Hugging Face's logo

optimum-internal-testing
/
neuron-testing-cache

Model card Files
xet
Community
dacorvo HF Staff commited on
Commit
303a319
·
verified ·
1 Parent(s): b91964b

Synchronizing local compiler cache.

Browse files
Files changed (24) hide show
  1. .gitattributes +7 -0
  2. neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.2.dev2/llama4_text/meta-llama/Llama-4-Scout-17B-16E-Instruct/1ad2eec2d90dfdb5deb7.json +220 -0
  3. neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.2.dev2/qwen3/Qwen/Qwen3-Embedding-0.6B/9eb34185e11a0b7c5ac2.json +87 -0
  4. neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.2.dev2/qwen3/Qwen/Qwen3-Embedding-0.6B/d652fbacbf3a73cb1caa.json +87 -0
  5. neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.3.dev1/llama/unsloth/Llama-3.2-1B-Instruct/889961c656ed9f2df481.json +63 -0
  6. neuronxcc-2.21.18209.0+043b1bf7/MODULE_04355e038292f7e78776+a02c3a36/compile_flags.json +1 -0
  7. neuronxcc-2.21.18209.0+043b1bf7/MODULE_04355e038292f7e78776+a02c3a36/model.done +0 -0
  8. neuronxcc-2.21.18209.0+043b1bf7/MODULE_04355e038292f7e78776+a02c3a36/model.hlo_module.pb +3 -0
  9. neuronxcc-2.21.18209.0+043b1bf7/MODULE_04355e038292f7e78776+a02c3a36/model.neff +3 -0
  10. neuronxcc-2.21.18209.0+043b1bf7/MODULE_04355e038292f7e78776+a02c3a36/wrapped_neff.hlo +3 -0
  11. neuronxcc-2.21.18209.0+043b1bf7/MODULE_3ae9ea3e6d448b5136e5+24129607/compile_flags.json +1 -0
  12. neuronxcc-2.21.18209.0+043b1bf7/MODULE_3ae9ea3e6d448b5136e5+24129607/model.done +0 -0
  13. neuronxcc-2.21.18209.0+043b1bf7/MODULE_3ae9ea3e6d448b5136e5+24129607/model.hlo_module.pb +3 -0
  14. neuronxcc-2.21.18209.0+043b1bf7/MODULE_3ae9ea3e6d448b5136e5+24129607/model.neff +3 -0
  15. neuronxcc-2.21.18209.0+043b1bf7/MODULE_6cb05e0d4a5faa35e069+63e4cd48/compile_flags.json +1 -0
  16. neuronxcc-2.21.18209.0+043b1bf7/MODULE_6cb05e0d4a5faa35e069+63e4cd48/model.done +0 -0
  17. neuronxcc-2.21.18209.0+043b1bf7/MODULE_6cb05e0d4a5faa35e069+63e4cd48/model.hlo_module.pb +3 -0
  18. neuronxcc-2.21.18209.0+043b1bf7/MODULE_6cb05e0d4a5faa35e069+63e4cd48/model.neff +3 -0
  19. neuronxcc-2.21.18209.0+043b1bf7/MODULE_6cb05e0d4a5faa35e069+63e4cd48/wrapped_neff.hlo +3 -0
  20. neuronxcc-2.21.18209.0+043b1bf7/MODULE_bd37a4fdfc1529582241+cc83d24a/compile_flags.json +1 -0
  21. neuronxcc-2.21.18209.0+043b1bf7/MODULE_bd37a4fdfc1529582241+cc83d24a/model.done +0 -0
  22. neuronxcc-2.21.18209.0+043b1bf7/MODULE_bd37a4fdfc1529582241+cc83d24a/model.hlo_module.pb +3 -0
  23. neuronxcc-2.21.18209.0+043b1bf7/MODULE_bd37a4fdfc1529582241+cc83d24a/model.neff +3 -0
  24. neuronxcc-2.21.18209.0+043b1bf7/MODULE_bd37a4fdfc1529582241+cc83d24a/wrapped_neff.hlo +3 -0
.gitattributes CHANGED
@@ -5500,3 +5500,10 @@ neuronxcc-2.21.18209.0+043b1bf7/MODULE_d4fe40d7566b309ac86a+a02c3a36/model.neff
5500
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_d4fe40d7566b309ac86a+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
5501
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_0eca6f3c78de14ecd159+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
5502
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_0eca6f3c78de14ecd159+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
 
5500
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_d4fe40d7566b309ac86a+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
5501
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_0eca6f3c78de14ecd159+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
5502
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_0eca6f3c78de14ecd159+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
5503
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_04355e038292f7e78776+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
5504
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_04355e038292f7e78776+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
5505
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_3ae9ea3e6d448b5136e5+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
5506
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_6cb05e0d4a5faa35e069+63e4cd48/model.neff filter=lfs diff=lfs merge=lfs -text
5507
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_6cb05e0d4a5faa35e069+63e4cd48/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
5508
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_bd37a4fdfc1529582241+cc83d24a/model.neff filter=lfs diff=lfs merge=lfs -text
5509
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_bd37a4fdfc1529582241+cc83d24a/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.2.dev2/llama4_text/meta-llama/Llama-4-Scout-17B-16E-Instruct/1ad2eec2d90dfdb5deb7.json ADDED
@@ -0,0 +1,220 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "dtype": "bfloat16",
11
+ "floor_scale": 8192,
12
+ "for_llm_compressor": false,
13
+ "head_dim": 128,
14
+ "hidden_act": "silu",
15
+ "hidden_size": 5120,
16
+ "initializer_range": 0.02,
17
+ "interleave_moe_layer_step": 1,
18
+ "intermediate_size": 8192,
19
+ "intermediate_size_mlp": 16384,
20
+ "layer_types": [
21
+ "chunked_attention",
22
+ "chunked_attention",
23
+ "chunked_attention",
24
+ "full_attention",
25
+ "chunked_attention",
26
+ "chunked_attention",
27
+ "chunked_attention",
28
+ "full_attention",
29
+ "chunked_attention",
30
+ "chunked_attention",
31
+ "chunked_attention",
32
+ "full_attention",
33
+ "chunked_attention",
34
+ "chunked_attention",
35
+ "chunked_attention",
36
+ "full_attention",
37
+ "chunked_attention",
38
+ "chunked_attention",
39
+ "chunked_attention",
40
+ "full_attention",
41
+ "chunked_attention",
42
+ "chunked_attention",
43
+ "chunked_attention",
44
+ "full_attention",
45
+ "chunked_attention",
46
+ "chunked_attention",
47
+ "chunked_attention",
48
+ "full_attention",
49
+ "chunked_attention",
50
+ "chunked_attention",
51
+ "chunked_attention",
52
+ "full_attention",
53
+ "chunked_attention",
54
+ "chunked_attention",
55
+ "chunked_attention",
56
+ "full_attention",
57
+ "chunked_attention",
58
+ "chunked_attention",
59
+ "chunked_attention",
60
+ "full_attention",
61
+ "chunked_attention",
62
+ "chunked_attention",
63
+ "chunked_attention",
64
+ "full_attention",
65
+ "chunked_attention",
66
+ "chunked_attention",
67
+ "chunked_attention",
68
+ "full_attention"
69
+ ],
70
+ "max_position_embeddings": 10485760,
71
+ "model_type": "llama4_text",
72
+ "moe_layers": [
73
+ 0,
74
+ 1,
75
+ 2,
76
+ 3,
77
+ 4,
78
+ 5,
79
+ 6,
80
+ 7,
81
+ 8,
82
+ 9,
83
+ 10,
84
+ 11,
85
+ 12,
86
+ 13,
87
+ 14,
88
+ 15,
89
+ 16,
90
+ 17,
91
+ 18,
92
+ 19,
93
+ 20,
94
+ 21,
95
+ 22,
96
+ 23,
97
+ 24,
98
+ 25,
99
+ 26,
100
+ 27,
101
+ 28,
102
+ 29,
103
+ 30,
104
+ 31,
105
+ 32,
106
+ 33,
107
+ 34,
108
+ 35,
109
+ 36,
110
+ 37,
111
+ 38,
112
+ 39,
113
+ 40,
114
+ 41,
115
+ 42,
116
+ 43,
117
+ 44,
118
+ 45,
119
+ 46,
120
+ 47
121
+ ],
122
+ "neuron": {
123
+ "_serialized_key": "NxDNeuronConfig",
124
+ "batch_size": 1,
125
+ "capacity_factor": null,
126
+ "checkpoint_id": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
127
+ "checkpoint_revision": "92f3b1597a195b523d8d9e5700e57e4fbb8f20d3",
128
+ "continuous_batching": false,
129
+ "ep_degree": 1,
130
+ "fused_qkv": false,
131
+ "glu_mlp": true,
132
+ "local_ranks_size": 16,
133
+ "max_batch_size": 1,
134
+ "max_context_length": 4096,
135
+ "max_topk": 256,
136
+ "n_active_tokens": 4096,
137
+ "neuronxcc_version": "2.21.18209.0+043b1bf7",
138
+ "on_device_sampling": true,
139
+ "optimum_neuron_version": "0.4.2.dev2",
140
+ "output_logits": false,
141
+ "pp_degree": 1,
142
+ "sequence_length": 4096,
143
+ "speculation_length": 0,
144
+ "start_rank_id": 0,
145
+ "target": "trn1",
146
+ "torch_dtype": "bfloat16",
147
+ "tp_degree": 16
148
+ },
149
+ "no_rope_layers": [
150
+ 1,
151
+ 1,
152
+ 1,
153
+ 0,
154
+ 1,
155
+ 1,
156
+ 1,
157
+ 0,
158
+ 1,
159
+ 1,
160
+ 1,
161
+ 0,
162
+ 1,
163
+ 1,
164
+ 1,
165
+ 0,
166
+ 1,
167
+ 1,
168
+ 1,
169
+ 0,
170
+ 1,
171
+ 1,
172
+ 1,
173
+ 0,
174
+ 1,
175
+ 1,
176
+ 1,
177
+ 0,
178
+ 1,
179
+ 1,
180
+ 1,
181
+ 0,
182
+ 1,
183
+ 1,
184
+ 1,
185
+ 0,
186
+ 1,
187
+ 1,
188
+ 1,
189
+ 0,
190
+ 1,
191
+ 1,
192
+ 1,
193
+ 0,
194
+ 1,
195
+ 1,
196
+ 1,
197
+ 0
198
+ ],
199
+ "num_attention_heads": 40,
200
+ "num_experts_per_tok": 1,
201
+ "num_hidden_layers": 48,
202
+ "num_key_value_heads": 8,
203
+ "num_local_experts": 16,
204
+ "output_router_logits": false,
205
+ "rms_norm_eps": 1e-05,
206
+ "rope_scaling": {
207
+ "factor": 16.0,
208
+ "high_freq_factor": 1.0,
209
+ "low_freq_factor": 1.0,
210
+ "original_max_position_embeddings": 8192,
211
+ "rope_type": "llama3"
212
+ },
213
+ "rope_theta": 500000.0,
214
+ "router_aux_loss_coef": 0.001,
215
+ "router_jitter_noise": 0.0,
216
+ "tie_word_embeddings": false,
217
+ "use_cache": true,
218
+ "use_qk_norm": true,
219
+ "vocab_size": 202048
220
+ }
neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.2.dev2/qwen3/Qwen/Qwen3-Embedding-0.6B/9eb34185e11a0b7c5ac2.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen3-Embedding-0.6B",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Qwen3ForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "dtype": "bfloat16",
11
+ "head_dim": 128,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 1024,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 3072,
16
+ "layer_types": [
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
+ ],
46
+ "max_position_embeddings": 32768,
47
+ "max_window_layers": 28,
48
+ "model_type": "qwen3",
49
+ "neuron": {
50
+ "_serialized_key": "NxDNeuronConfig",
51
+ "batch_size": 4,
52
+ "capacity_factor": null,
53
+ "checkpoint_id": "Qwen/Qwen3-Embedding-0.6B",
54
+ "checkpoint_revision": "c54f2e6e80b2d7b7de06f51cec4959f6b3e03418",
55
+ "continuous_batching": false,
56
+ "ep_degree": 1,
57
+ "fused_qkv": true,
58
+ "glu_mlp": true,
59
+ "local_ranks_size": 24,
60
+ "max_batch_size": 4,
61
+ "max_context_length": 1024,
62
+ "max_topk": 256,
63
+ "n_active_tokens": 1024,
64
+ "neuronxcc_version": "2.21.18209.0+043b1bf7",
65
+ "on_device_sampling": false,
66
+ "optimum_neuron_version": "0.4.2.dev2",
67
+ "output_logits": false,
68
+ "pp_degree": 1,
69
+ "sequence_length": 1024,
70
+ "speculation_length": 0,
71
+ "start_rank_id": 0,
72
+ "target": "trn1",
73
+ "torch_dtype": "bfloat16",
74
+ "tp_degree": 24
75
+ },
76
+ "num_attention_heads": 16,
77
+ "num_hidden_layers": 28,
78
+ "num_key_value_heads": 8,
79
+ "rms_norm_eps": 1e-06,
80
+ "rope_scaling": null,
81
+ "rope_theta": 1000000,
82
+ "sliding_window": null,
83
+ "tie_word_embeddings": true,
84
+ "use_cache": true,
85
+ "use_sliding_window": false,
86
+ "vocab_size": 151669
87
+ }
neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.2.dev2/qwen3/Qwen/Qwen3-Embedding-0.6B/d652fbacbf3a73cb1caa.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen3-Embedding-0.6B",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Qwen3ForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "dtype": "bfloat16",
11
+ "head_dim": 128,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 1024,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 3072,
16
+ "layer_types": [
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
+ ],
46
+ "max_position_embeddings": 32768,
47
+ "max_window_layers": 28,
48
+ "model_type": "qwen3",
49
+ "neuron": {
50
+ "_serialized_key": "NxDNeuronConfig",
51
+ "batch_size": 6,
52
+ "capacity_factor": null,
53
+ "checkpoint_id": "Qwen/Qwen3-Embedding-0.6B",
54
+ "checkpoint_revision": "c54f2e6e80b2d7b7de06f51cec4959f6b3e03418",
55
+ "continuous_batching": false,
56
+ "ep_degree": 1,
57
+ "fused_qkv": true,
58
+ "glu_mlp": true,
59
+ "local_ranks_size": 24,
60
+ "max_batch_size": 6,
61
+ "max_context_length": 1024,
62
+ "max_topk": 256,
63
+ "n_active_tokens": 1024,
64
+ "neuronxcc_version": "2.21.18209.0+043b1bf7",
65
+ "on_device_sampling": false,
66
+ "optimum_neuron_version": "0.4.2.dev2",
67
+ "output_logits": false,
68
+ "pp_degree": 1,
69
+ "sequence_length": 1024,
70
+ "speculation_length": 0,
71
+ "start_rank_id": 0,
72
+ "target": "trn1",
73
+ "torch_dtype": "bfloat16",
74
+ "tp_degree": 24
75
+ },
76
+ "num_attention_heads": 16,
77
+ "num_hidden_layers": 28,
78
+ "num_key_value_heads": 8,
79
+ "rms_norm_eps": 1e-06,
80
+ "rope_scaling": null,
81
+ "rope_theta": 1000000,
82
+ "sliding_window": null,
83
+ "tie_word_embeddings": true,
84
+ "use_cache": true,
85
+ "use_sliding_window": false,
86
+ "vocab_size": 151669
87
+ }
neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.3.dev1/llama/unsloth/Llama-3.2-1B-Instruct/889961c656ed9f2df481.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "unsloth/Llama-3.2-1B-Instruct",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "LlamaForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "dtype": "bfloat16",
11
+ "head_dim": 64,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 2048,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 8192,
16
+ "max_position_embeddings": 131072,
17
+ "mlp_bias": false,
18
+ "model_type": "llama",
19
+ "neuron": {
20
+ "_serialized_key": "NxDNeuronConfig",
21
+ "batch_size": 4,
22
+ "capacity_factor": null,
23
+ "checkpoint_id": "unsloth/Llama-3.2-1B-Instruct",
24
+ "checkpoint_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
25
+ "continuous_batching": true,
26
+ "ep_degree": 1,
27
+ "fused_qkv": true,
28
+ "glu_mlp": true,
29
+ "local_ranks_size": 2,
30
+ "max_batch_size": 4,
31
+ "max_context_length": 4096,
32
+ "max_topk": 256,
33
+ "n_active_tokens": 4096,
34
+ "neuronxcc_version": "2.21.18209.0+043b1bf7",
35
+ "on_device_sampling": true,
36
+ "optimum_neuron_version": "0.4.3.dev1",
37
+ "output_logits": false,
38
+ "pp_degree": 1,
39
+ "sequence_length": 4096,
40
+ "speculation_length": 0,
41
+ "start_rank_id": 0,
42
+ "target": "trn1",
43
+ "torch_dtype": "bfloat16",
44
+ "tp_degree": 2
45
+ },
46
+ "num_attention_heads": 32,
47
+ "num_hidden_layers": 16,
48
+ "num_key_value_heads": 8,
49
+ "pretraining_tp": 1,
50
+ "rms_norm_eps": 1e-05,
51
+ "rope_scaling": {
52
+ "factor": 32.0,
53
+ "high_freq_factor": 4.0,
54
+ "low_freq_factor": 1.0,
55
+ "original_max_position_embeddings": 8192,
56
+ "rope_type": "llama3"
57
+ },
58
+ "rope_theta": 500000.0,
59
+ "tie_word_embeddings": true,
60
+ "unsloth_fixed": true,
61
+ "use_cache": true,
62
+ "vocab_size": 128256
63
+ }
neuronxcc-2.21.18209.0+043b1bf7/MODULE_04355e038292f7e78776+a02c3a36/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/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.21.18209.0+043b1bf7/MODULE_04355e038292f7e78776+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_04355e038292f7e78776+a02c3a36/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ebab4bcb5f71de3a8c7e9ecf957d2d1a47a45c30507a1af8ee396079595387a9
3
+ size 61563
neuronxcc-2.21.18209.0+043b1bf7/MODULE_04355e038292f7e78776+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0df619b9ece922dd62783e4b0ac5ab759bb549419e1ed33401616afaf9702869
3
+ size 205824
neuronxcc-2.21.18209.0+043b1bf7/MODULE_04355e038292f7e78776+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:08038f75d84ba51623775c09b83eb1bf4df5d414b77d197fcaf802a163ec3b72
3
+ size 212110
neuronxcc-2.21.18209.0+043b1bf7/MODULE_3ae9ea3e6d448b5136e5+24129607/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/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.21.18209.0+043b1bf7/MODULE_3ae9ea3e6d448b5136e5+24129607/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_3ae9ea3e6d448b5136e5+24129607/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e9b8c8ce8bf2645b1a2d4c1e7dde81c2e4b92aa5eddf85da33bdc53680115649
3
+ size 64986
neuronxcc-2.21.18209.0+043b1bf7/MODULE_3ae9ea3e6d448b5136e5+24129607/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:60e5e835f471c5a629ce6f9292ef58d7007582c84cf5067cd674fe3c91b38e94
3
+ size 226304
neuronxcc-2.21.18209.0+043b1bf7/MODULE_6cb05e0d4a5faa35e069+63e4cd48/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn1", "--enable-saturate-infinity", "--auto-cast=none", "--model-type=transformer", "-O1", "--logfile=/tmp/nxdi_test_9edca7e2-41cb-40c3-8325-b0fb5293fb62/compiler_workdir/AttentionModelWrapper/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.21.18209.0+043b1bf7/MODULE_6cb05e0d4a5faa35e069+63e4cd48/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_6cb05e0d4a5faa35e069+63e4cd48/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4590b82eb167fd46963abdbb717cc18cb9511961a3393d100eeacdbdc45b0243
3
+ size 11280
neuronxcc-2.21.18209.0+043b1bf7/MODULE_6cb05e0d4a5faa35e069+63e4cd48/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:84d88ad3925c3af294657c0f04d0b2f5f8d09521b77b5e2a8bdc563e4c00d8a0
3
+ size 1444864
neuronxcc-2.21.18209.0+043b1bf7/MODULE_6cb05e0d4a5faa35e069+63e4cd48/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:411bdbd53d0de363857053e108f7cb310c386b791210da632a7b31a555b4127a
3
+ size 1447845
neuronxcc-2.21.18209.0+043b1bf7/MODULE_bd37a4fdfc1529582241+cc83d24a/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn1", "--enable-saturate-infinity", "--auto-cast=none", "--model-type=transformer", "-O1", "--logfile=/tmp/nxdi_test_43ee6255-d046-4a57-9b05-519269f01d32/compiler_workdir/AttentionModelWrapper/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.21.18209.0+043b1bf7/MODULE_bd37a4fdfc1529582241+cc83d24a/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_bd37a4fdfc1529582241+cc83d24a/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:51e79a5cc325f9a317d158d931d75ec73ae1cfb16a4d83343a2593edd95d0044
3
+ size 8979
neuronxcc-2.21.18209.0+043b1bf7/MODULE_bd37a4fdfc1529582241+cc83d24a/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ba41cd0e0df956a61ecfcd0b97b8c43a49826105f5391801aa3021dc828a0434
3
+ size 246784
neuronxcc-2.21.18209.0+043b1bf7/MODULE_bd37a4fdfc1529582241+cc83d24a/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:287f57163eb5a2db82088b959bd8ac1d67bccb4d45ca4b94d631e30371e8b394
3
+ size 249608