dacorvo HF Staff commited on
Commit
7a45392
·
verified ·
1 Parent(s): e39d4b5

Synchronizing local compiler cache.

Browse files
.gitattributes CHANGED
@@ -11329,3 +11329,6 @@ neuronxcc-2.21.18209.0+043b1bf7/MODULE_ab70d82d52c94689be67+a9d440f5/wrapped_nef
11329
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_77c5ff5d456ecdab1250+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11330
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_d733b1ef9c8b8a12c314+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11331
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_d733b1ef9c8b8a12c314+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
11329
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_77c5ff5d456ecdab1250+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11330
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_d733b1ef9c8b8a12c314+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11331
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_d733b1ef9c8b8a12c314+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
11332
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_2bdeadd87b5eb53474b6+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11333
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_2bdeadd87b5eb53474b6+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
11334
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_8acc9a6488836e96530c+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.3.1.dev7/phi3/microsoft/phi-4/fbfff05a0dd32cc747d3.json ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": 1,
20
+ "capacity_factor": null,
21
+ "checkpoint_id": "microsoft/phi-4",
22
+ "checkpoint_revision": "187ef0342fff0eb3333be9f00389385e95ef0b61",
23
+ "continuous_batching": false,
24
+ "enable_bucketing": false,
25
+ "ep_degree": 1,
26
+ "fused_qkv": true,
27
+ "glu_mlp": true,
28
+ "local_ranks_size": 10,
29
+ "logical_nc_config": 1,
30
+ "max_batch_size": 1,
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": "0.3.1.dev7",
37
+ "output_logits": false,
38
+ "pp_degree": 1,
39
+ "sequence_length": 4096,
40
+ "speculation_length": 0,
41
+ "start_rank_id": 0,
42
+ "target": null,
43
+ "torch_dtype": "bfloat16",
44
+ "tp_degree": 10
45
+ },
46
+ "num_attention_heads": 40,
47
+ "num_hidden_layers": 40,
48
+ "num_key_value_heads": 10,
49
+ "original_max_position_embeddings": 16384,
50
+ "partial_rotary_factor": 1.0,
51
+ "resid_pdrop": 0.0,
52
+ "rms_norm_eps": 1e-05,
53
+ "rope_scaling": null,
54
+ "rope_theta": 250000,
55
+ "sliding_window": null,
56
+ "tie_word_embeddings": false,
57
+ "use_cache": true,
58
+ "vocab_size": 100352
59
+ }
neuronxcc-2.21.18209.0+043b1bf7/MODULE_2bdeadd87b5eb53474b6+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_2bdeadd87b5eb53474b6+a9d440f5/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_2bdeadd87b5eb53474b6+a9d440f5/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:997fce42d5fb854e9a2272d02f75583315792f09ae6cc56a044ddaebffd39bb6
3
+ size 1002831
neuronxcc-2.21.18209.0+043b1bf7/MODULE_2bdeadd87b5eb53474b6+a9d440f5/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bf09f8f750232e0a44e8b4a6eae76362092ce2961d0e62dce7886eecac7b6886
3
+ size 3810304
neuronxcc-2.21.18209.0+043b1bf7/MODULE_2bdeadd87b5eb53474b6+a9d440f5/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cd07df328a73e8167b3f05713cc18f54ebf9dd6c572a58ecb4477d942c827f7c
3
+ size 3995906
neuronxcc-2.21.18209.0+043b1bf7/MODULE_8acc9a6488836e96530c+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_8acc9a6488836e96530c+ed72d204/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_8acc9a6488836e96530c+ed72d204/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:837416c27dce4ad0bcd42782fba98fa3f6217454b0cd5663db42649337f5e062
3
+ size 1141233
neuronxcc-2.21.18209.0+043b1bf7/MODULE_8acc9a6488836e96530c+ed72d204/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:de332ed97c7488d101d2424ca87ef18cd051bb865e0a2dbbed0906938422bee5
3
+ size 2632704