Hugging Face's logo

optimum-internal-testing
/
neuron-testing-cache

Model card Files
xet
 Community
dacorvo HF Staff commited on
Commit
0b378e4
·
verified ·
1 Parent(s): 82b7757

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.dev3/b5b967f1344568be9f11f63a5204f3922ed120215715eeffa2c5fc0a90998346/730f7b70967dbc88e7bc.json +85 -0
  3. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev3/qwen2/Qwen/Qwen2.5-0.5B/730f7b70967dbc88e7bc.json +85 -0
  4. neuronxcc-2.21.33363.0+82129205/MODULE_9efcdde20b63764672c5+24129607/compile_flags.json +1 -0
  5. neuronxcc-2.21.33363.0+82129205/MODULE_9efcdde20b63764672c5+24129607/model.done +0 -0
  6. neuronxcc-2.21.33363.0+82129205/MODULE_9efcdde20b63764672c5+24129607/model.hlo_module.pb +3 -0
  7. neuronxcc-2.21.33363.0+82129205/MODULE_9efcdde20b63764672c5+24129607/model.neff +3 -0
  8. neuronxcc-2.21.33363.0+82129205/MODULE_f76d71fc849f0090bedd+a02c3a36/compile_flags.json +1 -0
  9. neuronxcc-2.21.33363.0+82129205/MODULE_f76d71fc849f0090bedd+a02c3a36/model.done +0 -0
  10. neuronxcc-2.21.33363.0+82129205/MODULE_f76d71fc849f0090bedd+a02c3a36/model.hlo_module.pb +3 -0
  11. neuronxcc-2.21.33363.0+82129205/MODULE_f76d71fc849f0090bedd+a02c3a36/model.neff +3 -0
  12. neuronxcc-2.21.33363.0+82129205/MODULE_f76d71fc849f0090bedd+a02c3a36/wrapped_neff.hlo +3 -0
.gitattributes CHANGED
@@ -7207,3 +7207,6 @@ neuronxcc-2.21.33363.0+82129205/MODULE_d56f37f16ff55f594311+24129607/model.neff
7207
  neuronxcc-2.21.33363.0+82129205/MODULE_b72fc7afaa1831a24a09+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
7208
  neuronxcc-2.21.33363.0+82129205/MODULE_b72fc7afaa1831a24a09+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
7209
  neuronxcc-2.21.33363.0+82129205/MODULE_ef6a56befefaaf69dc23+6170d8e1/model.neff filter=lfs diff=lfs merge=lfs -text
 
 
 
 
7207
  neuronxcc-2.21.33363.0+82129205/MODULE_b72fc7afaa1831a24a09+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
7208
  neuronxcc-2.21.33363.0+82129205/MODULE_b72fc7afaa1831a24a09+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
7209
  neuronxcc-2.21.33363.0+82129205/MODULE_ef6a56befefaaf69dc23+6170d8e1/model.neff filter=lfs diff=lfs merge=lfs -text
7210
+ neuronxcc-2.21.33363.0+82129205/MODULE_9efcdde20b63764672c5+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
7211
+ neuronxcc-2.21.33363.0+82129205/MODULE_f76d71fc849f0090bedd+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
7212
+ neuronxcc-2.21.33363.0+82129205/MODULE_f76d71fc849f0090bedd+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev3/b5b967f1344568be9f11f63a5204f3922ed120215715eeffa2c5fc0a90998346/730f7b70967dbc88e7bc.json ADDED
@@ -0,0 +1,85 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "head_dim": 128,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 896,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 4864,
15
+ "layer_types": [
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
+ "full_attention"
40
+ ],
41
+ "max_position_embeddings": 32768,
42
+ "max_window_layers": 24,
43
+ "model_type": "qwen2",
44
+ "neuron": {
45
+ "_serialized_key": "NxDNeuronConfig",
46
+ "batch_size": 4,
47
+ "capacity_factor": null,
48
+ "checkpoint_id": "Qwen/Qwen2.5-0.5B",
49
+ "checkpoint_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
50
+ "continuous_batching": true,
51
+ "ep_degree": 1,
52
+ "fused_qkv": false,
53
+ "glu_mlp": true,
54
+ "local_ranks_size": 2,
55
+ "max_batch_size": 4,
56
+ "max_context_length": 1024,
57
+ "max_topk": 256,
58
+ "n_active_tokens": 1024,
59
+ "neuronxcc_version": "2.21.33363.0+82129205",
60
+ "on_device_sampling": false,
61
+ "optimum_neuron_version": "0.4.6.dev3",
62
+ "output_logits": false,
63
+ "pp_degree": 1,
64
+ "prefill_chunk_size": 0,
65
+ "sequence_length": 1024,
66
+ "speculation_length": 0,
67
+ "start_rank_id": 0,
68
+ "target": "trn1",
69
+ "torch_dtype": "bfloat16",
70
+ "tp_degree": 2
71
+ },
72
+ "num_attention_heads": 14,
73
+ "num_hidden_layers": 24,
74
+ "num_key_value_heads": 2,
75
+ "original_head_dim": 64,
76
+ "rms_norm_eps": 1e-06,
77
+ "rope_scaling": null,
78
+ "rope_theta": 1000000.0,
79
+ "sliding_window": null,
80
+ "tie_word_embeddings": true,
81
+ "use_cache": true,
82
+ "use_mrope": false,
83
+ "use_sliding_window": false,
84
+ "vocab_size": 151936
85
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev3/qwen2/Qwen/Qwen2.5-0.5B/730f7b70967dbc88e7bc.json ADDED
@@ -0,0 +1,85 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "head_dim": 128,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 896,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 4864,
15
+ "layer_types": [
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
+ "full_attention"
40
+ ],
41
+ "max_position_embeddings": 32768,
42
+ "max_window_layers": 24,
43
+ "model_type": "qwen2",
44
+ "neuron": {
45
+ "_serialized_key": "NxDNeuronConfig",
46
+ "batch_size": 4,
47
+ "capacity_factor": null,
48
+ "checkpoint_id": "Qwen/Qwen2.5-0.5B",
49
+ "checkpoint_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
50
+ "continuous_batching": true,
51
+ "ep_degree": 1,
52
+ "fused_qkv": false,
53
+ "glu_mlp": true,
54
+ "local_ranks_size": 2,
55
+ "max_batch_size": 4,
56
+ "max_context_length": 1024,
57
+ "max_topk": 256,
58
+ "n_active_tokens": 1024,
59
+ "neuronxcc_version": "2.21.33363.0+82129205",
60
+ "on_device_sampling": false,
61
+ "optimum_neuron_version": "0.4.6.dev3",
62
+ "output_logits": false,
63
+ "pp_degree": 1,
64
+ "prefill_chunk_size": 0,
65
+ "sequence_length": 1024,
66
+ "speculation_length": 0,
67
+ "start_rank_id": 0,
68
+ "target": "trn1",
69
+ "torch_dtype": "bfloat16",
70
+ "tp_degree": 2
71
+ },
72
+ "num_attention_heads": 14,
73
+ "num_hidden_layers": 24,
74
+ "num_key_value_heads": 2,
75
+ "original_head_dim": 64,
76
+ "rms_norm_eps": 1e-06,
77
+ "rope_scaling": null,
78
+ "rope_theta": 1000000.0,
79
+ "sliding_window": null,
80
+ "tie_word_embeddings": true,
81
+ "use_cache": true,
82
+ "use_mrope": false,
83
+ "use_sliding_window": false,
84
+ "vocab_size": 151936
85
+ }
neuronxcc-2.21.33363.0+82129205/MODULE_9efcdde20b63764672c5+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_9efcdde20b63764672c5+24129607/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_9efcdde20b63764672c5+24129607/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ebbe4626004fa10d6de1f06dc8d2e7c2b20ba375db45da9b3d52347af3436a18
3
+ size 626700
neuronxcc-2.21.33363.0+82129205/MODULE_9efcdde20b63764672c5+24129607/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0424d734825f1448b117fd4a8aa2b6ea297c1fb6df5f2d27b2f9395967770a7b
3
+ size 3738624
neuronxcc-2.21.33363.0+82129205/MODULE_f76d71fc849f0090bedd+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_f76d71fc849f0090bedd+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_f76d71fc849f0090bedd+a02c3a36/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5ced8c217c6169c1356494d79c639b30d9d2ec7b84b936225053548165d02ec1
3
+ size 647980
neuronxcc-2.21.33363.0+82129205/MODULE_f76d71fc849f0090bedd+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4c4bca76936a8731934fc7c9f83797809507aa82e5105ed003f9679c7289854e
3
+ size 1752064
neuronxcc-2.21.33363.0+82129205/MODULE_f76d71fc849f0090bedd+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a4a1a9ce12c55205617d73c0834e4a768a00fe041f8dc6a511e5b6556ac9c01b
3
+ size 1909815