Hugging Face's logo

optimum-internal-testing
/
neuron-testing-cache

Model card Files
xet
 Community
tengomucho HF Staff commited on
Commit
0b1890c
·
verified ·
1 Parent(s): b514470

Synchronizing local compiler cache.

Browse files
Files changed (11) hide show
  1. .gitattributes +3 -0
  2. neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/4cb7aff9e2a15c151396f2b684013e39d6739f0dec83e5c9dabbfe9d5fcf77b7/196ef14bd9a3e0d3035c.json +83 -0
  3. neuronxcc-2.26.6360.0+6f180f47/MODULE_7b5dd0edc4e64fd3bf6d+12dd9119/compile_flags.json +1 -0
  4. neuronxcc-2.26.6360.0+6f180f47/MODULE_7b5dd0edc4e64fd3bf6d+12dd9119/model.done +0 -0
  5. neuronxcc-2.26.6360.0+6f180f47/MODULE_7b5dd0edc4e64fd3bf6d+12dd9119/model.hlo_module.pb +3 -0
  6. neuronxcc-2.26.6360.0+6f180f47/MODULE_7b5dd0edc4e64fd3bf6d+12dd9119/model.neff +3 -0
  7. neuronxcc-2.26.6360.0+6f180f47/MODULE_ea27c4a705f821f7e05c+84337dd9/compile_flags.json +1 -0
  8. neuronxcc-2.26.6360.0+6f180f47/MODULE_ea27c4a705f821f7e05c+84337dd9/model.done +0 -0
  9. neuronxcc-2.26.6360.0+6f180f47/MODULE_ea27c4a705f821f7e05c+84337dd9/model.hlo_module.pb +3 -0
  10. neuronxcc-2.26.6360.0+6f180f47/MODULE_ea27c4a705f821f7e05c+84337dd9/model.neff +3 -0
  11. neuronxcc-2.26.6360.0+6f180f47/MODULE_ea27c4a705f821f7e05c+84337dd9/wrapped_neff.hlo +3 -0
.gitattributes CHANGED
@@ -7949,3 +7949,6 @@ neuronxcc-2.26.6360.0+6f180f47/MODULE_f8c69d6a263539b4580d+84337dd9/wrapped_neff
7949
  neuronxcc-2.26.6360.0+6f180f47/MODULE_424317c91cab55cd13d0+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
7950
  neuronxcc-2.26.6360.0+6f180f47/MODULE_66f5037492c24f9dd1b9+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
7951
  neuronxcc-2.26.6360.0+6f180f47/MODULE_66f5037492c24f9dd1b9+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
7949
  neuronxcc-2.26.6360.0+6f180f47/MODULE_424317c91cab55cd13d0+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
7950
  neuronxcc-2.26.6360.0+6f180f47/MODULE_66f5037492c24f9dd1b9+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
7951
  neuronxcc-2.26.6360.0+6f180f47/MODULE_66f5037492c24f9dd1b9+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
7952
+ neuronxcc-2.26.6360.0+6f180f47/MODULE_7b5dd0edc4e64fd3bf6d+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
7953
+ neuronxcc-2.26.6360.0+6f180f47/MODULE_ea27c4a705f821f7e05c+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
7954
+ neuronxcc-2.26.6360.0+6f180f47/MODULE_ea27c4a705f821f7e05c+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/4cb7aff9e2a15c151396f2b684013e39d6739f0dec83e5c9dabbfe9d5fcf77b7/196ef14bd9a3e0d3035c.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": 1,
46
+ "capacity_factor": null,
47
+ "checkpoint_id": "Qwen/Qwen2.5-0.5B",
48
+ "checkpoint_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
49
+ "continuous_batching": false,
50
+ "ep_degree": 1,
51
+ "fused_qkv": false,
52
+ "glu_mlp": true,
53
+ "local_ranks_size": 2,
54
+ "max_batch_size": 1,
55
+ "max_context_length": 8192,
56
+ "max_topk": 256,
57
+ "n_active_tokens": 8192,
58
+ "neuronxcc_version": "2.26.6360.0+6f180f47",
59
+ "on_device_sampling": true,
60
+ "optimum_neuron_version": "0.4.7.dev0",
61
+ "output_logits": false,
62
+ "pp_degree": 1,
63
+ "prefill_chunk_size": 1024,
64
+ "sequence_length": 8192,
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-2.26.6360.0+6f180f47/MODULE_7b5dd0edc4e64fd3bf6d+12dd9119/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", "--verbose=35", "--logfile=/tmp/nxd_model/chunked_prefill/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.26.6360.0+6f180f47/MODULE_7b5dd0edc4e64fd3bf6d+12dd9119/model.done ADDED
File without changes
neuronxcc-2.26.6360.0+6f180f47/MODULE_7b5dd0edc4e64fd3bf6d+12dd9119/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6d631860beebc21f121b53d2ba5c05b9cd0491830485941f2eb27d4ff33e67ad
3
+ size 2910636
neuronxcc-2.26.6360.0+6f180f47/MODULE_7b5dd0edc4e64fd3bf6d+12dd9119/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4fe3315e8e0f130b09ef4320cd09a1cf0a23dc93ef876f123c40ed04c97a61eb
3
+ size 15350784
neuronxcc-2.26.6360.0+6f180f47/MODULE_ea27c4a705f821f7e05c+84337dd9/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", "--verbose=35", "--logfile=/tmp/nxd_model/token_generation/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.26.6360.0+6f180f47/MODULE_ea27c4a705f821f7e05c+84337dd9/model.done ADDED
File without changes
neuronxcc-2.26.6360.0+6f180f47/MODULE_ea27c4a705f821f7e05c+84337dd9/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2083c7fcb588165b88cc6468624745d18861045c0d551cad18d6311e56a29bcc
3
+ size 2189940
neuronxcc-2.26.6360.0+6f180f47/MODULE_ea27c4a705f821f7e05c+84337dd9/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:805a6602427ab16981f3113b72fbf03d958eaad462dc0cbd85395ba28b31f6c5
3
+ size 1690624
neuronxcc-2.26.6360.0+6f180f47/MODULE_ea27c4a705f821f7e05c+84337dd9/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d2008234c39eeeab051793895549ccd9a132f059147037b8e5269c7d82a57151
3
+ size 1846930