860ba18e95034cf093+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev1/smollm3/HuggingFaceTB/SmolLM3-3B/765088ddf9ced28aa9a1.json ADDED
@@ -0,0 +1,149 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "async_mode": false,
59
+ "attn_kernel_enabled": false,
60
+ "batch_size": 16,
61
+ "capacity_factor": null,
62
+ "cc_pipeline_tiling_factor": 2,
63
+ "checkpoint_id": "HuggingFaceTB/SmolLM3-3B",
64
+ "checkpoint_revision": "988f1dde6bd2a36e3e7f45777a36507ea004fe22",
65
+ "continuous_batching": true,
66
+ "enable_bucketing": false,
67
+ "ep_degree": 1,
68
+ "flash_decoding_enabled": false,
69
+ "fused_qkv": true,
70
+ "glu_mlp": true,
71
+ "is_chunked_prefill": false,
72
+ "local_ranks_size": 2,
73
+ "logical_nc_config": 1,
74
+ "max_batch_size": 16,
75
+ "max_context_length": 4096,
76
+ "max_topk": 256,
77
+ "mlp_kernel_enabled": false,
78
+ "mlp_kernel_fuse_residual_add": false,
79
+ "n_active_tokens": 4096,
80
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
81
+ "num_cores_per_group": 1,
82
+ "on_device_sampling": true,
83
+ "optimum_neuron_version": "0.3.1.dev1",
84
+ "output_logits": false,
85
+ "padding_side": "right",
86
+ "pp_degree": 1,
87
+ "qk_layernorm": false,
88
+ "qkv_kernel_enabled": false,
89
+ "rpl_reduce_dtype": "bfloat16",
90
+ "sequence_length": 4096,
91
+ "sequence_parallel_enabled": false,
92
+ "speculation_length": 0,
93
+ "start_rank_id": 0,
94
+ "target": null,
95
+ "torch_dtype": "bfloat16",
96
+ "tp_degree": 2,
97
+ "vocab_parallel": false
98
+ },
99
+ "no_rope_layer_interval": 4,
100
+ "no_rope_layers": [
101
+ 1,
102
+ 1,
103
+ 1,
104
+ 0,
105
+ 1,
106
+ 1,
107
+ 1,
108
+ 0,
109
+ 1,
110
+ 1,
111
+ 1,
112
+ 0,
113
+ 1,
114
+ 1,
115
+ 1,
116
+ 0,
117
+ 1,
118
+ 1,
119
+ 1,
120
+ 0,
121
+ 1,
122
+ 1,
123
+ 1,
124
+ 0,
125
+ 1,
126
+ 1,
127
+ 1,
128
+ 0,
129
+ 1,
130
+ 1,
131
+ 1,
132
+ 0,
133
+ 1,
134
+ 1,
135
+ 1,
136
+ 0
137
+ ],
138
+ "num_attention_heads": 16,
139
+ "num_hidden_layers": 36,
140
+ "num_key_value_heads": 4,
141
+ "pretraining_tp": 2,
142
+ "rms_norm_eps": 1e-06,
143
+ "rope_scaling": null,
144
+ "rope_theta": 5000000.0,
145
+ "sliding_window": null,
146
+ "use_cache": false,
147
+ "use_sliding_window": false,
148
+ "vocab_size": 128256
149
+ }
neuronxcc-2.19.8089.0+8ab9f450/MODULE_5d8b3bbf6ba518b8162f+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-2.19.8089.0+8ab9f450/MODULE_5d8b3bbf6ba518b8162f+ed72d204/model.done ADDED
File without changes
neuronxcc-2.19.8089.0+8ab9f450/MODULE_5d8b3bbf6ba518b8162f+ed72d204/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:71939c494e52b35bec2295aff99bd7f69bc8ef24a7f82332fe817463fd82d580
3
+ size 1036280
neuronxcc-2.19.8089.0+8ab9f450/MODULE_5d8b3bbf6ba518b8162f+ed72d204/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f538962d0b2643238cbf7c34852046be3bba640e7007ac75555267e10dea7471
3
+ size 10681344
neuronxcc-2.19.8089.0+8ab9f450/MODULE_aa860ba18e95034cf093+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-2.19.8089.0+8ab9f450/MODULE_aa860ba18e95034cf093+a9d440f5/model.done ADDED
File without changes
neuronxcc-2.19.8089.0+8ab9f450/MODULE_aa860ba18e95034cf093+a9d440f5/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dee545e981e023f9ea361f2f60a26fa52fa682607225bf3b6eb0ea11fb36a86f
3
+ size 862478
neuronxcc-2.19.8089.0+8ab9f450/MODULE_aa860ba18e95034cf093+a9d440f5/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:54f46da234144703e575f63dbcc2d7c49b49b21c1a380b1677f907f501f3fd4f
3
+ size 6913024
neuronxcc-2.19.8089.0+8ab9f450/MODULE_aa860ba18e95034cf093+a9d440f5/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:876ed0fa40920c6776e14c4b6ef4b23753c635e562d81003a9284c26a28b7ef1
3
+ size 7069565