Hugging Face's logo

optimum-internal-testing
/
neuron-testing-cache

Model card Files
xet
 Community
dacorvo HF Staff commited on
Commit
fe8f22d
·
verified ·
1 Parent(s): eacb3d8

Synchronizing local compiler cache.

Browse files
Files changed (17) hide show
  1. .gitattributes +3 -0
  2. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev3/527899cc3b15a98022089573f6e42212494e0dbda65770e25bddf8fe48c6bd4a/8f586dd55a9038556565.json +59 -0
  3. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev3/granite/ibm-granite/granite-3.1-2b-base/8f586dd55a9038556565.json +59 -0
  4. neuronxcc-2.21.33363.0+82129205/MODULE_17476209562158013765+e30acd3a/model.neff +0 -0
  5. neuronxcc-2.21.33363.0+82129205/MODULE_8ff1216675c7144590c1+c4f887dc/model.neff +1 -1
  6. neuronxcc-2.21.33363.0+82129205/MODULE_8ff1216675c7144590c1+c4f887dc/wrapped_neff.hlo +1 -1
  7. neuronxcc-2.21.33363.0+82129205/MODULE_9295293099bd829ab6ce+24129607/compile_flags.json +1 -0
  8. neuronxcc-2.21.33363.0+82129205/MODULE_9295293099bd829ab6ce+24129607/model.done +0 -0
  9. neuronxcc-2.21.33363.0+82129205/MODULE_9295293099bd829ab6ce+24129607/model.hlo_module.pb +3 -0
  10. neuronxcc-2.21.33363.0+82129205/MODULE_9295293099bd829ab6ce+24129607/model.neff +3 -0
  11. neuronxcc-2.21.33363.0+82129205/MODULE_ad16674dd3fc5a51c226+a02c3a36/model.neff +1 -1
  12. neuronxcc-2.21.33363.0+82129205/MODULE_ad16674dd3fc5a51c226+a02c3a36/wrapped_neff.hlo +1 -1
  13. neuronxcc-2.21.33363.0+82129205/MODULE_cad34e8b42cec5152517+a02c3a36/compile_flags.json +1 -0
  14. neuronxcc-2.21.33363.0+82129205/MODULE_cad34e8b42cec5152517+a02c3a36/model.done +0 -0
  15. neuronxcc-2.21.33363.0+82129205/MODULE_cad34e8b42cec5152517+a02c3a36/model.hlo_module.pb +3 -0
  16. neuronxcc-2.21.33363.0+82129205/MODULE_cad34e8b42cec5152517+a02c3a36/model.neff +3 -0
  17. neuronxcc-2.21.33363.0+82129205/MODULE_cad34e8b42cec5152517+a02c3a36/wrapped_neff.hlo +3 -0
.gitattributes CHANGED
@@ -7768,3 +7768,6 @@ neuronxcc-2.21.33363.0+82129205/MODULE_c68932cfa906e878e55b+24129607/model.neff
7768
  neuronxcc-2.21.33363.0+82129205/MODULE_9c84a2481334a261eed0+6170d8e1/model.neff filter=lfs diff=lfs merge=lfs -text
7769
  neuronxcc-2.21.33363.0+82129205/MODULE_f1b8afb5fe5b5d394c2b+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
7770
  neuronxcc-2.21.33363.0+82129205/MODULE_f1b8afb5fe5b5d394c2b+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
7768
  neuronxcc-2.21.33363.0+82129205/MODULE_9c84a2481334a261eed0+6170d8e1/model.neff filter=lfs diff=lfs merge=lfs -text
7769
  neuronxcc-2.21.33363.0+82129205/MODULE_f1b8afb5fe5b5d394c2b+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
7770
  neuronxcc-2.21.33363.0+82129205/MODULE_f1b8afb5fe5b5d394c2b+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
7771
+ neuronxcc-2.21.33363.0+82129205/MODULE_9295293099bd829ab6ce+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
7772
+ neuronxcc-2.21.33363.0+82129205/MODULE_cad34e8b42cec5152517+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
7773
+ neuronxcc-2.21.33363.0+82129205/MODULE_cad34e8b42cec5152517+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev3/527899cc3b15a98022089573f6e42212494e0dbda65770e25bddf8fe48c6bd4a/8f586dd55a9038556565.json ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "ibm-granite/granite-3.1-2b-base",
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": 4,
24
+ "capacity_factor": null,
25
+ "checkpoint_id": "ibm-granite/granite-3.1-2b-base",
26
+ "checkpoint_revision": "bcf4e33e7debde263041db41869f9cdbeaae3f01",
27
+ "continuous_batching": true,
28
+ "ep_degree": 1,
29
+ "fused_qkv": true,
30
+ "glu_mlp": true,
31
+ "local_ranks_size": 2,
32
+ "max_batch_size": 4,
33
+ "max_context_length": 1024,
34
+ "max_topk": 256,
35
+ "n_active_tokens": 1024,
36
+ "neuronxcc_version": "2.21.33363.0+82129205",
37
+ "on_device_sampling": true,
38
+ "optimum_neuron_version": "0.4.6.dev3",
39
+ "output_logits": false,
40
+ "pp_degree": 1,
41
+ "prefill_chunk_size": 0,
42
+ "sequence_length": 1024,
43
+ "speculation_length": 0,
44
+ "start_rank_id": 0,
45
+ "target": "trn1",
46
+ "torch_dtype": "bfloat16",
47
+ "tp_degree": 2
48
+ },
49
+ "num_attention_heads": 32,
50
+ "num_hidden_layers": 40,
51
+ "num_key_value_heads": 8,
52
+ "residual_multiplier": 0.22,
53
+ "rms_norm_eps": 1e-05,
54
+ "rope_scaling": null,
55
+ "rope_theta": 5000000.0,
56
+ "tie_word_embeddings": true,
57
+ "use_cache": true,
58
+ "vocab_size": 49152
59
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev3/granite/ibm-granite/granite-3.1-2b-base/8f586dd55a9038556565.json ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "ibm-granite/granite-3.1-2b-base",
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": 4,
24
+ "capacity_factor": null,
25
+ "checkpoint_id": "ibm-granite/granite-3.1-2b-base",
26
+ "checkpoint_revision": "bcf4e33e7debde263041db41869f9cdbeaae3f01",
27
+ "continuous_batching": true,
28
+ "ep_degree": 1,
29
+ "fused_qkv": true,
30
+ "glu_mlp": true,
31
+ "local_ranks_size": 2,
32
+ "max_batch_size": 4,
33
+ "max_context_length": 1024,
34
+ "max_topk": 256,
35
+ "n_active_tokens": 1024,
36
+ "neuronxcc_version": "2.21.33363.0+82129205",
37
+ "on_device_sampling": true,
38
+ "optimum_neuron_version": "0.4.6.dev3",
39
+ "output_logits": false,
40
+ "pp_degree": 1,
41
+ "prefill_chunk_size": 0,
42
+ "sequence_length": 1024,
43
+ "speculation_length": 0,
44
+ "start_rank_id": 0,
45
+ "target": "trn1",
46
+ "torch_dtype": "bfloat16",
47
+ "tp_degree": 2
48
+ },
49
+ "num_attention_heads": 32,
50
+ "num_hidden_layers": 40,
51
+ "num_key_value_heads": 8,
52
+ "residual_multiplier": 0.22,
53
+ "rms_norm_eps": 1e-05,
54
+ "rope_scaling": null,
55
+ "rope_theta": 5000000.0,
56
+ "tie_word_embeddings": true,
57
+ "use_cache": true,
58
+ "vocab_size": 49152
59
+ }
neuronxcc-2.21.33363.0+82129205/MODULE_17476209562158013765+e30acd3a/model.neff CHANGED
Binary files a/neuronxcc-2.21.33363.0+82129205/MODULE_17476209562158013765+e30acd3a/model.neff and b/neuronxcc-2.21.33363.0+82129205/MODULE_17476209562158013765+e30acd3a/model.neff differ
 
neuronxcc-2.21.33363.0+82129205/MODULE_8ff1216675c7144590c1+c4f887dc/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:936f7c19a4ff86454a5b83a26a3fb7b07cdcc62518970b6c78f0bf293cc21420
3
  size 30598144
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c7ff4622dc34bd380f361bf88e7c6e21a704134669270047adc1998da3107993
3
  size 30598144
neuronxcc-2.21.33363.0+82129205/MODULE_8ff1216675c7144590c1+c4f887dc/wrapped_neff.hlo CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c5009b951afa0dace7b72350a1ffd6ed43bfb3919fd73d44c0de35f45a53cc56
3
  size 30708309
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:22de134cd282af4925855d2ae8c1af6e8c9998d51db17f71bb976d60a34e050f
3
  size 30708309
neuronxcc-2.21.33363.0+82129205/MODULE_9295293099bd829ab6ce+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_9295293099bd829ab6ce+24129607/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_9295293099bd829ab6ce+24129607/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9704f2e00d1dc90b3a87ce5351efe1ee3b40e72f27ddce05d01537d9ae161edf
3
+ size 991695
neuronxcc-2.21.33363.0+82129205/MODULE_9295293099bd829ab6ce+24129607/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:41a44ba502fc49c9c967eab78f4ec8fa7292c9442e4c60f33dca7eb6db15f05c
3
+ size 14009344
neuronxcc-2.21.33363.0+82129205/MODULE_ad16674dd3fc5a51c226+a02c3a36/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2b8be74d754d3877ff4c9586c81611124a2989d55ff021b4915372155a27fe1a
3
  size 3073024
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2f7ccb3c563a4c0fdae234892df68dcd6c60cd41674bf983c7a5693709d2af9c
3
  size 3073024
neuronxcc-2.21.33363.0+82129205/MODULE_ad16674dd3fc5a51c226+a02c3a36/wrapped_neff.hlo CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a1a2dfe758abc52214d97cbae6fd8bfe738893d16164557e687784ac1c1a80a9
3
  size 3147125
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1a6ac7c0556c0e218f9e95cabc999fa5d73cb79aabcc96ccbc57e3c5325e36e9
3
  size 3147125
neuronxcc-2.21.33363.0+82129205/MODULE_cad34e8b42cec5152517+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_cad34e8b42cec5152517+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_cad34e8b42cec5152517+a02c3a36/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:82c64fe4d4e1bc1cf90c8058db57e4b168df1044efe2c2ddb2f07ecdb06446de
3
+ size 1058450
neuronxcc-2.21.33363.0+82129205/MODULE_cad34e8b42cec5152517+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:96adc0c81a43b990fcbf3eb850dff902f48a4bde546921283c8bd496dde2790c
3
+ size 5161984
neuronxcc-2.21.33363.0+82129205/MODULE_cad34e8b42cec5152517+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1fdc84793964d5c2f9d969354bdd1de5598fe610cdbb94af6725deb48be72503
3
+ size 5346517