Hugging Face's logo

optimum-internal-testing
/
neuron-testing-cache

Model card Files
xet
Community
tengomucho HF Staff commited on
Commit
3c7be6b
·
verified ·
1 Parent(s): 176ab48

Synchronizing local compiler cache.

Browse files
Files changed (20) hide show
  1. .gitattributes +4 -0
  2. neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.0/granite/ibm-granite/granite-3.1-2b-instruct/9c5870a09bc8c7cc70d7.json +73 -0
  3. neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.0/llama/llamafactory/tiny-random-Llama-3/5feb73ac8fb94b76fa22.json +77 -0
  4. neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.0/llama/llamafactory/tiny-random-Llama-3/7de5b54b5e3cd467e86c.json +77 -0
  5. neuronxcc-2.19.8089.0+8ab9f450/MODULE_3acc4723387c8a92e62d+ed72d204/compile_flags.json +1 -0
  6. neuronxcc-2.19.8089.0+8ab9f450/MODULE_3acc4723387c8a92e62d+ed72d204/model.hlo_module.pb +3 -0
  7. neuronxcc-2.19.8089.0+8ab9f450/MODULE_3acc4723387c8a92e62d+ed72d204/model.log +1 -0
  8. neuronxcc-2.19.8089.0+8ab9f450/MODULE_4637147670044247985+e30acd3a/compile_flags.json +1 -0
  9. neuronxcc-2.19.8089.0+8ab9f450/MODULE_4637147670044247985+e30acd3a/model.done +0 -0
  10. neuronxcc-2.19.8089.0+8ab9f450/MODULE_4637147670044247985+e30acd3a/model.hlo_module.pb +3 -0
  11. neuronxcc-2.19.8089.0+8ab9f450/MODULE_4637147670044247985+e30acd3a/model.neff +3 -0
  12. neuronxcc-2.19.8089.0+8ab9f450/MODULE_6559760091786342695+e30acd3a/compile_flags.json +1 -0
  13. neuronxcc-2.19.8089.0+8ab9f450/MODULE_6559760091786342695+e30acd3a/model.done +0 -0
  14. neuronxcc-2.19.8089.0+8ab9f450/MODULE_6559760091786342695+e30acd3a/model.hlo_module.pb +3 -0
  15. neuronxcc-2.19.8089.0+8ab9f450/MODULE_6559760091786342695+e30acd3a/model.neff +3 -0
  16. neuronxcc-2.19.8089.0+8ab9f450/MODULE_7095b61940c7e3c6fe2a+a9d440f5/compile_flags.json +1 -0
  17. neuronxcc-2.19.8089.0+8ab9f450/MODULE_7095b61940c7e3c6fe2a+a9d440f5/model.done +0 -0
  18. neuronxcc-2.19.8089.0+8ab9f450/MODULE_7095b61940c7e3c6fe2a+a9d440f5/model.hlo_module.pb +3 -0
  19. neuronxcc-2.19.8089.0+8ab9f450/MODULE_7095b61940c7e3c6fe2a+a9d440f5/model.neff +3 -0
  20. neuronxcc-2.19.8089.0+8ab9f450/MODULE_7095b61940c7e3c6fe2a+a9d440f5/wrapped_neff.hlo +3 -0
.gitattributes CHANGED
@@ -3704,3 +3704,7 @@ neuronxcc-2.19.8089.0+8ab9f450/MODULE_66c6c213e336c170168a+a9d440f5/model.neff f
3704
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_66c6c213e336c170168a+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
3705
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_a4c480627efff1a9a8e9+431f5505/model.neff filter=lfs diff=lfs merge=lfs -text
3706
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_fbe6a99f0ecb0d3eff11+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
3704
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_66c6c213e336c170168a+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
3705
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_a4c480627efff1a9a8e9+431f5505/model.neff filter=lfs diff=lfs merge=lfs -text
3706
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_fbe6a99f0ecb0d3eff11+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
3707
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_4637147670044247985+e30acd3a/model.neff filter=lfs diff=lfs merge=lfs -text
3708
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_6559760091786342695+e30acd3a/model.neff filter=lfs diff=lfs merge=lfs -text
3709
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_7095b61940c7e3c6fe2a+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
3710
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_7095b61940c7e3c6fe2a+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.0/granite/ibm-granite/granite-3.1-2b-instruct/9c5870a09bc8c7cc70d7.json ADDED
@@ -0,0 +1,73 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "ibm-granite/granite-3.1-2b-instruct",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "GraniteForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.1,
10
+ "attention_multiplier": 0.015625,
11
+ "embedding_multiplier": 12.0,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 2048,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 8192,
16
+ "logits_scaling": 8.0,
17
+ "max_position_embeddings": 131072,
18
+ "mlp_bias": false,
19
+ "model_type": "granite",
20
+ "neuron": {
21
+ "_serialized_key": "NxDNeuronConfig",
22
+ "async_mode": false,
23
+ "attn_kernel_enabled": false,
24
+ "batch_size": 4,
25
+ "capacity_factor": null,
26
+ "cc_pipeline_tiling_factor": 2,
27
+ "checkpoint_id": "ibm-granite/granite-3.1-2b-instruct",
28
+ "checkpoint_revision": "bbc2aed595bd38bd770263dc3ab831db9794441d",
29
+ "continuous_batching": true,
30
+ "enable_bucketing": false,
31
+ "ep_degree": 1,
32
+ "flash_decoding_enabled": false,
33
+ "fused_qkv": true,
34
+ "glu_mlp": true,
35
+ "is_chunked_prefill": false,
36
+ "local_ranks_size": 2,
37
+ "logical_nc_config": 1,
38
+ "max_batch_size": 4,
39
+ "max_context_length": 4096,
40
+ "max_topk": 256,
41
+ "mlp_kernel_enabled": false,
42
+ "mlp_kernel_fuse_residual_add": false,
43
+ "n_active_tokens": 4096,
44
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
45
+ "num_cores_per_group": 1,
46
+ "on_device_sampling": true,
47
+ "optimum_neuron_version": "0.3.0",
48
+ "output_logits": false,
49
+ "padding_side": "right",
50
+ "pp_degree": 1,
51
+ "qk_layernorm": false,
52
+ "qkv_kernel_enabled": false,
53
+ "rpl_reduce_dtype": "bfloat16",
54
+ "sequence_length": 4096,
55
+ "sequence_parallel_enabled": false,
56
+ "speculation_length": 0,
57
+ "start_rank_id": 0,
58
+ "target": null,
59
+ "torch_dtype": "bfloat16",
60
+ "tp_degree": 2,
61
+ "vocab_parallel": false
62
+ },
63
+ "num_attention_heads": 32,
64
+ "num_hidden_layers": 40,
65
+ "num_key_value_heads": 8,
66
+ "residual_multiplier": 0.22,
67
+ "rms_norm_eps": 1e-05,
68
+ "rope_scaling": null,
69
+ "rope_theta": 5000000.0,
70
+ "tie_word_embeddings": true,
71
+ "use_cache": true,
72
+ "vocab_size": 49155
73
+ }
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.0/llama/llamafactory/tiny-random-Llama-3/5feb73ac8fb94b76fa22.json ADDED
@@ -0,0 +1,77 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "llamafactory/tiny-random-Llama-3",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "LlamaForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "head_dim": 4,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 16,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 64,
15
+ "max_position_embeddings": 131072,
16
+ "mlp_bias": false,
17
+ "model_type": "llama",
18
+ "neuron": {
19
+ "_serialized_key": "NxDNeuronConfig",
20
+ "async_mode": false,
21
+ "attn_kernel_enabled": false,
22
+ "batch_size": 1,
23
+ "capacity_factor": null,
24
+ "cc_pipeline_tiling_factor": 2,
25
+ "checkpoint_id": "llamafactory/tiny-random-Llama-3",
26
+ "checkpoint_revision": "bf2a2e3bf199ad2ee96f02a3c00246c608db22a8",
27
+ "continuous_batching": false,
28
+ "enable_bucketing": false,
29
+ "ep_degree": 1,
30
+ "flash_decoding_enabled": false,
31
+ "fused_qkv": true,
32
+ "glu_mlp": true,
33
+ "is_chunked_prefill": false,
34
+ "local_ranks_size": 2,
35
+ "logical_nc_config": 1,
36
+ "max_batch_size": 1,
37
+ "max_context_length": 100,
38
+ "max_topk": 256,
39
+ "mlp_kernel_enabled": false,
40
+ "mlp_kernel_fuse_residual_add": false,
41
+ "n_active_tokens": 100,
42
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
43
+ "num_cores_per_group": 1,
44
+ "on_device_sampling": true,
45
+ "optimum_neuron_version": "0.3.0",
46
+ "output_logits": false,
47
+ "padding_side": "right",
48
+ "pp_degree": 1,
49
+ "qk_layernorm": false,
50
+ "qkv_kernel_enabled": false,
51
+ "rpl_reduce_dtype": "float16",
52
+ "sequence_length": 100,
53
+ "sequence_parallel_enabled": false,
54
+ "speculation_length": 0,
55
+ "start_rank_id": 0,
56
+ "target": null,
57
+ "torch_dtype": "float16",
58
+ "tp_degree": 2,
59
+ "vocab_parallel": false
60
+ },
61
+ "num_attention_heads": 4,
62
+ "num_hidden_layers": 2,
63
+ "num_key_value_heads": 4,
64
+ "pretraining_tp": 1,
65
+ "rms_norm_eps": 1e-05,
66
+ "rope_scaling": {
67
+ "factor": 8.0,
68
+ "high_freq_factor": 4.0,
69
+ "low_freq_factor": 1.0,
70
+ "original_max_position_embeddings": 8192,
71
+ "rope_type": "llama3"
72
+ },
73
+ "rope_theta": 500000.0,
74
+ "tie_word_embeddings": false,
75
+ "use_cache": true,
76
+ "vocab_size": 128256
77
+ }
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.0/llama/llamafactory/tiny-random-Llama-3/7de5b54b5e3cd467e86c.json ADDED
@@ -0,0 +1,77 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "llamafactory/tiny-random-Llama-3",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "LlamaForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "head_dim": 4,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 16,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 64,
15
+ "max_position_embeddings": 131072,
16
+ "mlp_bias": false,
17
+ "model_type": "llama",
18
+ "neuron": {
19
+ "_serialized_key": "NxDNeuronConfig",
20
+ "async_mode": false,
21
+ "attn_kernel_enabled": false,
22
+ "batch_size": 1,
23
+ "capacity_factor": null,
24
+ "cc_pipeline_tiling_factor": 2,
25
+ "checkpoint_id": "llamafactory/tiny-random-Llama-3",
26
+ "checkpoint_revision": "bf2a2e3bf199ad2ee96f02a3c00246c608db22a8",
27
+ "continuous_batching": false,
28
+ "enable_bucketing": false,
29
+ "ep_degree": 1,
30
+ "flash_decoding_enabled": false,
31
+ "fused_qkv": true,
32
+ "glu_mlp": true,
33
+ "is_chunked_prefill": false,
34
+ "local_ranks_size": 2,
35
+ "logical_nc_config": 1,
36
+ "max_batch_size": 1,
37
+ "max_context_length": 100,
38
+ "max_topk": 256,
39
+ "mlp_kernel_enabled": false,
40
+ "mlp_kernel_fuse_residual_add": false,
41
+ "n_active_tokens": 100,
42
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
43
+ "num_cores_per_group": 1,
44
+ "on_device_sampling": true,
45
+ "optimum_neuron_version": "0.3.0",
46
+ "output_logits": false,
47
+ "padding_side": "right",
48
+ "pp_degree": 1,
49
+ "qk_layernorm": false,
50
+ "qkv_kernel_enabled": false,
51
+ "rpl_reduce_dtype": "bfloat16",
52
+ "sequence_length": 100,
53
+ "sequence_parallel_enabled": false,
54
+ "speculation_length": 0,
55
+ "start_rank_id": 0,
56
+ "target": null,
57
+ "torch_dtype": "bfloat16",
58
+ "tp_degree": 2,
59
+ "vocab_parallel": false
60
+ },
61
+ "num_attention_heads": 4,
62
+ "num_hidden_layers": 2,
63
+ "num_key_value_heads": 4,
64
+ "pretraining_tp": 1,
65
+ "rms_norm_eps": 1e-05,
66
+ "rope_scaling": {
67
+ "factor": 8.0,
68
+ "high_freq_factor": 4.0,
69
+ "low_freq_factor": 1.0,
70
+ "original_max_position_embeddings": 8192,
71
+ "rope_type": "llama3"
72
+ },
73
+ "rope_theta": 500000.0,
74
+ "tie_word_embeddings": false,
75
+ "use_cache": true,
76
+ "vocab_size": 128256
77
+ }
neuronxcc-2.19.8089.0+8ab9f450/MODULE_3acc4723387c8a92e62d+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_3acc4723387c8a92e62d+ed72d204/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1c6cbcc26a65c26942a9749a3cd405ef9646b8f94bc6bc5477773bf3ee2352e9
3
+ size 33681414
neuronxcc-2.19.8089.0+8ab9f450/MODULE_3acc4723387c8a92e62d+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_3acc4723387c8a92e62d+ed72d204.hlo_module.pb', '--output', '/tmp/nxd_model/context_encoding_model/_tp0_bk0/model.MODULE_3acc4723387c8a92e62d+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_4637147670044247985+e30acd3a/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn1"]
neuronxcc-2.19.8089.0+8ab9f450/MODULE_4637147670044247985+e30acd3a/model.done ADDED
File without changes
neuronxcc-2.19.8089.0+8ab9f450/MODULE_4637147670044247985+e30acd3a/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6b4c980b2835a377ebef6a5ad2aefd95d44eb4fe744ec8c1cbca983f727260f2
3
+ size 76121
neuronxcc-2.19.8089.0+8ab9f450/MODULE_4637147670044247985+e30acd3a/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b69acaa64aa30337bf9c48eab405bee022bac6987c8d8e8449adcab6457aaffc
3
+ size 6841344
neuronxcc-2.19.8089.0+8ab9f450/MODULE_6559760091786342695+e30acd3a/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn1"]
neuronxcc-2.19.8089.0+8ab9f450/MODULE_6559760091786342695+e30acd3a/model.done ADDED
File without changes
neuronxcc-2.19.8089.0+8ab9f450/MODULE_6559760091786342695+e30acd3a/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ea670fb0b946bd19bcf39b2a0c56f4260c22f22595a6d758950c91a83b2bcd06
3
+ size 75924
neuronxcc-2.19.8089.0+8ab9f450/MODULE_6559760091786342695+e30acd3a/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6b85019441d66bb825381cf048c9405f15afaa985fcc92e7cfa87ad2e3292501
3
+ size 6800384
neuronxcc-2.19.8089.0+8ab9f450/MODULE_7095b61940c7e3c6fe2a+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_7095b61940c7e3c6fe2a+a9d440f5/model.done ADDED
File without changes
neuronxcc-2.19.8089.0+8ab9f450/MODULE_7095b61940c7e3c6fe2a+a9d440f5/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:66aa37f6b23bfcf647589e454ad25e3be44c0d85bbcdd00bce430fa84a19f50a
3
+ size 133857
neuronxcc-2.19.8089.0+8ab9f450/MODULE_7095b61940c7e3c6fe2a+a9d440f5/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:77a49de03b6f6b018324126abd848c7644104a5e7f58d4376eda6c531ba46798
3
+ size 3841024
neuronxcc-2.19.8089.0+8ab9f450/MODULE_7095b61940c7e3c6fe2a+a9d440f5/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2cd29ca3ed583447ce3154c3b2fb1cee9761a04327d88261102b411f4135b3cf
3
+ size 3869835