Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
 Community
709
tengomucho HF Staff commited on
Commit
27a53b9
·
verified ·
1 Parent(s): 03d5d30

Synchronizing local compiler cache.

Browse files
Files changed (21) hide show
  1. .gitattributes +6 -0
  2. neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/c44eac5d03bacf7643a3782a45cab2feff584652bee95cd3c43c7d2600b2f3c5/8452568ed9fe0799584a.json +64 -0
  3. neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/llama/unsloth/Llama-3.1-8B-Instruct/8452568ed9fe0799584a.json +64 -0
  4. neuronxcc-2.26.6360.0+6f180f47/MODULE_45c0bed0ef53eb70769d+665810b0/compile_flags.json +1 -0
  5. neuronxcc-2.26.6360.0+6f180f47/MODULE_45c0bed0ef53eb70769d+665810b0/model.done +0 -0
  6. neuronxcc-2.26.6360.0+6f180f47/MODULE_45c0bed0ef53eb70769d+665810b0/model.hlo_module.pb +3 -0
  7. neuronxcc-2.26.6360.0+6f180f47/MODULE_45c0bed0ef53eb70769d+665810b0/model.neff +3 -0
  8. neuronxcc-2.26.6360.0+6f180f47/MODULE_49b256952b6e47c8eee6+665810b0/compile_flags.json +1 -0
  9. neuronxcc-2.26.6360.0+6f180f47/MODULE_49b256952b6e47c8eee6+665810b0/model.done +0 -0
  10. neuronxcc-2.26.6360.0+6f180f47/MODULE_49b256952b6e47c8eee6+665810b0/model.hlo_module.pb +3 -0
  11. neuronxcc-2.26.6360.0+6f180f47/MODULE_49b256952b6e47c8eee6+665810b0/model.neff +3 -0
  12. neuronxcc-2.26.6360.0+6f180f47/MODULE_a361b21eaf4d1aecc789+4a27301b/compile_flags.json +1 -0
  13. neuronxcc-2.26.6360.0+6f180f47/MODULE_a361b21eaf4d1aecc789+4a27301b/model.done +0 -0
  14. neuronxcc-2.26.6360.0+6f180f47/MODULE_a361b21eaf4d1aecc789+4a27301b/model.hlo_module.pb +3 -0
  15. neuronxcc-2.26.6360.0+6f180f47/MODULE_a361b21eaf4d1aecc789+4a27301b/model.neff +3 -0
  16. neuronxcc-2.26.6360.0+6f180f47/MODULE_a361b21eaf4d1aecc789+4a27301b/wrapped_neff.hlo +3 -0
  17. neuronxcc-2.26.6360.0+6f180f47/MODULE_e549b39486d25778527d+4a27301b/compile_flags.json +1 -0
  18. neuronxcc-2.26.6360.0+6f180f47/MODULE_e549b39486d25778527d+4a27301b/model.done +0 -0
  19. neuronxcc-2.26.6360.0+6f180f47/MODULE_e549b39486d25778527d+4a27301b/model.hlo_module.pb +3 -0
  20. neuronxcc-2.26.6360.0+6f180f47/MODULE_e549b39486d25778527d+4a27301b/model.neff +3 -0
  21. neuronxcc-2.26.6360.0+6f180f47/MODULE_e549b39486d25778527d+4a27301b/wrapped_neff.hlo +3 -0
.gitattributes CHANGED
@@ -18314,3 +18314,9 @@ neuronxcc-2.26.6360.0+6f180f47/MODULE_75f36dd32d471c2241a3+4a27301b/wrapped_neff
18314
  neuronxcc-2.26.6360.0+6f180f47/MODULE_3b158dae1812ab33bdca+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
18315
  neuronxcc-2.26.6360.0+6f180f47/MODULE_76d4cad1258633c54d0d+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
18316
  neuronxcc-2.26.6360.0+6f180f47/MODULE_76d4cad1258633c54d0d+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
18314
  neuronxcc-2.26.6360.0+6f180f47/MODULE_3b158dae1812ab33bdca+519f203d/model.neff filter=lfs diff=lfs merge=lfs -text
18315
  neuronxcc-2.26.6360.0+6f180f47/MODULE_76d4cad1258633c54d0d+84337dd9/model.neff filter=lfs diff=lfs merge=lfs -text
18316
  neuronxcc-2.26.6360.0+6f180f47/MODULE_76d4cad1258633c54d0d+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
18317
+ neuronxcc-2.26.6360.0+6f180f47/MODULE_45c0bed0ef53eb70769d+665810b0/model.neff filter=lfs diff=lfs merge=lfs -text
18318
+ neuronxcc-2.26.6360.0+6f180f47/MODULE_49b256952b6e47c8eee6+665810b0/model.neff filter=lfs diff=lfs merge=lfs -text
18319
+ neuronxcc-2.26.6360.0+6f180f47/MODULE_a361b21eaf4d1aecc789+4a27301b/model.neff filter=lfs diff=lfs merge=lfs -text
18320
+ neuronxcc-2.26.6360.0+6f180f47/MODULE_a361b21eaf4d1aecc789+4a27301b/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
18321
+ neuronxcc-2.26.6360.0+6f180f47/MODULE_e549b39486d25778527d+4a27301b/model.neff filter=lfs diff=lfs merge=lfs -text
18322
+ neuronxcc-2.26.6360.0+6f180f47/MODULE_e549b39486d25778527d+4a27301b/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/c44eac5d03bacf7643a3782a45cab2feff584652bee95cd3c43c7d2600b2f3c5/8452568ed9fe0799584a.json ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "unsloth/Llama-3.1-8B-Instruct",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "LlamaForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "dtype": "bfloat16",
11
+ "head_dim": 128,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 4096,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 14336,
16
+ "max_position_embeddings": 131072,
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": "unsloth/Llama-3.1-8B-Instruct",
24
+ "checkpoint_revision": "4699cc75b550f9c6f3173fb80f4703b62d946aa5",
25
+ "continuous_batching": false,
26
+ "ep_degree": 1,
27
+ "fused_qkv": true,
28
+ "glu_mlp": true,
29
+ "local_ranks_size": 1,
30
+ "max_batch_size": 1,
31
+ "max_context_length": 4096,
32
+ "max_topk": 256,
33
+ "n_active_tokens": 4096,
34
+ "neuronxcc_version": "2.26.6360.0+6f180f47",
35
+ "on_device_sampling": true,
36
+ "optimum_neuron_version": "0.4.7.dev0",
37
+ "output_logits": false,
38
+ "pp_degree": 1,
39
+ "prefill_chunk_size": 1024,
40
+ "sequence_length": 4096,
41
+ "speculation_length": 0,
42
+ "start_rank_id": 0,
43
+ "target": "trn2",
44
+ "torch_dtype": "bfloat16",
45
+ "tp_degree": 1
46
+ },
47
+ "num_attention_heads": 32,
48
+ "num_hidden_layers": 32,
49
+ "num_key_value_heads": 8,
50
+ "pretraining_tp": 1,
51
+ "rms_norm_eps": 1e-05,
52
+ "rope_scaling": {
53
+ "factor": 8.0,
54
+ "high_freq_factor": 4.0,
55
+ "low_freq_factor": 1.0,
56
+ "original_max_position_embeddings": 8192,
57
+ "rope_type": "llama3"
58
+ },
59
+ "rope_theta": 500000.0,
60
+ "tie_word_embeddings": false,
61
+ "unsloth_fixed": true,
62
+ "use_cache": true,
63
+ "vocab_size": 128256
64
+ }
neuronxcc-2.26.6360.0+6f180f47/0_REGISTRY/0.4.7.dev0/llama/unsloth/Llama-3.1-8B-Instruct/8452568ed9fe0799584a.json ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "unsloth/Llama-3.1-8B-Instruct",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "LlamaForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "dtype": "bfloat16",
11
+ "head_dim": 128,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 4096,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 14336,
16
+ "max_position_embeddings": 131072,
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": "unsloth/Llama-3.1-8B-Instruct",
24
+ "checkpoint_revision": "4699cc75b550f9c6f3173fb80f4703b62d946aa5",
25
+ "continuous_batching": false,
26
+ "ep_degree": 1,
27
+ "fused_qkv": true,
28
+ "glu_mlp": true,
29
+ "local_ranks_size": 1,
30
+ "max_batch_size": 1,
31
+ "max_context_length": 4096,
32
+ "max_topk": 256,
33
+ "n_active_tokens": 4096,
34
+ "neuronxcc_version": "2.26.6360.0+6f180f47",
35
+ "on_device_sampling": true,
36
+ "optimum_neuron_version": "0.4.7.dev0",
37
+ "output_logits": false,
38
+ "pp_degree": 1,
39
+ "prefill_chunk_size": 1024,
40
+ "sequence_length": 4096,
41
+ "speculation_length": 0,
42
+ "start_rank_id": 0,
43
+ "target": "trn2",
44
+ "torch_dtype": "bfloat16",
45
+ "tp_degree": 1
46
+ },
47
+ "num_attention_heads": 32,
48
+ "num_hidden_layers": 32,
49
+ "num_key_value_heads": 8,
50
+ "pretraining_tp": 1,
51
+ "rms_norm_eps": 1e-05,
52
+ "rope_scaling": {
53
+ "factor": 8.0,
54
+ "high_freq_factor": 4.0,
55
+ "low_freq_factor": 1.0,
56
+ "original_max_position_embeddings": 8192,
57
+ "rope_type": "llama3"
58
+ },
59
+ "rope_theta": 500000.0,
60
+ "tie_word_embeddings": false,
61
+ "unsloth_fixed": true,
62
+ "use_cache": true,
63
+ "vocab_size": 128256
64
+ }
neuronxcc-2.26.6360.0+6f180f47/MODULE_45c0bed0ef53eb70769d+665810b0/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn2", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=2", "--verbose=35", "--logfile=/tmp/nxd_model/chunked_prefill/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.26.6360.0+6f180f47/MODULE_45c0bed0ef53eb70769d+665810b0/model.done ADDED
File without changes
neuronxcc-2.26.6360.0+6f180f47/MODULE_45c0bed0ef53eb70769d+665810b0/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:30d901717677b45da4031b4966382ca6dc1388d5ca3f4c51736425de2571b5cb
3
+ size 3864154
neuronxcc-2.26.6360.0+6f180f47/MODULE_45c0bed0ef53eb70769d+665810b0/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9dd2de91c1e05ddfa81c191c4838eb8c88b00858c55d11602393e8a65e10cc76
3
+ size 11822490
neuronxcc-2.26.6360.0+6f180f47/MODULE_49b256952b6e47c8eee6+665810b0/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn2", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=2", "--verbose=35", "--logfile=/tmp/nxd_model/chunked_prefill/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.26.6360.0+6f180f47/MODULE_49b256952b6e47c8eee6+665810b0/model.done ADDED
File without changes
neuronxcc-2.26.6360.0+6f180f47/MODULE_49b256952b6e47c8eee6+665810b0/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9b9672f81b6e5c2c243c2e11778ce8a0a2583a88da0e45a0f92d013a0e191178
3
+ size 2831198
neuronxcc-2.26.6360.0+6f180f47/MODULE_49b256952b6e47c8eee6+665810b0/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4187401636809981d251d583523f42436eea8a193662e22a0f04239806e71b7f
3
+ size 5340742
neuronxcc-2.26.6360.0+6f180f47/MODULE_a361b21eaf4d1aecc789+4a27301b/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn2", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=2", "--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_a361b21eaf4d1aecc789+4a27301b/model.done ADDED
File without changes
neuronxcc-2.26.6360.0+6f180f47/MODULE_a361b21eaf4d1aecc789+4a27301b/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:05300fecbb9305be6b0b14452ebbe4f064a2ebbccc9724ddaaec7b6a9d5c34ac
3
+ size 2885108
neuronxcc-2.26.6360.0+6f180f47/MODULE_a361b21eaf4d1aecc789+4a27301b/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:48e608b17f04dfd8416b80d0c02935de810192107ed409f34b2cd86ee2c13284
3
+ size 18730935
neuronxcc-2.26.6360.0+6f180f47/MODULE_a361b21eaf4d1aecc789+4a27301b/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ead3c445372e1d69642e123c974cec8eceb62612b43167219326cea4fb12afec
3
+ size 18881118
neuronxcc-2.26.6360.0+6f180f47/MODULE_e549b39486d25778527d+4a27301b/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn2", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=2", "--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_e549b39486d25778527d+4a27301b/model.done ADDED
File without changes
neuronxcc-2.26.6360.0+6f180f47/MODULE_e549b39486d25778527d+4a27301b/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:46dea878124ca60699796c4ce53cada51d808c394fe53f82a3b36d6553d79321
3
+ size 2167725
neuronxcc-2.26.6360.0+6f180f47/MODULE_e549b39486d25778527d+4a27301b/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3fe73a264e3f5fedcb0181cf564991f265d6ad8668a8ae1c70070101f9afc463
3
+ size 1599979
neuronxcc-2.26.6360.0+6f180f47/MODULE_e549b39486d25778527d+4a27301b/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8a5258b70c6fb1805a87c5471f19635597eeecf5aaf323a3689a4958017c6a23
3
+ size 1701504