dacorvo HF Staff commited on
Commit
4f5ed72
·
verified ·
1 Parent(s): 6289468

Synchronizing local compiler cache.

Browse files
.gitattributes CHANGED
@@ -17810,3 +17810,6 @@ neuronxcc-2.26.6360.0+6f180f47/MODULE_628c8002276b5a6987cf+12dd9119/model.neff f
17810
  neuronxcc-2.26.6360.0+6f180f47/MODULE_10fe634e4c07aa0d76f9+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
17811
  neuronxcc-2.26.6360.0+6f180f47/MODULE_10fe634e4c07aa0d76f9+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
17812
  neuronxcc-2.26.6360.0+6f180f47/MODULE_cf54d7d60d523a1dc614+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
 
 
 
 
17810
  neuronxcc-2.26.6360.0+6f180f47/MODULE_10fe634e4c07aa0d76f9+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
17811
  neuronxcc-2.26.6360.0+6f180f47/MODULE_10fe634e4c07aa0d76f9+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
17812
  neuronxcc-2.26.6360.0+6f180f47/MODULE_cf54d7d60d523a1dc614+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
17813
+ neuronxcc-2.26.6360.0+6f180f47/MODULE_40935637ee993b1f4cd0+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
17814
+ neuronxcc-2.26.6360.0+6f180f47/MODULE_82a0ae250162a817ed30+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
17815
+ neuronxcc-2.26.6360.0+6f180f47/MODULE_82a0ae250162a817ed30+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/83b64303132921a55cb0a422f2894a8b4d9c6143ff8379ddb5933ecd8a35fb8d/60e3bee13e1e519658ae.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
+ "dtype": "bfloat16",
11
+ "embd_pdrop": 0.0,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 5120,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 17920,
16
+ "max_position_embeddings": 16384,
17
+ "model_type": "phi3",
18
+ "neuron": {
19
+ "_serialized_key": "NxDNeuronConfig",
20
+ "batch_size": 16,
21
+ "capacity_factor": null,
22
+ "checkpoint_id": "microsoft/phi-4",
23
+ "checkpoint_revision": "2db69c1c3e91a05d2c64a3185acfbaf36f744e25",
24
+ "continuous_batching": true,
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": "2.26.6360.0+6f180f47",
34
+ "on_device_sampling": true,
35
+ "optimum_neuron_version": "0.4.7.dev0",
36
+ "output_logits": false,
37
+ "pp_degree": 1,
38
+ "prefill_chunk_size": 1024,
39
+ "sequence_length": 4096,
40
+ "speculation_length": 0,
41
+ "start_rank_id": 0,
42
+ "target": "trn1",
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-2.26.6360.0+6f180f47/MODULE_40935637ee993b1f4cd0+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_40935637ee993b1f4cd0+12dd9119/model.done ADDED
File without changes
neuronxcc-2.26.6360.0+6f180f47/MODULE_40935637ee993b1f4cd0+12dd9119/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9bb77455608c6a56888b660f0c1f95e22bad203280dc06dc1f41b361303603b8
3
+ size 5964938
neuronxcc-2.26.6360.0+6f180f47/MODULE_40935637ee993b1f4cd0+12dd9119/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:74a651af481bdf1ec7166f206e89c7e1ee285f338b451f1310d4e552cabd3d88
3
+ size 5336064
neuronxcc-2.26.6360.0+6f180f47/MODULE_82a0ae250162a817ed30+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_82a0ae250162a817ed30+84337dd9/model.done ADDED
File without changes
neuronxcc-2.26.6360.0+6f180f47/MODULE_82a0ae250162a817ed30+84337dd9/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c22b57d1b20a3c06eeb5b3169873fc4f91a1b28b48656ab30f5d52f916ea8a7c
3
+ size 4093747
neuronxcc-2.26.6360.0+6f180f47/MODULE_82a0ae250162a817ed30+84337dd9/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:88203227613710a017480cf6ec0e3ced4e5988da12d685667ccc0955f7400f77
3
+ size 6718464
neuronxcc-2.26.6360.0+6f180f47/MODULE_82a0ae250162a817ed30+84337dd9/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3a341bd9378c2b58e8c783a609b63716896f92e177367ec8165078a5b9d97359
3
+ size 6903643