Hugging Face's logo

optimum-internal-testing
/
neuron-testing-cache

Model card Files
xet
 Community
optimum-internal-testing-user commited on
Commit
d864856
·
verified ·
1 Parent(s): df87cfd

Synchronizing local compiler cache.

Browse files
Files changed (12) hide show
  1. .gitattributes +3 -0
  2. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev2/4cb7aff9e2a15c151396f2b684013e39d6739f0dec83e5c9dabbfe9d5fcf77b7/856d1b148c5b09824fd2.json +83 -0
  3. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev2/qwen2/Qwen/Qwen2.5-0.5B/856d1b148c5b09824fd2.json +83 -0
  4. neuronxcc-2.21.33363.0+82129205/MODULE_18044ed346a94d78250a+24129607/compile_flags.json +1 -0
  5. neuronxcc-2.21.33363.0+82129205/MODULE_18044ed346a94d78250a+24129607/model.done +0 -0
  6. neuronxcc-2.21.33363.0+82129205/MODULE_18044ed346a94d78250a+24129607/model.hlo_module.pb +3 -0
  7. neuronxcc-2.21.33363.0+82129205/MODULE_18044ed346a94d78250a+24129607/model.neff +3 -0
  8. neuronxcc-2.21.33363.0+82129205/MODULE_cafe5e1cc124bf1d35e0+a02c3a36/compile_flags.json +1 -0
  9. neuronxcc-2.21.33363.0+82129205/MODULE_cafe5e1cc124bf1d35e0+a02c3a36/model.done +0 -0
  10. neuronxcc-2.21.33363.0+82129205/MODULE_cafe5e1cc124bf1d35e0+a02c3a36/model.hlo_module.pb +3 -0
  11. neuronxcc-2.21.33363.0+82129205/MODULE_cafe5e1cc124bf1d35e0+a02c3a36/model.neff +3 -0
  12. neuronxcc-2.21.33363.0+82129205/MODULE_cafe5e1cc124bf1d35e0+a02c3a36/wrapped_neff.hlo +3 -0
.gitattributes CHANGED
@@ -7116,3 +7116,6 @@ neuronxcc-2.21.33363.0+82129205/MODULE_691f0ef9fca273527dbe+24129607/model.neff
7116
  neuronxcc-2.21.33363.0+82129205/MODULE_a3c6a617f4311e09ad49+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
7117
  neuronxcc-2.21.33363.0+82129205/MODULE_a3c6a617f4311e09ad49+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
7118
  neuronxcc-2.21.33363.0+82129205/MODULE_ab293424ba3cf7b684c0+6170d8e1/model.neff filter=lfs diff=lfs merge=lfs -text
 
 
 
 
7116
  neuronxcc-2.21.33363.0+82129205/MODULE_a3c6a617f4311e09ad49+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
7117
  neuronxcc-2.21.33363.0+82129205/MODULE_a3c6a617f4311e09ad49+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
7118
  neuronxcc-2.21.33363.0+82129205/MODULE_ab293424ba3cf7b684c0+6170d8e1/model.neff filter=lfs diff=lfs merge=lfs -text
7119
+ neuronxcc-2.21.33363.0+82129205/MODULE_18044ed346a94d78250a+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
7120
+ neuronxcc-2.21.33363.0+82129205/MODULE_cafe5e1cc124bf1d35e0+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
7121
+ neuronxcc-2.21.33363.0+82129205/MODULE_cafe5e1cc124bf1d35e0+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev2/4cb7aff9e2a15c151396f2b684013e39d6739f0dec83e5c9dabbfe9d5fcf77b7/856d1b148c5b09824fd2.json ADDED
@@ -0,0 +1,83 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen2.5-0.5B",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Qwen2ForCausalLM"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "dtype": "bfloat16",
10
+ "hidden_act": "silu",
11
+ "hidden_size": 896,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 4864,
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
+ ],
40
+ "max_position_embeddings": 32768,
41
+ "max_window_layers": 24,
42
+ "model_type": "qwen2",
43
+ "neuron": {
44
+ "_serialized_key": "NxDNeuronConfig",
45
+ "batch_size": 4,
46
+ "capacity_factor": null,
47
+ "checkpoint_id": "Qwen/Qwen2.5-0.5B",
48
+ "checkpoint_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
49
+ "continuous_batching": true,
50
+ "ep_degree": 1,
51
+ "fused_qkv": false,
52
+ "glu_mlp": true,
53
+ "local_ranks_size": 2,
54
+ "max_batch_size": 4,
55
+ "max_context_length": 1024,
56
+ "max_topk": 256,
57
+ "n_active_tokens": 1024,
58
+ "neuronxcc_version": "2.21.33363.0+82129205",
59
+ "on_device_sampling": false,
60
+ "optimum_neuron_version": "0.4.6.dev2",
61
+ "output_logits": false,
62
+ "pp_degree": 1,
63
+ "prefill_chunk_size": 0,
64
+ "sequence_length": 1024,
65
+ "speculation_length": 0,
66
+ "start_rank_id": 0,
67
+ "target": "trn1",
68
+ "torch_dtype": "bfloat16",
69
+ "tp_degree": 2
70
+ },
71
+ "num_attention_heads": 14,
72
+ "num_hidden_layers": 24,
73
+ "num_key_value_heads": 2,
74
+ "rms_norm_eps": 1e-06,
75
+ "rope_scaling": null,
76
+ "rope_theta": 1000000.0,
77
+ "sliding_window": null,
78
+ "tie_word_embeddings": true,
79
+ "use_cache": true,
80
+ "use_mrope": false,
81
+ "use_sliding_window": false,
82
+ "vocab_size": 151936
83
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev2/qwen2/Qwen/Qwen2.5-0.5B/856d1b148c5b09824fd2.json ADDED
@@ -0,0 +1,83 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen2.5-0.5B",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Qwen2ForCausalLM"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "dtype": "bfloat16",
10
+ "hidden_act": "silu",
11
+ "hidden_size": 896,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 4864,
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
+ ],
40
+ "max_position_embeddings": 32768,
41
+ "max_window_layers": 24,
42
+ "model_type": "qwen2",
43
+ "neuron": {
44
+ "_serialized_key": "NxDNeuronConfig",
45
+ "batch_size": 4,
46
+ "capacity_factor": null,
47
+ "checkpoint_id": "Qwen/Qwen2.5-0.5B",
48
+ "checkpoint_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
49
+ "continuous_batching": true,
50
+ "ep_degree": 1,
51
+ "fused_qkv": false,
52
+ "glu_mlp": true,
53
+ "local_ranks_size": 2,
54
+ "max_batch_size": 4,
55
+ "max_context_length": 1024,
56
+ "max_topk": 256,
57
+ "n_active_tokens": 1024,
58
+ "neuronxcc_version": "2.21.33363.0+82129205",
59
+ "on_device_sampling": false,
60
+ "optimum_neuron_version": "0.4.6.dev2",
61
+ "output_logits": false,
62
+ "pp_degree": 1,
63
+ "prefill_chunk_size": 0,
64
+ "sequence_length": 1024,
65
+ "speculation_length": 0,
66
+ "start_rank_id": 0,
67
+ "target": "trn1",
68
+ "torch_dtype": "bfloat16",
69
+ "tp_degree": 2
70
+ },
71
+ "num_attention_heads": 14,
72
+ "num_hidden_layers": 24,
73
+ "num_key_value_heads": 2,
74
+ "rms_norm_eps": 1e-06,
75
+ "rope_scaling": null,
76
+ "rope_theta": 1000000.0,
77
+ "sliding_window": null,
78
+ "tie_word_embeddings": true,
79
+ "use_cache": true,
80
+ "use_mrope": false,
81
+ "use_sliding_window": false,
82
+ "vocab_size": 151936
83
+ }
neuronxcc-2.21.33363.0+82129205/MODULE_18044ed346a94d78250a+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-2.21.33363.0+82129205/MODULE_18044ed346a94d78250a+24129607/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_18044ed346a94d78250a+24129607/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5b4a65dc71348db4d05d4fe2fbb3a05cdda74ca1c56e08940b2578439b7a6989
3
+ size 593552
neuronxcc-2.21.33363.0+82129205/MODULE_18044ed346a94d78250a+24129607/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fa360ad1aae0ed22de9993b117c014503fb298068845ce199bbbcb75469f728f
3
+ size 3769344
neuronxcc-2.21.33363.0+82129205/MODULE_cafe5e1cc124bf1d35e0+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-2.21.33363.0+82129205/MODULE_cafe5e1cc124bf1d35e0+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_cafe5e1cc124bf1d35e0+a02c3a36/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e883165574aed0354166f6d95aca4f607724646d2e006e98dba0278cf6ccbda4
3
+ size 616389
neuronxcc-2.21.33363.0+82129205/MODULE_cafe5e1cc124bf1d35e0+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6fb45d02aa525a3b0f2863b416ca17dc1c4cced5f0d8ae87dab2304bb03b0fd5
3
+ size 1793024
neuronxcc-2.21.33363.0+82129205/MODULE_cafe5e1cc124bf1d35e0+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:de701c4101e32e15296821c9187b1e6983882193bbd7bf7f06950ce3995ce8ba
3
+ size 1949215