Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
 Community
662
dacorvo HF Staff commited on
Commit
0218751
·
verified ·
1 Parent(s): e93e1bc

Synchronizing local compiler cache.

Browse files
Files changed (12) hide show
  1. .gitattributes +3 -0
  2. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev2/081b62fbc286ccc2ed90270d25c62dfa498c91de399d534fd84b3643c32d6f2d/8381b4eabe09e839d5d5.json +56 -0
  3. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev2/llama/meta-llama/Llama-2-13b-hf/8381b4eabe09e839d5d5.json +56 -0
  4. neuronxcc-2.21.33363.0+82129205/MODULE_743ec6d0e2a18aabb72c+24129607/compile_flags.json +1 -0
  5. neuronxcc-2.21.33363.0+82129205/MODULE_743ec6d0e2a18aabb72c+24129607/model.done +0 -0
  6. neuronxcc-2.21.33363.0+82129205/MODULE_743ec6d0e2a18aabb72c+24129607/model.hlo_module.pb +3 -0
  7. neuronxcc-2.21.33363.0+82129205/MODULE_743ec6d0e2a18aabb72c+24129607/model.neff +3 -0
  8. neuronxcc-2.21.33363.0+82129205/MODULE_ad3140bf704a4662dc10+a02c3a36/compile_flags.json +1 -0
  9. neuronxcc-2.21.33363.0+82129205/MODULE_ad3140bf704a4662dc10+a02c3a36/model.done +0 -0
  10. neuronxcc-2.21.33363.0+82129205/MODULE_ad3140bf704a4662dc10+a02c3a36/model.hlo_module.pb +3 -0
  11. neuronxcc-2.21.33363.0+82129205/MODULE_ad3140bf704a4662dc10+a02c3a36/model.neff +3 -0
  12. neuronxcc-2.21.33363.0+82129205/MODULE_ad3140bf704a4662dc10+a02c3a36/wrapped_neff.hlo +3 -0
.gitattributes CHANGED
@@ -16994,3 +16994,6 @@ neuronxcc-2.21.33363.0+82129205/MODULE_6df3765fc0718b9095f7+afeb10ce/model.neff
16994
  neuronxcc-2.21.33363.0+82129205/MODULE_6df3765fc0718b9095f7+afeb10ce/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
16995
  neuronxcc-2.21.33363.0+82129205/MODULE_5b87bb35c84bca1759ee+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
16996
  neuronxcc-2.21.33363.0+82129205/MODULE_5b87bb35c84bca1759ee+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
16994
  neuronxcc-2.21.33363.0+82129205/MODULE_6df3765fc0718b9095f7+afeb10ce/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
16995
  neuronxcc-2.21.33363.0+82129205/MODULE_5b87bb35c84bca1759ee+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
16996
  neuronxcc-2.21.33363.0+82129205/MODULE_5b87bb35c84bca1759ee+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
16997
+ neuronxcc-2.21.33363.0+82129205/MODULE_743ec6d0e2a18aabb72c+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
16998
+ neuronxcc-2.21.33363.0+82129205/MODULE_ad3140bf704a4662dc10+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
16999
+ neuronxcc-2.21.33363.0+82129205/MODULE_ad3140bf704a4662dc10+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev2/081b62fbc286ccc2ed90270d25c62dfa498c91de399d534fd84b3643c32d6f2d/8381b4eabe09e839d5d5.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "meta-llama/Llama-2-13b-hf",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "LlamaForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "dtype": "float16",
11
+ "head_dim": 128,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 5120,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 13824,
16
+ "max_position_embeddings": 4096,
17
+ "mlp_bias": false,
18
+ "model_type": "llama",
19
+ "neuron": {
20
+ "_serialized_key": "NxDNeuronConfig",
21
+ "batch_size": 1,
22
+ "capacity_factor": null,
23
+ "checkpoint_id": "meta-llama/Llama-2-13b-hf",
24
+ "checkpoint_revision": "5c31dfb671ce7cfe2d7bb7c04375e44c55e815b1",
25
+ "continuous_batching": false,
26
+ "ep_degree": 1,
27
+ "fused_qkv": true,
28
+ "glu_mlp": true,
29
+ "local_ranks_size": 8,
30
+ "max_batch_size": 1,
31
+ "max_context_length": 2048,
32
+ "max_topk": 256,
33
+ "n_active_tokens": 2048,
34
+ "neuronxcc_version": "2.21.33363.0+82129205",
35
+ "on_device_sampling": true,
36
+ "optimum_neuron_version": "0.4.6.dev2",
37
+ "output_logits": false,
38
+ "pp_degree": 1,
39
+ "sequence_length": 2048,
40
+ "speculation_length": 0,
41
+ "start_rank_id": 0,
42
+ "target": "trn1",
43
+ "torch_dtype": "float16",
44
+ "tp_degree": 8
45
+ },
46
+ "num_attention_heads": 40,
47
+ "num_hidden_layers": 40,
48
+ "num_key_value_heads": 40,
49
+ "pretraining_tp": 1,
50
+ "rms_norm_eps": 1e-05,
51
+ "rope_scaling": null,
52
+ "rope_theta": 10000.0,
53
+ "tie_word_embeddings": false,
54
+ "use_cache": true,
55
+ "vocab_size": 32000
56
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev2/llama/meta-llama/Llama-2-13b-hf/8381b4eabe09e839d5d5.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "meta-llama/Llama-2-13b-hf",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "LlamaForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "dtype": "float16",
11
+ "head_dim": 128,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 5120,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 13824,
16
+ "max_position_embeddings": 4096,
17
+ "mlp_bias": false,
18
+ "model_type": "llama",
19
+ "neuron": {
20
+ "_serialized_key": "NxDNeuronConfig",
21
+ "batch_size": 1,
22
+ "capacity_factor": null,
23
+ "checkpoint_id": "meta-llama/Llama-2-13b-hf",
24
+ "checkpoint_revision": "5c31dfb671ce7cfe2d7bb7c04375e44c55e815b1",
25
+ "continuous_batching": false,
26
+ "ep_degree": 1,
27
+ "fused_qkv": true,
28
+ "glu_mlp": true,
29
+ "local_ranks_size": 8,
30
+ "max_batch_size": 1,
31
+ "max_context_length": 2048,
32
+ "max_topk": 256,
33
+ "n_active_tokens": 2048,
34
+ "neuronxcc_version": "2.21.33363.0+82129205",
35
+ "on_device_sampling": true,
36
+ "optimum_neuron_version": "0.4.6.dev2",
37
+ "output_logits": false,
38
+ "pp_degree": 1,
39
+ "sequence_length": 2048,
40
+ "speculation_length": 0,
41
+ "start_rank_id": 0,
42
+ "target": "trn1",
43
+ "torch_dtype": "float16",
44
+ "tp_degree": 8
45
+ },
46
+ "num_attention_heads": 40,
47
+ "num_hidden_layers": 40,
48
+ "num_key_value_heads": 40,
49
+ "pretraining_tp": 1,
50
+ "rms_norm_eps": 1e-05,
51
+ "rope_scaling": null,
52
+ "rope_theta": 10000.0,
53
+ "tie_word_embeddings": false,
54
+ "use_cache": true,
55
+ "vocab_size": 32000
56
+ }
neuronxcc-2.21.33363.0+82129205/MODULE_743ec6d0e2a18aabb72c+24129607/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/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.21.33363.0+82129205/MODULE_743ec6d0e2a18aabb72c+24129607/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_743ec6d0e2a18aabb72c+24129607/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9cb76b6197266fb795454a7dee50f963485fa52b4f6b5976a73e9555f7e3a1ea
3
+ size 809294
neuronxcc-2.21.33363.0+82129205/MODULE_743ec6d0e2a18aabb72c+24129607/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:06c103ee46d11c52ad9df1a49edeeb5470904e4184f13e728372aa1e4785c230
3
+ size 35247104
neuronxcc-2.21.33363.0+82129205/MODULE_ad3140bf704a4662dc10+a02c3a36/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/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.21.33363.0+82129205/MODULE_ad3140bf704a4662dc10+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_ad3140bf704a4662dc10+a02c3a36/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:15989298c033fa9a8630a83b102f7329938cecca29443a6e9d4994bd381fdf3d
3
+ size 911385
neuronxcc-2.21.33363.0+82129205/MODULE_ad3140bf704a4662dc10+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c321ec7f9d6c273bb294b000482fa16bc9a0cbb66166ca5a7d45b7ce56983122
3
+ size 5540864
neuronxcc-2.21.33363.0+82129205/MODULE_ad3140bf704a4662dc10+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d5865301f554809637efc3125450e525ee25bec5f8752f41ed2d017285116d02
3
+ size 5725360