dacorvo HF Staff commited on
Commit
901906a
·
verified ·
1 Parent(s): f42d8b2

Synchronizing local compiler cache.

Browse files
.gitattributes CHANGED
@@ -17795,3 +17795,6 @@ neuronxcc-2.26.6360.0+6f180f47/MODULE_5e4597bdb685b3f0017d+8f3c2f7c/wrapped_neff
17795
  neuronxcc-2.26.6360.0+6f180f47/MODULE_1f54590e695b4002553c+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
17796
  neuronxcc-2.26.6360.0+6f180f47/MODULE_f75715e2ead36890b407+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
17797
  neuronxcc-2.26.6360.0+6f180f47/MODULE_f75715e2ead36890b407+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
17795
  neuronxcc-2.26.6360.0+6f180f47/MODULE_1f54590e695b4002553c+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
17796
  neuronxcc-2.26.6360.0+6f180f47/MODULE_f75715e2ead36890b407+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
17797
  neuronxcc-2.26.6360.0+6f180f47/MODULE_f75715e2ead36890b407+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
17798
+ neuronxcc-2.26.6360.0+6f180f47/MODULE_abc7f6e1bd3fc58a04bf+12dd9119/model.neff filter=lfs diff=lfs merge=lfs -text
17799
+ neuronxcc-2.26.6360.0+6f180f47/MODULE_cfc98d6eaa0cefa6f7ca+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
17800
+ neuronxcc-2.26.6360.0+6f180f47/MODULE_cfc98d6eaa0cefa6f7ca+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/8f68cf9739bc231106be6b50c08f4f0965334ec793da7c9e467444105d80178b/3843ce73852f189eb6de.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "dtype": "bfloat16",
15
+ "embd_pdrop": 0.0,
16
+ "hidden_act": "silu",
17
+ "hidden_size": 3072,
18
+ "initializer_range": 0.02,
19
+ "intermediate_size": 8192,
20
+ "max_position_embeddings": 4096,
21
+ "model_type": "phi3",
22
+ "neuron": {
23
+ "_serialized_key": "NxDNeuronConfig",
24
+ "batch_size": 4,
25
+ "capacity_factor": null,
26
+ "checkpoint_id": "microsoft/Phi-3-mini-4k-instruct",
27
+ "checkpoint_revision": "f39ac1d28e925b323eae81227eaba4464caced4e",
28
+ "continuous_batching": true,
29
+ "ep_degree": 1,
30
+ "fused_qkv": true,
31
+ "glu_mlp": true,
32
+ "local_ranks_size": 2,
33
+ "max_batch_size": 4,
34
+ "max_context_length": 4096,
35
+ "max_topk": 256,
36
+ "n_active_tokens": 4096,
37
+ "neuronxcc_version": "2.26.6360.0+6f180f47",
38
+ "on_device_sampling": true,
39
+ "optimum_neuron_version": "0.4.7.dev0",
40
+ "output_logits": false,
41
+ "pp_degree": 1,
42
+ "prefill_chunk_size": 1024,
43
+ "sequence_length": 4096,
44
+ "speculation_length": 0,
45
+ "start_rank_id": 0,
46
+ "target": "trn1",
47
+ "torch_dtype": "bfloat16",
48
+ "tp_degree": 2
49
+ },
50
+ "num_attention_heads": 32,
51
+ "num_hidden_layers": 32,
52
+ "num_key_value_heads": 32,
53
+ "original_max_position_embeddings": 4096,
54
+ "partial_rotary_factor": 1.0,
55
+ "resid_pdrop": 0.0,
56
+ "rms_norm_eps": 1e-05,
57
+ "rope_scaling": null,
58
+ "rope_theta": 10000.0,
59
+ "sliding_window": 2047,
60
+ "tie_word_embeddings": false,
61
+ "use_cache": true,
62
+ "vocab_size": 32064
63
+ }
neuronxcc-2.26.6360.0+6f180f47/MODULE_abc7f6e1bd3fc58a04bf+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_abc7f6e1bd3fc58a04bf+12dd9119/model.done ADDED
File without changes
neuronxcc-2.26.6360.0+6f180f47/MODULE_abc7f6e1bd3fc58a04bf+12dd9119/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7c3dd8463f3731b47145e6b45c0ea6601a8d043aa4d559beb29f8b896a0c55d3
3
+ size 4322119
neuronxcc-2.26.6360.0+6f180f47/MODULE_abc7f6e1bd3fc58a04bf+12dd9119/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:57753da0af150619d588c58213456e955fbfa1005c6342a1ee662672261b5c8e
3
+ size 4373504
neuronxcc-2.26.6360.0+6f180f47/MODULE_cfc98d6eaa0cefa6f7ca+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_cfc98d6eaa0cefa6f7ca+84337dd9/model.done ADDED
File without changes
neuronxcc-2.26.6360.0+6f180f47/MODULE_cfc98d6eaa0cefa6f7ca+84337dd9/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7dce43c24ef3968c14de42c981ccf1a5d7c15a2465334f0c0e569229f25ddbfc
3
+ size 3035405
neuronxcc-2.26.6360.0+6f180f47/MODULE_cfc98d6eaa0cefa6f7ca+84337dd9/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7a61c58b0da0997defe99f246bd27a491869d9a5c7064f50e12c6a883663f537
3
+ size 11817984
neuronxcc-2.26.6360.0+6f180f47/MODULE_cfc98d6eaa0cefa6f7ca+84337dd9/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e8ced1e50efa018e80e8a8c42a8b83a37b787cff3346764775af4e9c140830a5
3
+ size 11965000