dacorvo HF Staff commited on
Commit
6a37f47
·
verified ·
1 Parent(s): b16c1af

Synchronizing local compiler cache.

Browse files
.gitattributes CHANGED
@@ -12424,3 +12424,6 @@ neuronxcc-2.21.18209.0+043b1bf7/MODULE_a9765a0de4686386ff33+a02c3a36/wrapped_nef
12424
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_73440ea37044d9b55412+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
12425
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_73440ea37044d9b55412+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
12426
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_c0d8c793345bb142d034+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
 
 
 
 
12424
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_73440ea37044d9b55412+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
12425
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_73440ea37044d9b55412+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
12426
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_c0d8c793345bb142d034+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
12427
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_54197861b0e068020a8d+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
12428
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_7c1393a02d9cc9ac0a99+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
12429
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_7c1393a02d9cc9ac0a99+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.1.dev0/phi3/microsoft/phi-4/b860319d54420e1b4506.json ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "ep_degree": 1,
25
+ "fused_qkv": true,
26
+ "glu_mlp": true,
27
+ "local_ranks_size": 10,
28
+ "max_batch_size": 16,
29
+ "max_context_length": 4096,
30
+ "max_topk": 256,
31
+ "n_active_tokens": 4096,
32
+ "neuronxcc_version": "2.21.18209.0+043b1bf7",
33
+ "on_device_sampling": true,
34
+ "optimum_neuron_version": "0.4.1.dev0",
35
+ "output_logits": false,
36
+ "pp_degree": 1,
37
+ "sequence_length": 4096,
38
+ "speculation_length": 0,
39
+ "start_rank_id": 0,
40
+ "target": "trn1",
41
+ "torch_dtype": "bfloat16",
42
+ "tp_degree": 10
43
+ },
44
+ "num_attention_heads": 40,
45
+ "num_hidden_layers": 40,
46
+ "num_key_value_heads": 10,
47
+ "original_max_position_embeddings": 16384,
48
+ "partial_rotary_factor": 1.0,
49
+ "resid_pdrop": 0.0,
50
+ "rms_norm_eps": 1e-05,
51
+ "rope_scaling": null,
52
+ "rope_theta": 250000,
53
+ "sliding_window": null,
54
+ "tie_word_embeddings": false,
55
+ "use_cache": true,
56
+ "vocab_size": 100352
57
+ }
neuronxcc-2.21.18209.0+043b1bf7/MODULE_54197861b0e068020a8d+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-2.21.18209.0+043b1bf7/MODULE_54197861b0e068020a8d+24129607/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_54197861b0e068020a8d+24129607/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:129dc1f0cb03c89b7876e28b739d2c72925c2676d2b457b75902cd7cd2d988ce
3
+ size 1229197
neuronxcc-2.21.18209.0+043b1bf7/MODULE_54197861b0e068020a8d+24129607/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cb5d517808b0e70f9769b5cbe14ec043e182782e9138a3375c9747f39fa08acd
3
+ size 2673664
neuronxcc-2.21.18209.0+043b1bf7/MODULE_7c1393a02d9cc9ac0a99+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-2.21.18209.0+043b1bf7/MODULE_7c1393a02d9cc9ac0a99+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_7c1393a02d9cc9ac0a99+a02c3a36/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_7c1393a02d9cc9ac0a99+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:430ae826d1a278fe7903f55e1185ec7eb56522e0cecdd878cafe3895bf454b33
3
+ size 7384064
neuronxcc-2.21.18209.0+043b1bf7/MODULE_7c1393a02d9cc9ac0a99+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e1b62e092e3e6915e7b253661114cfe907068e467d7fcf572ca223d58bb44476
3
+ size 7569386