dacorvo HF Staff commited on
Commit
cd8a8e4
·
verified ·
1 Parent(s): ed853d6

Synchronizing local compiler cache.

Browse files
.gitattributes CHANGED
@@ -11584,3 +11584,6 @@ neuronxcc-2.21.18209.0+043b1bf7/MODULE_ceeba48b4110c6ca7ed9+a9d440f5/wrapped_nef
11584
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_a1150bac79f4debde799+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11585
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_f4334a48bbd53d500c22+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11586
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_f4334a48bbd53d500c22+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
11584
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_a1150bac79f4debde799+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11585
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_f4334a48bbd53d500c22+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11586
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_f4334a48bbd53d500c22+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
11587
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_4f8e0e5a14253ffb6e90+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11588
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_bb13eca870b25e01eeb2+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11589
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_bb13eca870b25e01eeb2+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.0.dev0/phi3/microsoft/phi-4/4465a2fffcc5d9b8d1f6.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "microsoft/phi-4",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Phi3ForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "embd_pdrop": 0.0,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 5120,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 17920,
15
+ "max_position_embeddings": 16384,
16
+ "model_type": "phi3",
17
+ "neuron": {
18
+ "_serialized_key": "NxDNeuronConfig",
19
+ "batch_size": 16,
20
+ "capacity_factor": null,
21
+ "checkpoint_id": "microsoft/phi-4",
22
+ "checkpoint_revision": "187ef0342fff0eb3333be9f00389385e95ef0b61",
23
+ "continuous_batching": true,
24
+ "enable_bucketing": false,
25
+ "ep_degree": 1,
26
+ "fused_qkv": true,
27
+ "glu_mlp": true,
28
+ "local_ranks_size": 10,
29
+ "max_batch_size": 16,
30
+ "max_context_length": 4096,
31
+ "max_topk": 256,
32
+ "n_active_tokens": 4096,
33
+ "neuronxcc_version": "2.21.18209.0+043b1bf7",
34
+ "on_device_sampling": true,
35
+ "optimum_neuron_version": "0.4.0.dev0",
36
+ "output_logits": false,
37
+ "pp_degree": 1,
38
+ "sequence_length": 4096,
39
+ "speculation_length": 0,
40
+ "start_rank_id": 0,
41
+ "target": "trn1",
42
+ "torch_dtype": "bfloat16",
43
+ "tp_degree": 10
44
+ },
45
+ "num_attention_heads": 40,
46
+ "num_hidden_layers": 40,
47
+ "num_key_value_heads": 10,
48
+ "original_max_position_embeddings": 16384,
49
+ "partial_rotary_factor": 1.0,
50
+ "resid_pdrop": 0.0,
51
+ "rms_norm_eps": 1e-05,
52
+ "rope_scaling": null,
53
+ "rope_theta": 250000,
54
+ "sliding_window": null,
55
+ "tie_word_embeddings": false,
56
+ "use_cache": true,
57
+ "vocab_size": 100352
58
+ }
neuronxcc-2.21.18209.0+043b1bf7/MODULE_4f8e0e5a14253ffb6e90+ed72d204/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_model/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.21.18209.0+043b1bf7/MODULE_4f8e0e5a14253ffb6e90+ed72d204/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_4f8e0e5a14253ffb6e90+ed72d204/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:03abd395f93d0627b561d1dfb780363f1e1b99ae77c5cd92dab5143a6d91a6e1
3
+ size 1229197
neuronxcc-2.21.18209.0+043b1bf7/MODULE_4f8e0e5a14253ffb6e90+ed72d204/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5c357975907f81035e4e25374fab22a00443cd46b0527a3c82342b5b3aa9fef9
3
+ size 2673664
neuronxcc-2.21.18209.0+043b1bf7/MODULE_bb13eca870b25e01eeb2+a9d440f5/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_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.21.18209.0+043b1bf7/MODULE_bb13eca870b25e01eeb2+a9d440f5/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_bb13eca870b25e01eeb2+a9d440f5/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ab6efabe8587bf226e2799bbccd757e76760fe7a8e655f72b275ab28a8c23f63
3
+ size 1003448
neuronxcc-2.21.18209.0+043b1bf7/MODULE_bb13eca870b25e01eeb2+a9d440f5/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7d7ec8208e5fe5725431c5506f71beef112eeb756a91b96a9f786783918f1be9
3
+ size 7384064
neuronxcc-2.21.18209.0+043b1bf7/MODULE_bb13eca870b25e01eeb2+a9d440f5/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a1f9e3d7ce437822dcbdb7e4907446a5c5ccabb3c0839210c8e4da78f2d4e49f
3
+ size 7569386