Hugging Face's logo

optimum-internal-testing
/
neuron-testing-cache

Model card Files
xet
Community
dacorvo HF Staff commited on
Commit
0e4ed53
·
verified ·
1 Parent(s): 948a752

Synchronizing local compiler cache.

Browse files
Files changed (12) hide show
  1. .gitattributes +3 -0
  2. neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.2.dev1/llama/unsloth/Llama-3.2-1B-Instruct/7d08bf5fb4f9a14c4981.json +63 -0
  3. neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.2.dev1/qwen3/Qwen/Qwen3-0.6B/4ddf27fb5dfd4909a319.json +87 -0
  4. neuronxcc-2.21.18209.0+043b1bf7/MODULE_d1e28d83b0b75daadd33+24129607/compile_flags.json +1 -0
  5. neuronxcc-2.21.18209.0+043b1bf7/MODULE_d1e28d83b0b75daadd33+24129607/model.done +0 -0
  6. neuronxcc-2.21.18209.0+043b1bf7/MODULE_d1e28d83b0b75daadd33+24129607/model.hlo_module.pb +3 -0
  7. neuronxcc-2.21.18209.0+043b1bf7/MODULE_d1e28d83b0b75daadd33+24129607/model.neff +3 -0
  8. neuronxcc-2.21.18209.0+043b1bf7/MODULE_d4fe40d7566b309ac86a+a02c3a36/compile_flags.json +1 -0
  9. neuronxcc-2.21.18209.0+043b1bf7/MODULE_d4fe40d7566b309ac86a+a02c3a36/model.done +0 -0
  10. neuronxcc-2.21.18209.0+043b1bf7/MODULE_d4fe40d7566b309ac86a+a02c3a36/model.hlo_module.pb +3 -0
  11. neuronxcc-2.21.18209.0+043b1bf7/MODULE_d4fe40d7566b309ac86a+a02c3a36/model.neff +3 -0
  12. neuronxcc-2.21.18209.0+043b1bf7/MODULE_d4fe40d7566b309ac86a+a02c3a36/wrapped_neff.hlo +3 -0
.gitattributes CHANGED
@@ -5495,3 +5495,6 @@ neuronxcc-2.21.18209.0+043b1bf7/MODULE_962a482f90524ff403f5+24129607/model.neff
5495
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_accf1fdc5eb1cee6c292+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
5496
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_accf1fdc5eb1cee6c292+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
5497
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_b79a786451ff8fc0676d+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
 
 
 
 
5495
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_accf1fdc5eb1cee6c292+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
5496
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_accf1fdc5eb1cee6c292+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
5497
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_b79a786451ff8fc0676d+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
5498
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_d1e28d83b0b75daadd33+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
5499
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_d4fe40d7566b309ac86a+a02c3a36/model.neff 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
neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.2.dev1/llama/unsloth/Llama-3.2-1B-Instruct/7d08bf5fb4f9a14c4981.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": "0.4.2.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+043b1bf7/0_REGISTRY/0.4.2.dev1/qwen3/Qwen/Qwen3-0.6B/4ddf27fb5dfd4909a319.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen3-0.6B",
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": 40960,
47
+ "max_window_layers": 28,
48
+ "model_type": "qwen3",
49
+ "neuron": {
50
+ "_serialized_key": "NxDNeuronConfig",
51
+ "batch_size": 2,
52
+ "capacity_factor": null,
53
+ "checkpoint_id": "Qwen/Qwen3-0.6B",
54
+ "checkpoint_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
55
+ "continuous_batching": true,
56
+ "ep_degree": 1,
57
+ "fused_qkv": true,
58
+ "glu_mlp": true,
59
+ "local_ranks_size": 2,
60
+ "max_batch_size": 2,
61
+ "max_context_length": 2048,
62
+ "max_topk": 256,
63
+ "n_active_tokens": 2048,
64
+ "neuronxcc_version": "2.21.18209.0+043b1bf7",
65
+ "on_device_sampling": false,
66
+ "optimum_neuron_version": "0.4.2.dev1",
67
+ "output_logits": false,
68
+ "pp_degree": 1,
69
+ "sequence_length": 2048,
70
+ "speculation_length": 0,
71
+ "start_rank_id": 0,
72
+ "target": "trn1",
73
+ "torch_dtype": "bfloat16",
74
+ "tp_degree": 2
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": 151936
87
+ }
neuronxcc-2.21.18209.0+043b1bf7/MODULE_d1e28d83b0b75daadd33+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_d1e28d83b0b75daadd33+24129607/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_d1e28d83b0b75daadd33+24129607/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf35cc5188092a5ed97a2fd36b936a577ef7dc5f8e103e38714748065ecdd328
3
+ size 693532
neuronxcc-2.21.18209.0+043b1bf7/MODULE_d1e28d83b0b75daadd33+24129607/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b0bbf7638bc657c8f8d72fdd57dcfbbc3d7ef74e645f02f878fc8178e8b3971f
3
+ size 11265024
neuronxcc-2.21.18209.0+043b1bf7/MODULE_d4fe40d7566b309ac86a+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_d4fe40d7566b309ac86a+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_d4fe40d7566b309ac86a+a02c3a36/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0dd18126105abb5dddc19543d22a99b11b3ae776516a3845ea8956ffaadb15a0
3
+ size 702870
neuronxcc-2.21.18209.0+043b1bf7/MODULE_d4fe40d7566b309ac86a+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:503663d1e9e83515a143dbe1a2fa6d4151c04250e2d0a53da66ea3d1a2e75847
3
+ size 2059264
neuronxcc-2.21.18209.0+043b1bf7/MODULE_d4fe40d7566b309ac86a+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:848c9cf439fa7e607aff664a598fefc3dee96cfabe42dba3cdfbcb7241202510
3
+ size 2196458