Hugging Face's logo

optimum-internal-testing
/
neuron-testing-cache

Model card Files
xet
 Community
dacorvo HF Staff commited on
Commit
f5bc7a7
·
verified ·
1 Parent(s): 1934562

Synchronizing local compiler cache.

Browse files
Files changed (7) hide show
  1. .gitattributes +2 -0
  2. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev1/qwen2/Qwen/Qwen2.5-0.5B/39d0494022f8af95d63a.json +83 -0
  3. neuronxcc-2.21.33363.0+82129205/MODULE_fd476e174a39b55eeb18+a02c3a36/compile_flags.json +1 -0
  4. neuronxcc-2.21.33363.0+82129205/MODULE_fd476e174a39b55eeb18+a02c3a36/model.done +0 -0
  5. neuronxcc-2.21.33363.0+82129205/MODULE_fd476e174a39b55eeb18+a02c3a36/model.hlo_module.pb +3 -0
  6. neuronxcc-2.21.33363.0+82129205/MODULE_fd476e174a39b55eeb18+a02c3a36/model.neff +3 -0
  7. neuronxcc-2.21.33363.0+82129205/MODULE_fd476e174a39b55eeb18+a02c3a36/wrapped_neff.hlo +3 -0
.gitattributes CHANGED
@@ -5811,3 +5811,5 @@ neuronxcc-2.21.33363.0+82129205/MODULE_f9c594cf85ceddc7f3ad+24129607/model.neff
5811
  neuronxcc-2.21.33363.0+82129205/MODULE_60946426f567b03182de+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
5812
  neuronxcc-2.21.33363.0+82129205/MODULE_906d1a9a7651a30426b8+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
5813
  neuronxcc-2.21.33363.0+82129205/MODULE_906d1a9a7651a30426b8+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
5811
  neuronxcc-2.21.33363.0+82129205/MODULE_60946426f567b03182de+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
5812
  neuronxcc-2.21.33363.0+82129205/MODULE_906d1a9a7651a30426b8+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
5813
  neuronxcc-2.21.33363.0+82129205/MODULE_906d1a9a7651a30426b8+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
5814
+ neuronxcc-2.21.33363.0+82129205/MODULE_fd476e174a39b55eeb18+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
5815
+ neuronxcc-2.21.33363.0+82129205/MODULE_fd476e174a39b55eeb18+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev1/qwen2/Qwen/Qwen2.5-0.5B/39d0494022f8af95d63a.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": 4096,
56
+ "max_topk": 256,
57
+ "n_active_tokens": 4096,
58
+ "neuronxcc_version": "2.21.33363.0+82129205",
59
+ "on_device_sampling": false,
60
+ "optimum_neuron_version": "0.4.4.dev1",
61
+ "output_logits": false,
62
+ "pp_degree": 1,
63
+ "sequence_length": 4096,
64
+ "sequence_parallel_enabled": false,
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_fd476e174a39b55eeb18+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_fd476e174a39b55eeb18+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_fd476e174a39b55eeb18+a02c3a36/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3ca0a9c8719d71e823180edfc6eba4a8842e74134383b6713cb51cd6134d5fb7
3
+ size 596952
neuronxcc-2.21.33363.0+82129205/MODULE_fd476e174a39b55eeb18+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e72ea34ac3d917e71710c6a1013878ffd1eebe91a544f4cb6e39b7bdaf7d800b
3
+ size 1936384
neuronxcc-2.21.33363.0+82129205/MODULE_fd476e174a39b55eeb18+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9831cd4d45d394bbf42f0c26cf3bea9f5e3d97db6212e0980cc3f897547307e7
3
+ size 2092575