Synchronizing local compiler cache of microsoft/Phi-3-mini-4k-instruct

#8
by badaoui HF Staff - opened
.gitattributes CHANGED
@@ -63,3 +63,7 @@ neuronxcc-2.19.8089.0+8ab9f450/MODULE_6dc18b51a0bd41e6db9b+ed72d204/model.neff f
63
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_bbfac8bf25e0db852ad6+431f5505/model.neff filter=lfs diff=lfs merge=lfs -text
64
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_c6adfb0a32e0702731a6+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
65
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_c6adfb0a32e0702731a6+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
63
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_bbfac8bf25e0db852ad6+431f5505/model.neff filter=lfs diff=lfs merge=lfs -text
64
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_c6adfb0a32e0702731a6+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
65
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_c6adfb0a32e0702731a6+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
66
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_a9b53f328920f947cbb4+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
67
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_f8ebb02d7e7d5ffc1e26+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
68
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_f8ebb02d7e7d5ffc1e26+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
69
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_faa2ec7532645bda3415+431f5505/model.neff filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.0/phi3/microsoft/Phi-3-mini-4k-instruct/1da0c23fe2ea7d4c9e3c.json ADDED
@@ -0,0 +1,77 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "microsoft/Phi-3-mini-4k-instruct",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Phi3ForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "auto_map": {
11
+ "AutoConfig": "microsoft/Phi-3-mini-4k-instruct--configuration_phi3.Phi3Config",
12
+ "AutoModelForCausalLM": "microsoft/Phi-3-mini-4k-instruct--modeling_phi3.Phi3ForCausalLM"
13
+ },
14
+ "embd_pdrop": 0.0,
15
+ "hidden_act": "silu",
16
+ "hidden_size": 3072,
17
+ "initializer_range": 0.02,
18
+ "intermediate_size": 8192,
19
+ "max_position_embeddings": 4096,
20
+ "model_type": "phi3",
21
+ "neuron": {
22
+ "_serialized_key": "NxDNeuronConfig",
23
+ "async_mode": false,
24
+ "attn_kernel_enabled": false,
25
+ "batch_size": 1,
26
+ "capacity_factor": null,
27
+ "cc_pipeline_tiling_factor": 2,
28
+ "checkpoint_id": "microsoft/Phi-3-mini-4k-instruct",
29
+ "checkpoint_revision": "0a67737cc96d2554230f90338b163bc6380a2a85",
30
+ "continuous_batching": false,
31
+ "enable_bucketing": false,
32
+ "ep_degree": 1,
33
+ "flash_decoding_enabled": false,
34
+ "fused_qkv": true,
35
+ "glu_mlp": true,
36
+ "is_chunked_prefill": false,
37
+ "local_ranks_size": 12,
38
+ "logical_nc_config": 1,
39
+ "max_batch_size": 1,
40
+ "max_context_length": 128,
41
+ "max_topk": 256,
42
+ "mlp_kernel_enabled": false,
43
+ "mlp_kernel_fuse_residual_add": false,
44
+ "n_active_tokens": 128,
45
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
46
+ "num_cores_per_group": 1,
47
+ "on_device_sampling": true,
48
+ "optimum_neuron_version": "0.3.0",
49
+ "output_logits": false,
50
+ "padding_side": "right",
51
+ "pp_degree": 1,
52
+ "qk_layernorm": false,
53
+ "qkv_kernel_enabled": false,
54
+ "rpl_reduce_dtype": "bfloat16",
55
+ "sequence_length": 128,
56
+ "sequence_parallel_enabled": false,
57
+ "speculation_length": 0,
58
+ "start_rank_id": 0,
59
+ "target": null,
60
+ "torch_dtype": "bfloat16",
61
+ "tp_degree": 12,
62
+ "vocab_parallel": false
63
+ },
64
+ "num_attention_heads": 32,
65
+ "num_hidden_layers": 32,
66
+ "num_key_value_heads": 32,
67
+ "original_max_position_embeddings": 4096,
68
+ "partial_rotary_factor": 1.0,
69
+ "resid_pdrop": 0.0,
70
+ "rms_norm_eps": 1e-05,
71
+ "rope_scaling": null,
72
+ "rope_theta": 10000.0,
73
+ "sliding_window": 2047,
74
+ "tie_word_embeddings": false,
75
+ "use_cache": true,
76
+ "vocab_size": 32064
77
+ }
neuronxcc-2.19.8089.0+8ab9f450/MODULE_a9b53f328920f947cbb4+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_a9b53f328920f947cbb4+ed72d204/model.done ADDED
File without changes
neuronxcc-2.19.8089.0+8ab9f450/MODULE_a9b53f328920f947cbb4+ed72d204/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f04e03ceb01334236308579fd06341f90f584d95874b0896b983cdc0f2d4717d
3
+ size 645203
neuronxcc-2.19.8089.0+8ab9f450/MODULE_a9b53f328920f947cbb4+ed72d204/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1d8b0c97cf776e0c90e90a659794b8bf04b67abadefd08c855e3b6763b29f727
3
+ size 973824
neuronxcc-2.19.8089.0+8ab9f450/MODULE_f8ebb02d7e7d5ffc1e26+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_f8ebb02d7e7d5ffc1e26+a9d440f5/model.done ADDED
File without changes
neuronxcc-2.19.8089.0+8ab9f450/MODULE_f8ebb02d7e7d5ffc1e26+a9d440f5/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b5030290560ecf77494fdcb00e8e145c9ad60886d33d5816fb141de73d833f4f
3
+ size 696832
neuronxcc-2.19.8089.0+8ab9f450/MODULE_f8ebb02d7e7d5ffc1e26+a9d440f5/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:99fc108ecbe01343be8ad1a7a199afdc98b993710c93dff18f7088a32f810db9
3
+ size 1332224
neuronxcc-2.19.8089.0+8ab9f450/MODULE_f8ebb02d7e7d5ffc1e26+a9d440f5/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7719f0d76343e8a0a78b26b2bece577f3558c2f43a05ac4011a2f02e72178132
3
+ size 1469075
neuronxcc-2.19.8089.0+8ab9f450/MODULE_faa2ec7532645bda3415+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_faa2ec7532645bda3415+431f5505/model.done ADDED
File without changes
neuronxcc-2.19.8089.0+8ab9f450/MODULE_faa2ec7532645bda3415+431f5505/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c56c71cf9341f0e3ad556d73b598fc31832dddb693587637b64d6c1cacfe4418
3
+ size 135302
neuronxcc-2.19.8089.0+8ab9f450/MODULE_faa2ec7532645bda3415+431f5505/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:becf364eb78b142130361f7747cd14997119da1d2bf95b97828903a8de9a63d1
3
+ size 994304