dacorvo HF Staff commited on
Commit
909dfa7
·
verified ·
1 Parent(s): ac5b7fe

Synchronizing local compiler cache.

Browse files
.gitattributes CHANGED
@@ -9464,7 +9464,7 @@ neuronxcc-2.19.8089.0+8ab9f450/MODULE_89d879f69388d23e479e+431f5505/model.neff f
9464
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_db36d0641992b80c1f6e+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
9465
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_f74433a8b539947c0765+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
9466
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_f74433a8b539947c0765+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
9467
- neuronxcc-2.19.8089.0+8ab9f450/MODULE_0ca0c4ac0634e856a577+431f5505/model.neff filter=lfs diff=lfs merge=lfs -text
9468
- neuronxcc-2.19.8089.0+8ab9f450/MODULE_a3f54bd0729e7d8f986e+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
9469
- neuronxcc-2.19.8089.0+8ab9f450/MODULE_a3f54bd0729e7d8f986e+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
9470
- neuronxcc-2.19.8089.0+8ab9f450/MODULE_e10842bf0119ff0ac35c+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
 
9464
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_db36d0641992b80c1f6e+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
9465
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_f74433a8b539947c0765+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
9466
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_f74433a8b539947c0765+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
9467
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_72dd0544832241165775+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
9468
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_72dd0544832241165775+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
9469
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_7794dad30e291e50bd8c+431f5505/model.neff filter=lfs diff=lfs merge=lfs -text
9470
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_b77e1c00470ecdc2be46+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.0.dev5/phi3/microsoft/phi-4/d9f6bd26c449d7c7cbde.json ADDED
@@ -0,0 +1,73 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "async_mode": false,
20
+ "attn_kernel_enabled": false,
21
+ "batch_size": 1,
22
+ "capacity_factor": null,
23
+ "cc_pipeline_tiling_factor": 2,
24
+ "checkpoint_id": "microsoft/phi-4",
25
+ "checkpoint_revision": "187ef0342fff0eb3333be9f00389385e95ef0b61",
26
+ "continuous_batching": false,
27
+ "enable_bucketing": false,
28
+ "ep_degree": 1,
29
+ "flash_decoding_enabled": false,
30
+ "fused_qkv": true,
31
+ "glu_mlp": true,
32
+ "is_chunked_prefill": false,
33
+ "local_ranks_size": 10,
34
+ "logical_nc_config": 1,
35
+ "max_batch_size": 1,
36
+ "max_context_length": 4096,
37
+ "max_topk": 256,
38
+ "mlp_kernel_enabled": false,
39
+ "mlp_kernel_fuse_residual_add": false,
40
+ "n_active_tokens": 4096,
41
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
42
+ "num_cores_per_group": 1,
43
+ "on_device_sampling": true,
44
+ "optimum_neuron_version": "0.3.0.dev5",
45
+ "output_logits": false,
46
+ "padding_side": "right",
47
+ "pp_degree": 1,
48
+ "qk_layernorm": false,
49
+ "qkv_kernel_enabled": false,
50
+ "rpl_reduce_dtype": "bfloat16",
51
+ "sequence_length": 4096,
52
+ "sequence_parallel_enabled": false,
53
+ "speculation_length": 0,
54
+ "start_rank_id": 0,
55
+ "target": null,
56
+ "torch_dtype": "bfloat16",
57
+ "tp_degree": 10,
58
+ "vocab_parallel": false
59
+ },
60
+ "num_attention_heads": 40,
61
+ "num_hidden_layers": 40,
62
+ "num_key_value_heads": 10,
63
+ "original_max_position_embeddings": 16384,
64
+ "partial_rotary_factor": 1.0,
65
+ "resid_pdrop": 0.0,
66
+ "rms_norm_eps": 1e-05,
67
+ "rope_scaling": null,
68
+ "rope_theta": 250000,
69
+ "sliding_window": null,
70
+ "tie_word_embeddings": false,
71
+ "use_cache": true,
72
+ "vocab_size": 100352
73
+ }
neuronxcc-2.19.8089.0+8ab9f450/MODULE_72dd0544832241165775+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-2.19.8089.0+8ab9f450/MODULE_72dd0544832241165775+a9d440f5/model.done ADDED
File without changes
neuronxcc-2.19.8089.0+8ab9f450/MODULE_72dd0544832241165775+a9d440f5/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:392c939ea75252587b2f8eb72734f3521d6274c9f50a3c4cef8f8a48ba7a777e
3
+ size 990986
neuronxcc-2.19.8089.0+8ab9f450/MODULE_72dd0544832241165775+a9d440f5/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a83afde56c33969da7e8e5207a8150658a294e00c198efb7f0586254f396c2a6
3
+ size 2827264
neuronxcc-2.19.8089.0+8ab9f450/MODULE_72dd0544832241165775+a9d440f5/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ce3d6e229c87529f55bb80d5d9a2f50315de2c2077ab419c19ff4adcd8b30d54
3
+ size 3002100
neuronxcc-2.19.8089.0+8ab9f450/MODULE_7794dad30e291e50bd8c+431f5505/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ "--model-type=transformer -O1 --lnc=1 --internal-hlo2tensorizer-options=--experimental-unsafe-fp8e4m3fn-as-fp8e4m3 --logfile=/tmp/nxd_model/layout_opt/log-neuron-cc.txt"
neuronxcc-2.19.8089.0+8ab9f450/MODULE_7794dad30e291e50bd8c+431f5505/model.done ADDED
File without changes
neuronxcc-2.19.8089.0+8ab9f450/MODULE_7794dad30e291e50bd8c+431f5505/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8f6de06f93645c426553d03113f20943f5ff57fc97e62d34ccb006223fc1df21
3
+ size 173762
neuronxcc-2.19.8089.0+8ab9f450/MODULE_7794dad30e291e50bd8c+431f5505/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:345ad6eeccdcfc377f2f85708aaa3a1272bff850b58fc2c3ed164a536feb118f
3
+ size 3195904
neuronxcc-2.19.8089.0+8ab9f450/MODULE_b77e1c00470ecdc2be46+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-2.19.8089.0+8ab9f450/MODULE_b77e1c00470ecdc2be46+ed72d204/model.done ADDED
File without changes
neuronxcc-2.19.8089.0+8ab9f450/MODULE_b77e1c00470ecdc2be46+ed72d204/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:982788e36d5f668898ac6b5f364d0c291a82a6d87dbb3c363daebcd9c62d51ac
3
+ size 1111899
neuronxcc-2.19.8089.0+8ab9f450/MODULE_b77e1c00470ecdc2be46+ed72d204/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8a70d7a8675e89b6ca9c191fa325e8fb121d8d7cc9959bbf33ab34b49cf3c627
3
+ size 2929664