dacorvo HF Staff commited on
Commit
f931939
·
verified ·
1 Parent(s): 16f89f7

Synchronizing local compiler cache.

Browse files
.gitattributes CHANGED
@@ -11551,3 +11551,6 @@ neuronxcc-2.21.18209.0+043b1bf7/MODULE_7c86760b1ab3a4639e3f+ed72d204/model.neff
11551
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_14497e2829240163f2f3+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11552
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_949c917c789ef5ab7819+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11553
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_949c917c789ef5ab7819+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
11551
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_14497e2829240163f2f3+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11552
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_949c917c789ef5ab7819+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11553
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_949c917c789ef5ab7819+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
11554
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_04a651791e36c2cfd4d1+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11555
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_04a651791e36c2cfd4d1+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
11556
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_8c48d2b8e39a5ae21620+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.0.dev0/phi3/microsoft/Phi-3-mini-4k-instruct/f47ac1f3702213f7b53f.json ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "configuration_phi3.Phi3Config",
12
+ "AutoModelForCausalLM": "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
+ "batch_size": 1,
24
+ "capacity_factor": null,
25
+ "checkpoint_id": "microsoft/Phi-3-mini-4k-instruct",
26
+ "checkpoint_revision": "0a67737cc96d2554230f90338b163bc6380a2a85",
27
+ "continuous_batching": false,
28
+ "enable_bucketing": false,
29
+ "ep_degree": 1,
30
+ "fused_qkv": true,
31
+ "glu_mlp": true,
32
+ "local_ranks_size": 2,
33
+ "max_batch_size": 1,
34
+ "max_context_length": 4096,
35
+ "max_topk": 256,
36
+ "n_active_tokens": 4096,
37
+ "neuronxcc_version": "2.21.18209.0+043b1bf7",
38
+ "on_device_sampling": true,
39
+ "optimum_neuron_version": "0.4.0.dev0",
40
+ "output_logits": false,
41
+ "pp_degree": 1,
42
+ "sequence_length": 4096,
43
+ "speculation_length": 0,
44
+ "start_rank_id": 0,
45
+ "target": "trn1",
46
+ "torch_dtype": "bfloat16",
47
+ "tp_degree": 2
48
+ },
49
+ "num_attention_heads": 32,
50
+ "num_hidden_layers": 32,
51
+ "num_key_value_heads": 32,
52
+ "original_max_position_embeddings": 4096,
53
+ "partial_rotary_factor": 1.0,
54
+ "resid_pdrop": 0.0,
55
+ "rms_norm_eps": 1e-05,
56
+ "rope_scaling": null,
57
+ "rope_theta": 10000.0,
58
+ "sliding_window": 2047,
59
+ "tie_word_embeddings": false,
60
+ "use_cache": true,
61
+ "vocab_size": 32064
62
+ }
neuronxcc-2.21.18209.0+043b1bf7/MODULE_04a651791e36c2cfd4d1+a9d440f5/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_04a651791e36c2cfd4d1+a9d440f5/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5fae139ec776f5d3046c6b588c29aa51cb82081f0b42a0a5ba65d846e0354248
3
+ size 7281664
neuronxcc-2.21.18209.0+043b1bf7/MODULE_04a651791e36c2cfd4d1+a9d440f5/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0a8b1afeaf36e7d4a3b6ff70729cfeec1599352abcc134e062c45b0c7b50782d
3
+ size 7428823
neuronxcc-2.21.18209.0+043b1bf7/MODULE_8c48d2b8e39a5ae21620+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_8c48d2b8e39a5ae21620+ed72d204/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_8c48d2b8e39a5ae21620+ed72d204/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ee0ea1bda6910a7a1aad626c7545e0dbbbd770fffe67424910c39b21c8db8a5f
3
+ size 853602
neuronxcc-2.21.18209.0+043b1bf7/MODULE_8c48d2b8e39a5ae21620+ed72d204/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ae324e57ec6455aaf3ce084783202a5148ec19950ab99c789647a0d112873143
3
+ size 5028864