Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
Community
629
dacorvo HF Staff commited on
Commit
6b83242
·
verified ·
1 Parent(s): 43f6ca6

Synchronizing local compiler cache.

Browse files
Files changed (30) hide show
  1. .gitattributes +4 -0
  2. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.3.0.dev4/llama/princeton-nlp/Sheared-LLaMA-1.3B/395a6f32985d1a6fe6bd.json +71 -0
  3. neuronxcc-2.17.194.0+d312836f/MODULE_24ebae90c3026e285413+165e9558/model.hlo_module.pb +1 -1
  4. neuronxcc-2.17.194.0+d312836f/MODULE_24ebae90c3026e285413+165e9558/model.neff +1 -1
  5. neuronxcc-2.17.194.0+d312836f/MODULE_24ebae90c3026e285413+165e9558/wrapped_neff.hlo +1 -1
  6. neuronxcc-2.17.194.0+d312836f/MODULE_47eabeb917b5a1092c42+165e9558/compile_flags.json +1 -0
  7. neuronxcc-2.17.194.0+d312836f/MODULE_47eabeb917b5a1092c42+165e9558/model.done +0 -0
  8. neuronxcc-2.17.194.0+d312836f/MODULE_47eabeb917b5a1092c42+165e9558/model.hlo_module.pb +3 -0
  9. neuronxcc-2.17.194.0+d312836f/MODULE_47eabeb917b5a1092c42+165e9558/model.neff +3 -0
  10. neuronxcc-2.17.194.0+d312836f/MODULE_47eabeb917b5a1092c42+165e9558/wrapped_neff.hlo +3 -0
  11. neuronxcc-2.17.194.0+d312836f/MODULE_4ee273ae3c24111ed563+bfe5714b/model.hlo_module.pb +1 -1
  12. neuronxcc-2.17.194.0+d312836f/MODULE_4ee273ae3c24111ed563+bfe5714b/model.neff +1 -1
  13. neuronxcc-2.17.194.0+d312836f/MODULE_854169d7dd7fe4ccf204+bfe5714b/model.hlo_module.pb +1 -1
  14. neuronxcc-2.17.194.0+d312836f/MODULE_854169d7dd7fe4ccf204+bfe5714b/model.neff +1 -1
  15. neuronxcc-2.17.194.0+d312836f/MODULE_b2c92680f8df8d069137+165e9558/model.hlo_module.pb +1 -1
  16. neuronxcc-2.17.194.0+d312836f/MODULE_b2c92680f8df8d069137+165e9558/model.neff +1 -1
  17. neuronxcc-2.17.194.0+d312836f/MODULE_b2c92680f8df8d069137+165e9558/wrapped_neff.hlo +1 -1
  18. neuronxcc-2.17.194.0+d312836f/MODULE_c34acb78ed4dd8ec151f+bfe5714b/compile_flags.json +1 -0
  19. neuronxcc-2.17.194.0+d312836f/MODULE_c34acb78ed4dd8ec151f+bfe5714b/model.done +0 -0
  20. neuronxcc-2.17.194.0+d312836f/MODULE_c34acb78ed4dd8ec151f+bfe5714b/model.hlo_module.pb +3 -0
  21. neuronxcc-2.17.194.0+d312836f/MODULE_c34acb78ed4dd8ec151f+bfe5714b/model.neff +3 -0
  22. neuronxcc-2.17.194.0+d312836f/MODULE_c85201a721ea86600d77+bfe5714b/model.hlo_module.pb +1 -1
  23. neuronxcc-2.17.194.0+d312836f/MODULE_c85201a721ea86600d77+bfe5714b/model.neff +1 -1
  24. neuronxcc-2.17.194.0+d312836f/MODULE_fc5c44298c77e23327ca+165e9558/model.hlo_module.pb +1 -1
  25. neuronxcc-2.17.194.0+d312836f/MODULE_fc5c44298c77e23327ca+165e9558/model.neff +1 -1
  26. neuronxcc-2.17.194.0+d312836f/MODULE_fc5c44298c77e23327ca+165e9558/wrapped_neff.hlo +1 -1
  27. neuronxcc-2.17.194.0+d312836f/MODULE_ff65700565ad6ef146a2+431f5505/compile_flags.json +1 -0
  28. neuronxcc-2.17.194.0+d312836f/MODULE_ff65700565ad6ef146a2+431f5505/model.done +0 -0
  29. neuronxcc-2.17.194.0+d312836f/MODULE_ff65700565ad6ef146a2+431f5505/model.hlo_module.pb +3 -0
  30. neuronxcc-2.17.194.0+d312836f/MODULE_ff65700565ad6ef146a2+431f5505/model.neff +3 -0
.gitattributes CHANGED
@@ -9112,3 +9112,7 @@ neuronxcc-2.17.194.0+d312836f/MODULE_6ac64ef4119d776ebae5+bfe5714b/model.neff fi
9112
  neuronxcc-2.17.194.0+d312836f/MODULE_c543518923444641ffef+165e9558/model.neff filter=lfs diff=lfs merge=lfs -text
9113
  neuronxcc-2.17.194.0+d312836f/MODULE_c543518923444641ffef+165e9558/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
9114
  neuronxcc-2.17.194.0+d312836f/MODULE_f4f73b6e4e0ecc8a6905+431f5505/model.neff filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
9112
  neuronxcc-2.17.194.0+d312836f/MODULE_c543518923444641ffef+165e9558/model.neff filter=lfs diff=lfs merge=lfs -text
9113
  neuronxcc-2.17.194.0+d312836f/MODULE_c543518923444641ffef+165e9558/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
9114
  neuronxcc-2.17.194.0+d312836f/MODULE_f4f73b6e4e0ecc8a6905+431f5505/model.neff filter=lfs diff=lfs merge=lfs -text
9115
+ neuronxcc-2.17.194.0+d312836f/MODULE_47eabeb917b5a1092c42+165e9558/model.neff filter=lfs diff=lfs merge=lfs -text
9116
+ neuronxcc-2.17.194.0+d312836f/MODULE_47eabeb917b5a1092c42+165e9558/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
9117
+ neuronxcc-2.17.194.0+d312836f/MODULE_c34acb78ed4dd8ec151f+bfe5714b/model.neff filter=lfs diff=lfs merge=lfs -text
9118
+ neuronxcc-2.17.194.0+d312836f/MODULE_ff65700565ad6ef146a2+431f5505/model.neff filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.3.0.dev4/llama/princeton-nlp/Sheared-LLaMA-1.3B/395a6f32985d1a6fe6bd.json ADDED
@@ -0,0 +1,71 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "princeton-nlp/Sheared-LLaMA-1.3B",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "LlamaForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "head_dim": 128,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 2048,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 5504,
15
+ "max_position_embeddings": 4096,
16
+ "mlp_bias": false,
17
+ "model_type": "llama",
18
+ "neuron": {
19
+ "_serialized_key": "NxDNeuronConfig",
20
+ "async_mode": false,
21
+ "attn_kernel_enabled": false,
22
+ "batch_size": 4,
23
+ "capacity_factor": null,
24
+ "cc_pipeline_tiling_factor": 2,
25
+ "checkpoint_id": "princeton-nlp/Sheared-LLaMA-1.3B",
26
+ "checkpoint_revision": "a4b76938edbf571ea7d7d9904861cbdca08809b4",
27
+ "continuous_batching": true,
28
+ "enable_bucketing": false,
29
+ "ep_degree": 1,
30
+ "flash_decoding_enabled": false,
31
+ "fused_qkv": true,
32
+ "glu_mlp": true,
33
+ "is_chunked_prefill": false,
34
+ "local_ranks_size": 2,
35
+ "logical_nc_config": 1,
36
+ "max_batch_size": 4,
37
+ "max_context_length": 4096,
38
+ "max_topk": 256,
39
+ "mlp_kernel_enabled": false,
40
+ "mlp_kernel_fuse_residual_add": false,
41
+ "n_active_tokens": 4096,
42
+ "neuronxcc_version": "2.17.194.0+d312836f",
43
+ "num_cores_per_group": 1,
44
+ "on_device_sampling": false,
45
+ "optimum_neuron_version": "0.3.0.dev4",
46
+ "output_logits": false,
47
+ "padding_side": "right",
48
+ "pp_degree": 1,
49
+ "qk_layernorm": false,
50
+ "qkv_kernel_enabled": false,
51
+ "rpl_reduce_dtype": "float16",
52
+ "sequence_length": 4096,
53
+ "sequence_parallel_enabled": false,
54
+ "speculation_length": 0,
55
+ "start_rank_id": 0,
56
+ "target": null,
57
+ "torch_dtype": "float16",
58
+ "tp_degree": 2,
59
+ "vocab_parallel": false
60
+ },
61
+ "num_attention_heads": 16,
62
+ "num_hidden_layers": 24,
63
+ "num_key_value_heads": 16,
64
+ "pretraining_tp": 1,
65
+ "rms_norm_eps": 1e-05,
66
+ "rope_scaling": null,
67
+ "rope_theta": 10000.0,
68
+ "tie_word_embeddings": false,
69
+ "use_cache": true,
70
+ "vocab_size": 32000
71
+ }
neuronxcc-2.17.194.0+d312836f/MODULE_24ebae90c3026e285413+165e9558/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8b9620bbcee180a5e4367c50b9fad491bb7ba647b87af1dac526bdbd5afcfcb9
3
  size 780838
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e1310afafb3287e5bff0d6cf0ace3b4ad3bd5a5aa3edd2bfa85d9978e95ebd2
3
  size 780838
neuronxcc-2.17.194.0+d312836f/MODULE_24ebae90c3026e285413+165e9558/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d273282b2415a8ab094a78cd734cf2ff37d985a1ff13928713d56d4123ea991b
3
  size 3830784
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b809c75a55343aeaee0ec0d12d3bca7fb5ffa5311018df853fe85ee4e9f0e1a7
3
  size 3830784
neuronxcc-2.17.194.0+d312836f/MODULE_24ebae90c3026e285413+165e9558/wrapped_neff.hlo CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:308754b6190112dd969aeb0bc0a1718972a05657c3c52635945b9ffc2582f63a
3
  size 3968601
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:edf2befb62ac89a047d5e40a6082a4be40b3de39eb3d47e5a96ab277f38702b9
3
  size 3968601
neuronxcc-2.17.194.0+d312836f/MODULE_47eabeb917b5a1092c42+165e9558/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ "--auto-cast=none --model-type=transformer --tensorizer-options='--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-dge-dma --vectorize-strided-dma ' -O2 --internal-num-neuroncores-per-sengine=1 --logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt --enable-internal-neff-wrapper"
neuronxcc-2.17.194.0+d312836f/MODULE_47eabeb917b5a1092c42+165e9558/model.done ADDED
File without changes
neuronxcc-2.17.194.0+d312836f/MODULE_47eabeb917b5a1092c42+165e9558/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6ad9c918cf2fb7e98406ac83564033df53d805abfe24c29caa7a3a6b387e4f29
3
+ size 493198
neuronxcc-2.17.194.0+d312836f/MODULE_47eabeb917b5a1092c42+165e9558/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:07a3c3129ca52259f017f90a71b07ba96652add5d958068489e7fc359733a6bf
3
+ size 3431424
neuronxcc-2.17.194.0+d312836f/MODULE_47eabeb917b5a1092c42+165e9558/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f5fa6d13656f192e3231da24253f35bd960d89448e61c0f6637b927717db0a82
3
+ size 3534035
neuronxcc-2.17.194.0+d312836f/MODULE_4ee273ae3c24111ed563+bfe5714b/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:de7a9478d2432f19ee226b6f93469f8e00cabdf9fb0054b2018c152d5141d29a
3
  size 924407
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:07a04a46167eebec5ab29ee1850c37d4897ebef1da46fc5a5c01b68d8fac0909
3
  size 924407
neuronxcc-2.17.194.0+d312836f/MODULE_4ee273ae3c24111ed563+bfe5714b/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d84528624e456b916c68762805b04eefe6919ec65efdcba6b5011629212a7548
3
  size 32635904
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5f669483aa76ba11d216560c146ae142fe85910e78b1f5c75cd8c9e8e9a6ecf3
3
  size 32635904
neuronxcc-2.17.194.0+d312836f/MODULE_854169d7dd7fe4ccf204+bfe5714b/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:166e96a30da3f5253aa08eb3799951515151a7eba4a632147371530199c79b1d
3
  size 924407
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d92562501eb731a91128371276e5ae809dee0f13e1016dd3e3a778f6349a9fd0
3
  size 924407
neuronxcc-2.17.194.0+d312836f/MODULE_854169d7dd7fe4ccf204+bfe5714b/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a51516e6b48e0b222c93b07eb813f2adf7f4a84e3e67c92c56b1544ae4702f68
3
  size 32635904
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4e91e527614dc566a59da35f213206bd1d5d63faa1f07e0ea5c88f39489554f1
3
  size 32635904
neuronxcc-2.17.194.0+d312836f/MODULE_b2c92680f8df8d069137+165e9558/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9eb4a0361b32b12f513b43656f0e2bb39da854f74c19cdd0a05072adaffe23ca
3
  size 780822
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f0a5abf446e9e51df15114a6e45c4b4518d43f80a633e436184342068fcff037
3
  size 780822
neuronxcc-2.17.194.0+d312836f/MODULE_b2c92680f8df8d069137+165e9558/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d0d08ab4943934367aeb1f55dac58462076c5822de08fb64a5c95a326881433c
3
  size 3134464
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2127a10be985fe9e0fa5cdcbb9872ecfa946930d13bf80ea9ee9f1e7872ad1e2
3
  size 3134464
neuronxcc-2.17.194.0+d312836f/MODULE_b2c92680f8df8d069137+165e9558/wrapped_neff.hlo CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f539c9a8a2a3274cc35dc1953397bc6a3bf0bc64b5ce9f84ed693084e647f045
3
  size 3272281
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:615d7f646a8a6eeff2ca4a93b9f79ce455b54dbd78c8d4063317128e940ed0f1
3
  size 3272281
neuronxcc-2.17.194.0+d312836f/MODULE_c34acb78ed4dd8ec151f+bfe5714b/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ "--auto-cast=none --model-type=transformer --tensorizer-options='--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-dge-dma --vectorize-strided-dma ' -O2 --internal-num-neuroncores-per-sengine=1 --logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"
neuronxcc-2.17.194.0+d312836f/MODULE_c34acb78ed4dd8ec151f+bfe5714b/model.done ADDED
File without changes
neuronxcc-2.17.194.0+d312836f/MODULE_c34acb78ed4dd8ec151f+bfe5714b/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:93644d2175c80a21c512b2fbc4f1346203c51079ea72c22fb3dfa4b44e21d271
3
+ size 633457
neuronxcc-2.17.194.0+d312836f/MODULE_c34acb78ed4dd8ec151f+bfe5714b/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e31c77747d84b9cf2e11e0469e7ba99a0e19ebde770025a0c5f55d32cdb040b6
3
+ size 30751744
neuronxcc-2.17.194.0+d312836f/MODULE_c85201a721ea86600d77+bfe5714b/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:281c3001907ab2b54aab495a821bc3bf1378fc99da416774db89f514c185f90f
3
  size 924407
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:65e9591f1dcda70919dc05772a5920bab62b4d8baa3e52b1b8040f8d410ade4c
3
  size 924407
neuronxcc-2.17.194.0+d312836f/MODULE_c85201a721ea86600d77+bfe5714b/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:124870ed611c97763d705c52d0ef48e19cc6ae73fa6b2404888d328ec80670bd
3
  size 32635904
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:09f09ba8b35b8fab4458ed0de0a0a27b7e38620db57ed1c126cfa272bc538b3b
3
  size 32635904
neuronxcc-2.17.194.0+d312836f/MODULE_fc5c44298c77e23327ca+165e9558/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f3ea7ba2e92928e91fdfb94c40841cd478587b155401cf896756a0840569dd12
3
  size 781222
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c645c0ef951f934a1f89cd515de836818d6936b9bbfa360ed05a5c2425385fca
3
  size 781222
neuronxcc-2.17.194.0+d312836f/MODULE_fc5c44298c77e23327ca+165e9558/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:eaf24fad34b29fba1ad292269ff9826601eb89d03ab408a16bff07956e5fd19a
3
  size 4977664
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e4176d320a84a26a51bf312299251126e7194d7c212ccbb46a163135350053ba
3
  size 4977664
neuronxcc-2.17.194.0+d312836f/MODULE_fc5c44298c77e23327ca+165e9558/wrapped_neff.hlo CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:559f910d2365e541ddd101afc671f97ec3cb569fb8a315da887370f2320e98ed
3
  size 5115481
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:17c574adedb171a1096b2482bfa593ced82d27774588ec50f22e422ceb167a13
3
  size 5115481
neuronxcc-2.17.194.0+d312836f/MODULE_ff65700565ad6ef146a2+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-2.17.194.0+d312836f/MODULE_ff65700565ad6ef146a2+431f5505/model.done ADDED
File without changes
neuronxcc-2.17.194.0+d312836f/MODULE_ff65700565ad6ef146a2+431f5505/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:47ac7513117074cf1539b5e99c31e6c6840b364ae487783e37c7f291104f7977
3
+ size 100901
neuronxcc-2.17.194.0+d312836f/MODULE_ff65700565ad6ef146a2+431f5505/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:30139a338a9b47e57fb4c5cde502c668417e659bd103a420a7a527e29ef5c4a5
3
+ size 1229824