Commit
41867c0
·
verified ·
1 Parent(s): 7dcaca4

Synchronizing local compiler cache.

Browse files
Files changed (21) hide show
  1. .gitattributes +3 -0
  2. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev1/granite/ibm-granite/granite-3.1-2b-instruct/a55342f00a2b27869b7c.json +58 -0
  3. neuronxcc-2.21.33363.0+82129205/MODULE_1934c2fc93e7793f0aef+a02c3a36/model.neff +1 -1
  4. neuronxcc-2.21.33363.0+82129205/MODULE_1934c2fc93e7793f0aef+a02c3a36/wrapped_neff.hlo +1 -1
  5. neuronxcc-2.21.33363.0+82129205/MODULE_238f936f6792c3472f34+24129607/compile_flags.json +1 -0
  6. neuronxcc-2.21.33363.0+82129205/MODULE_238f936f6792c3472f34+24129607/model.done +0 -0
  7. neuronxcc-2.21.33363.0+82129205/MODULE_238f936f6792c3472f34+24129607/model.hlo_module.pb +3 -0
  8. neuronxcc-2.21.33363.0+82129205/MODULE_238f936f6792c3472f34+24129607/model.neff +3 -0
  9. neuronxcc-2.21.33363.0+82129205/MODULE_30826b478ad3db0c312e+24129607/model.hlo_module.pb +1 -1
  10. neuronxcc-2.21.33363.0+82129205/MODULE_30826b478ad3db0c312e+24129607/model.neff +1 -1
  11. neuronxcc-2.21.33363.0+82129205/MODULE_3c75648f1b6d85bd72d5+a02c3a36/model.neff +1 -1
  12. neuronxcc-2.21.33363.0+82129205/MODULE_3c75648f1b6d85bd72d5+a02c3a36/wrapped_neff.hlo +1 -1
  13. neuronxcc-2.21.33363.0+82129205/MODULE_9c53a5ac284b649e893b+a02c3a36/model.neff +1 -1
  14. neuronxcc-2.21.33363.0+82129205/MODULE_9c53a5ac284b649e893b+a02c3a36/wrapped_neff.hlo +1 -1
  15. neuronxcc-2.21.33363.0+82129205/MODULE_b6a329fd6d4912085786+a02c3a36/model.neff +1 -1
  16. neuronxcc-2.21.33363.0+82129205/MODULE_b6a329fd6d4912085786+a02c3a36/wrapped_neff.hlo +1 -1
  17. neuronxcc-2.21.33363.0+82129205/MODULE_f6a3ff2ad752e338116c+a02c3a36/compile_flags.json +1 -0
  18. neuronxcc-2.21.33363.0+82129205/MODULE_f6a3ff2ad752e338116c+a02c3a36/model.done +0 -0
  19. neuronxcc-2.21.33363.0+82129205/MODULE_f6a3ff2ad752e338116c+a02c3a36/model.hlo_module.pb +3 -0
  20. neuronxcc-2.21.33363.0+82129205/MODULE_f6a3ff2ad752e338116c+a02c3a36/model.neff +3 -0
  21. neuronxcc-2.21.33363.0+82129205/MODULE_f6a3ff2ad752e338116c+a02c3a36/wrapped_neff.hlo +3 -0
.gitattributes CHANGED
@@ -5874,3 +5874,6 @@ neuronxcc-2.21.33363.0+82129205/MODULE_aa1a56237e3ad1974672+a02c3a36/wrapped_nef
5874
  neuronxcc-2.21.33363.0+82129205/MODULE_c3ed26835961cb74a993+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
5875
  neuronxcc-2.21.33363.0+82129205/MODULE_c3ed26835961cb74a993+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
5876
  neuronxcc-2.21.33363.0+82129205/MODULE_e31397d074de214d8505+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
 
 
 
 
5874
  neuronxcc-2.21.33363.0+82129205/MODULE_c3ed26835961cb74a993+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
5875
  neuronxcc-2.21.33363.0+82129205/MODULE_c3ed26835961cb74a993+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
5876
  neuronxcc-2.21.33363.0+82129205/MODULE_e31397d074de214d8505+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
5877
+ neuronxcc-2.21.33363.0+82129205/MODULE_238f936f6792c3472f34+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
5878
+ neuronxcc-2.21.33363.0+82129205/MODULE_f6a3ff2ad752e338116c+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
5879
+ neuronxcc-2.21.33363.0+82129205/MODULE_f6a3ff2ad752e338116c+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev1/granite/ibm-granite/granite-3.1-2b-instruct/a55342f00a2b27869b7c.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "ibm-granite/granite-3.1-2b-instruct",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "GraniteForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.1,
10
+ "attention_multiplier": 0.015625,
11
+ "dtype": "bfloat16",
12
+ "embedding_multiplier": 12.0,
13
+ "hidden_act": "silu",
14
+ "hidden_size": 2048,
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 8192,
17
+ "logits_scaling": 8.0,
18
+ "max_position_embeddings": 131072,
19
+ "mlp_bias": false,
20
+ "model_type": "granite",
21
+ "neuron": {
22
+ "_serialized_key": "NxDNeuronConfig",
23
+ "batch_size": 1,
24
+ "capacity_factor": null,
25
+ "checkpoint_id": "ibm-granite/granite-3.1-2b-instruct",
26
+ "checkpoint_revision": "bbc2aed595bd38bd770263dc3ab831db9794441d",
27
+ "continuous_batching": false,
28
+ "ep_degree": 1,
29
+ "fused_qkv": true,
30
+ "glu_mlp": true,
31
+ "local_ranks_size": 2,
32
+ "max_batch_size": 1,
33
+ "max_context_length": 8192,
34
+ "max_topk": 256,
35
+ "n_active_tokens": 8192,
36
+ "neuronxcc_version": "2.21.33363.0+82129205",
37
+ "on_device_sampling": true,
38
+ "optimum_neuron_version": "0.4.4.dev1",
39
+ "output_logits": false,
40
+ "pp_degree": 1,
41
+ "sequence_length": 8192,
42
+ "speculation_length": 0,
43
+ "start_rank_id": 0,
44
+ "target": "trn1",
45
+ "torch_dtype": "bfloat16",
46
+ "tp_degree": 2
47
+ },
48
+ "num_attention_heads": 32,
49
+ "num_hidden_layers": 40,
50
+ "num_key_value_heads": 8,
51
+ "residual_multiplier": 0.22,
52
+ "rms_norm_eps": 1e-05,
53
+ "rope_scaling": null,
54
+ "rope_theta": 5000000.0,
55
+ "tie_word_embeddings": true,
56
+ "use_cache": true,
57
+ "vocab_size": 49155
58
+ }
neuronxcc-2.21.33363.0+82129205/MODULE_1934c2fc93e7793f0aef+a02c3a36/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:bd1f0f7832922f80018693c7e5f05f2af87da8c5ccd61d46747d7d5dfdf2ddd0
3
  size 1936384
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:be0fff828fd38c1aee43143f27a0465c5581091a6665253c6dd7552af3f6cad4
3
  size 1936384
neuronxcc-2.21.33363.0+82129205/MODULE_1934c2fc93e7793f0aef+a02c3a36/wrapped_neff.hlo CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d39862519f89981012eab6c1d75d52c052563ea7b90f1f36150a854435daac50
3
  size 2092575
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c576c406877b5b57dd3bbbf4c1f199862c0a000db3a5183fa6daf8e6ec87c6d4
3
  size 2092575
neuronxcc-2.21.33363.0+82129205/MODULE_238f936f6792c3472f34+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_238f936f6792c3472f34+24129607/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_238f936f6792c3472f34+24129607/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3ea0d8cb15cbca6ee67fc130522321761e21811e03149600c91b7f97eee96ab2
3
+ size 961456
neuronxcc-2.21.33363.0+82129205/MODULE_238f936f6792c3472f34+24129607/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:baa8da791e88015319d46aecb25cebbb9b78184766aec862e17c9cf7aeaf8483
3
+ size 14644224
neuronxcc-2.21.33363.0+82129205/MODULE_30826b478ad3db0c312e+24129607/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1e8f102543359b90db4ce07789fb2411309e86c1c0421f0141e0ca5df84c22c4
3
  size 726810
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:750f3632ea38b335655fd28f42584cdcaa65b02786976fcd408f8314f9cd17b5
3
  size 726810
neuronxcc-2.21.33363.0+82129205/MODULE_30826b478ad3db0c312e+24129607/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ec265c44f447e9681fb65ed2166c579a88311c8bd934a107f0ac854d669dc27a
3
  size 71302144
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8c21b8d2c352b0e8f28c965f110de45f1d43dc3e5773994dd2808e2524d999b2
3
  size 71302144
neuronxcc-2.21.33363.0+82129205/MODULE_3c75648f1b6d85bd72d5+a02c3a36/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f8806f154792b185e952668bf467a6f45313fb98068845916efbb38366b36f80
3
  size 3073024
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7ef93799d48af862b4c7eaf33cd52ec052e50b7f66c1e98a1fbd8a034562a82a
3
  size 3073024
neuronxcc-2.21.33363.0+82129205/MODULE_3c75648f1b6d85bd72d5+a02c3a36/wrapped_neff.hlo CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9487ed19414eca5f4dfe60bfef8c7487cfa96943f8f65cbbdb3697963cd8a31e
3
  size 3147125
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f2b74f65fb1ccbe1abf57b8c48abb04bc9a50b25249e125bddd26f645ba8b860
3
  size 3147125
neuronxcc-2.21.33363.0+82129205/MODULE_9c53a5ac284b649e893b+a02c3a36/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d3ab0d36b2e071a0e3bb0ce9372831f5e81acd384623417a971849adb4f94849
3
  size 6933504
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a3ba84571c1e52b46b083f32a8e42fb79dcdea40145b30d06ab8a4ba8cde832c
3
  size 6933504
neuronxcc-2.21.33363.0+82129205/MODULE_9c53a5ac284b649e893b+a02c3a36/wrapped_neff.hlo CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9452993693d4e48409871d085542e91e2bcb96f00202fc7dedce45fda1047441
3
  size 7117697
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c248c978668e18e40e71cb03ba1b90aec28594c57cdc20e425ee53a1d76fb5fb
3
  size 7117697
neuronxcc-2.21.33363.0+82129205/MODULE_b6a329fd6d4912085786+a02c3a36/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f46abd0df3c7b95eb482563152cc52e30c9bd91a396ab8768218312ba40f18bc
3
  size 1721344
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:518ab130bd8f46ee96fbfe9e1ca3aa7ba8bdb781938f46f6ba02a73979d931e4
3
  size 1721344
neuronxcc-2.21.33363.0+82129205/MODULE_b6a329fd6d4912085786+a02c3a36/wrapped_neff.hlo CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:40a0d7b481ce53c5150d00ed17a768db35b4c5b40786bb0ad11458dfec33e14e
3
  size 1877650
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2245dcf5e066ce50a9c705d0a9374948b4f45a57aafe66fde7b227170e89da6a
3
  size 1877650
neuronxcc-2.21.33363.0+82129205/MODULE_f6a3ff2ad752e338116c+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_f6a3ff2ad752e338116c+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_f6a3ff2ad752e338116c+a02c3a36/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1db9fbaa948ee5a336ec5eb66ccf4bfbd49cf94ccf20ce7360119546014d2bd5
3
+ size 1088079
neuronxcc-2.21.33363.0+82129205/MODULE_f6a3ff2ad752e338116c+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1a05be195ea62a2460eef02acad4497d201f4a466a664c89753f52d28264bf57
3
+ size 5008384
neuronxcc-2.21.33363.0+82129205/MODULE_f6a3ff2ad752e338116c+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d01d53d9c2db3968aebe5732528cda022c4ac8775d540940642555b6d74d58d9
3
+ size 5192563