Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
Community
643
dacorvo HF Staff commited on
Commit
f92a9d0
·
verified ·
1 Parent(s): cfc8283

Synchronizing local compiler cache.

Browse files
Files changed (20) hide show
  1. .gitattributes +5 -0
  2. neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev6/llama/meta-llama/Llama-3.1-8B-Instruct/d236b3c4d4059a15699b.json +63 -0
  3. neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev6/llama4_text/meta-llama/Llama-4-Scout-17B-16E-Instruct/3801b431cfcb1a8590ad.json +221 -0
  4. neuronxcc-2.19.8089.0+8ab9f450/MODULE_1301e0f1c0ef8a0c8119+a9d440f5/compile_flags.json +1 -0
  5. neuronxcc-2.19.8089.0+8ab9f450/MODULE_1301e0f1c0ef8a0c8119+a9d440f5/model.done +0 -0
  6. neuronxcc-2.19.8089.0+8ab9f450/MODULE_1301e0f1c0ef8a0c8119+a9d440f5/model.hlo_module.pb +3 -0
  7. neuronxcc-2.19.8089.0+8ab9f450/MODULE_1301e0f1c0ef8a0c8119+a9d440f5/model.neff +3 -0
  8. neuronxcc-2.19.8089.0+8ab9f450/MODULE_1301e0f1c0ef8a0c8119+a9d440f5/wrapped_neff.hlo +3 -0
  9. neuronxcc-2.19.8089.0+8ab9f450/MODULE_51e9eb4f870408488b15+ed72d204/compile_flags.json +1 -0
  10. neuronxcc-2.19.8089.0+8ab9f450/MODULE_51e9eb4f870408488b15+ed72d204/model.hlo_module.pb +3 -0
  11. neuronxcc-2.19.8089.0+8ab9f450/MODULE_51e9eb4f870408488b15+ed72d204/model.log +1 -0
  12. neuronxcc-2.19.8089.0+8ab9f450/MODULE_b48e6343d27f2b6101d4+ed72d204/compile_flags.json +1 -0
  13. neuronxcc-2.19.8089.0+8ab9f450/MODULE_b48e6343d27f2b6101d4+ed72d204/model.done +0 -0
  14. neuronxcc-2.19.8089.0+8ab9f450/MODULE_b48e6343d27f2b6101d4+ed72d204/model.hlo_module.pb +3 -0
  15. neuronxcc-2.19.8089.0+8ab9f450/MODULE_b48e6343d27f2b6101d4+ed72d204/model.neff +3 -0
  16. neuronxcc-2.19.8089.0+8ab9f450/MODULE_db1c520c3c70c18709b3+a9d440f5/compile_flags.json +1 -0
  17. neuronxcc-2.19.8089.0+8ab9f450/MODULE_db1c520c3c70c18709b3+a9d440f5/model.done +0 -0
  18. neuronxcc-2.19.8089.0+8ab9f450/MODULE_db1c520c3c70c18709b3+a9d440f5/model.hlo_module.pb +3 -0
  19. neuronxcc-2.19.8089.0+8ab9f450/MODULE_db1c520c3c70c18709b3+a9d440f5/model.neff +3 -0
  20. neuronxcc-2.19.8089.0+8ab9f450/MODULE_db1c520c3c70c18709b3+a9d440f5/wrapped_neff.hlo +3 -0
.gitattributes CHANGED
@@ -11005,3 +11005,8 @@ neuronxcc-2.19.8089.0+8ab9f450/MODULE_4474796592207674754+5860ac52/model.neff fi
11005
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_9d66f1cd194a8c2e1666+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11006
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_9d66f1cd194a8c2e1666+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
11007
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_dec56b9dc8ce732b5136+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
11005
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_9d66f1cd194a8c2e1666+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11006
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_9d66f1cd194a8c2e1666+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
11007
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_dec56b9dc8ce732b5136+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11008
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_1301e0f1c0ef8a0c8119+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11009
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_1301e0f1c0ef8a0c8119+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
11010
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_b48e6343d27f2b6101d4+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11011
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_db1c520c3c70c18709b3+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11012
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_db1c520c3c70c18709b3+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev6/llama/meta-llama/Llama-3.1-8B-Instruct/d236b3c4d4059a15699b.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "meta-llama/Llama-3.1-8B-Instruct",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "LlamaForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "head_dim": 128,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 4096,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 14336,
15
+ "max_position_embeddings": 131072,
16
+ "mlp_bias": false,
17
+ "model_type": "llama",
18
+ "neuron": {
19
+ "_serialized_key": "NxDNeuronConfig",
20
+ "batch_size": 1,
21
+ "capacity_factor": null,
22
+ "checkpoint_id": "meta-llama/Llama-3.1-8B-Instruct",
23
+ "checkpoint_revision": "0e9e39f249a16976918f6564b8830bc894c89659",
24
+ "continuous_batching": false,
25
+ "enable_bucketing": false,
26
+ "ep_degree": 1,
27
+ "fused_qkv": true,
28
+ "glu_mlp": true,
29
+ "local_ranks_size": 2,
30
+ "logical_nc_config": 1,
31
+ "max_batch_size": 1,
32
+ "max_context_length": 4096,
33
+ "max_topk": 256,
34
+ "n_active_tokens": 4096,
35
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
36
+ "on_device_sampling": true,
37
+ "optimum_neuron_version": "0.3.1.dev6",
38
+ "output_logits": false,
39
+ "pp_degree": 1,
40
+ "sequence_length": 4096,
41
+ "speculation_length": 0,
42
+ "start_rank_id": 0,
43
+ "target": null,
44
+ "torch_dtype": "bfloat16",
45
+ "tp_degree": 2
46
+ },
47
+ "num_attention_heads": 32,
48
+ "num_hidden_layers": 32,
49
+ "num_key_value_heads": 8,
50
+ "pretraining_tp": 1,
51
+ "rms_norm_eps": 1e-05,
52
+ "rope_scaling": {
53
+ "factor": 8.0,
54
+ "high_freq_factor": 4.0,
55
+ "low_freq_factor": 1.0,
56
+ "original_max_position_embeddings": 8192,
57
+ "rope_type": "llama3"
58
+ },
59
+ "rope_theta": 500000.0,
60
+ "tie_word_embeddings": false,
61
+ "use_cache": true,
62
+ "vocab_size": 128256
63
+ }
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev6/llama4_text/meta-llama/Llama-4-Scout-17B-16E-Instruct/3801b431cfcb1a8590ad.json ADDED
@@ -0,0 +1,221 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
4
+ "_task": "text-generation",
5
+ "attention_bias": false,
6
+ "attention_chunk_size": 8192,
7
+ "attention_dropout": 0.0,
8
+ "attn_scale": 0.1,
9
+ "attn_temperature_tuning": true,
10
+ "floor_scale": 8192,
11
+ "for_llm_compressor": false,
12
+ "head_dim": 128,
13
+ "hidden_act": "silu",
14
+ "hidden_size": 5120,
15
+ "initializer_range": 0.02,
16
+ "interleave_moe_layer_step": 1,
17
+ "intermediate_size": 8192,
18
+ "intermediate_size_mlp": 16384,
19
+ "layer_types": [
20
+ "chunked_attention",
21
+ "chunked_attention",
22
+ "chunked_attention",
23
+ "full_attention",
24
+ "chunked_attention",
25
+ "chunked_attention",
26
+ "chunked_attention",
27
+ "full_attention",
28
+ "chunked_attention",
29
+ "chunked_attention",
30
+ "chunked_attention",
31
+ "full_attention",
32
+ "chunked_attention",
33
+ "chunked_attention",
34
+ "chunked_attention",
35
+ "full_attention",
36
+ "chunked_attention",
37
+ "chunked_attention",
38
+ "chunked_attention",
39
+ "full_attention",
40
+ "chunked_attention",
41
+ "chunked_attention",
42
+ "chunked_attention",
43
+ "full_attention",
44
+ "chunked_attention",
45
+ "chunked_attention",
46
+ "chunked_attention",
47
+ "full_attention",
48
+ "chunked_attention",
49
+ "chunked_attention",
50
+ "chunked_attention",
51
+ "full_attention",
52
+ "chunked_attention",
53
+ "chunked_attention",
54
+ "chunked_attention",
55
+ "full_attention",
56
+ "chunked_attention",
57
+ "chunked_attention",
58
+ "chunked_attention",
59
+ "full_attention",
60
+ "chunked_attention",
61
+ "chunked_attention",
62
+ "chunked_attention",
63
+ "full_attention",
64
+ "chunked_attention",
65
+ "chunked_attention",
66
+ "chunked_attention",
67
+ "full_attention"
68
+ ],
69
+ "max_position_embeddings": 10485760,
70
+ "model_type": "llama4_text",
71
+ "moe_layers": [
72
+ 0,
73
+ 1,
74
+ 2,
75
+ 3,
76
+ 4,
77
+ 5,
78
+ 6,
79
+ 7,
80
+ 8,
81
+ 9,
82
+ 10,
83
+ 11,
84
+ 12,
85
+ 13,
86
+ 14,
87
+ 15,
88
+ 16,
89
+ 17,
90
+ 18,
91
+ 19,
92
+ 20,
93
+ 21,
94
+ 22,
95
+ 23,
96
+ 24,
97
+ 25,
98
+ 26,
99
+ 27,
100
+ 28,
101
+ 29,
102
+ 30,
103
+ 31,
104
+ 32,
105
+ 33,
106
+ 34,
107
+ 35,
108
+ 36,
109
+ 37,
110
+ 38,
111
+ 39,
112
+ 40,
113
+ 41,
114
+ 42,
115
+ 43,
116
+ 44,
117
+ 45,
118
+ 46,
119
+ 47
120
+ ],
121
+ "neuron": {
122
+ "_serialized_key": "NxDNeuronConfig",
123
+ "batch_size": 8,
124
+ "capacity_factor": null,
125
+ "checkpoint_id": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
126
+ "checkpoint_revision": "92f3b1597a195b523d8d9e5700e57e4fbb8f20d3",
127
+ "continuous_batching": true,
128
+ "enable_bucketing": false,
129
+ "ep_degree": 1,
130
+ "fused_qkv": false,
131
+ "glu_mlp": true,
132
+ "local_ranks_size": 32,
133
+ "logical_nc_config": 1,
134
+ "max_batch_size": 8,
135
+ "max_context_length": 4096,
136
+ "max_topk": 256,
137
+ "n_active_tokens": 4096,
138
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
139
+ "on_device_sampling": true,
140
+ "optimum_neuron_version": "0.3.1.dev6",
141
+ "output_logits": false,
142
+ "pp_degree": 1,
143
+ "sequence_length": 4096,
144
+ "speculation_length": 0,
145
+ "start_rank_id": 0,
146
+ "target": null,
147
+ "torch_dtype": "bfloat16",
148
+ "tp_degree": 32
149
+ },
150
+ "no_rope_layers": [
151
+ 1,
152
+ 1,
153
+ 1,
154
+ 0,
155
+ 1,
156
+ 1,
157
+ 1,
158
+ 0,
159
+ 1,
160
+ 1,
161
+ 1,
162
+ 0,
163
+ 1,
164
+ 1,
165
+ 1,
166
+ 0,
167
+ 1,
168
+ 1,
169
+ 1,
170
+ 0,
171
+ 1,
172
+ 1,
173
+ 1,
174
+ 0,
175
+ 1,
176
+ 1,
177
+ 1,
178
+ 0,
179
+ 1,
180
+ 1,
181
+ 1,
182
+ 0,
183
+ 1,
184
+ 1,
185
+ 1,
186
+ 0,
187
+ 1,
188
+ 1,
189
+ 1,
190
+ 0,
191
+ 1,
192
+ 1,
193
+ 1,
194
+ 0,
195
+ 1,
196
+ 1,
197
+ 1,
198
+ 0
199
+ ],
200
+ "num_attention_heads": 40,
201
+ "num_experts_per_tok": 1,
202
+ "num_hidden_layers": 48,
203
+ "num_key_value_heads": 8,
204
+ "num_local_experts": 16,
205
+ "output_router_logits": false,
206
+ "rms_norm_eps": 1e-05,
207
+ "rope_scaling": {
208
+ "factor": 16.0,
209
+ "high_freq_factor": 1.0,
210
+ "low_freq_factor": 1.0,
211
+ "original_max_position_embeddings": 8192,
212
+ "rope_type": "llama3"
213
+ },
214
+ "rope_theta": 500000.0,
215
+ "router_aux_loss_coef": 0.001,
216
+ "router_jitter_noise": 0.0,
217
+ "tie_word_embeddings": false,
218
+ "use_cache": true,
219
+ "use_qk_norm": true,
220
+ "vocab_size": 202048
221
+ }
neuronxcc-2.19.8089.0+8ab9f450/MODULE_1301e0f1c0ef8a0c8119+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_1301e0f1c0ef8a0c8119+a9d440f5/model.done ADDED
File without changes
neuronxcc-2.19.8089.0+8ab9f450/MODULE_1301e0f1c0ef8a0c8119+a9d440f5/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8f6f71581d7bcf46edc5db6400259e56c4a2bb1051d3b4a08cf9db02a0ab14ea
3
+ size 104109321
neuronxcc-2.19.8089.0+8ab9f450/MODULE_1301e0f1c0ef8a0c8119+a9d440f5/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0496b09d273d0eca7c5efcd57b964feffea9fd13e0e2bfca494c9d8ff3ad9aae
3
+ size 8664064
neuronxcc-2.19.8089.0+8ab9f450/MODULE_1301e0f1c0ef8a0c8119+a9d440f5/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a5e0f89723e7745d497a32c6fb3ec44d8afcbf30f81407318dc8f829a14b55c2
3
+ size 8981458
neuronxcc-2.19.8089.0+8ab9f450/MODULE_51e9eb4f870408488b15+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_51e9eb4f870408488b15+ed72d204/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c393022e6e823ec43d06b515a17075ac3c35fc349a23cea155d3153ae499c6ec
3
+ size 817423
neuronxcc-2.19.8089.0+8ab9f450/MODULE_51e9eb4f870408488b15+ed72d204/model.log ADDED
@@ -0,0 +1 @@
 
 
1
+ Failed compilation with ['neuronx-cc', 'compile', '--framework=XLA', '/tmp/nxd_model/context_encoding_model/_tp0_bk0/model.MODULE_51e9eb4f870408488b15+ed72d204.hlo_module.pb', '--output', '/tmp/nxd_model/context_encoding_model/_tp0_bk0/model.MODULE_51e9eb4f870408488b15+ed72d204.neff', '--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', '--verbose=35']:
neuronxcc-2.19.8089.0+8ab9f450/MODULE_b48e6343d27f2b6101d4+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_b48e6343d27f2b6101d4+ed72d204/model.done ADDED
File without changes
neuronxcc-2.19.8089.0+8ab9f450/MODULE_b48e6343d27f2b6101d4+ed72d204/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:440b16d2730c153e45e5557860323e5bcbf4ea91cd5cb06f3f071c4f79020ece
3
+ size 104360398
neuronxcc-2.19.8089.0+8ab9f450/MODULE_b48e6343d27f2b6101d4+ed72d204/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:386d560f2d60f7b468c45c9607142eabd1459b7ea2d2aa7128c69efd4e02516b
3
+ size 31529984
neuronxcc-2.19.8089.0+8ab9f450/MODULE_db1c520c3c70c18709b3+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_db1c520c3c70c18709b3+a9d440f5/model.done ADDED
File without changes
neuronxcc-2.19.8089.0+8ab9f450/MODULE_db1c520c3c70c18709b3+a9d440f5/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7c2bfd6280d23de045a4030ae2079bc810a8d1f90b53487b9cca7b69da4331fd
3
+ size 669679
neuronxcc-2.19.8089.0+8ab9f450/MODULE_db1c520c3c70c18709b3+a9d440f5/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3b0280f72f50dbed7bf935ed842f009a36bca860e853ebf547132cbb33e02ddd
3
+ size 1137664
neuronxcc-2.19.8089.0+8ab9f450/MODULE_db1c520c3c70c18709b3+a9d440f5/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ba57169cb3a32231de393f833c1ca2e75a986b00752247df542fd61537a8bcb0
3
+ size 1273550