Hugging Face's logo

optimum-internal-testing
/
neuron-testing-cache

Model card Files
xet
Community
dacorvo HF Staff commited on
Commit
5974006
·
verified ·
1 Parent(s): 2595bac

Synchronizing local compiler cache.

Browse files
Files changed (23) hide show
  1. .gitattributes +4 -0
  2. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.3.0.dev1/granite/ibm-granite/granite-3.1-2b-instruct/6c659b1c4f864a345f17.json +73 -0
  3. neuronxcc-2.17.194.0+d312836f/MODULE_3fd5107b3bbf73dca906+165e9558/compile_flags.json +1 -0
  4. neuronxcc-2.17.194.0+d312836f/MODULE_3fd5107b3bbf73dca906+165e9558/model.hlo_module.pb +3 -0
  5. neuronxcc-2.17.194.0+d312836f/MODULE_4a2733ad0c48499b3cb1+431f5505/compile_flags.json +1 -0
  6. neuronxcc-2.17.194.0+d312836f/MODULE_4a2733ad0c48499b3cb1+431f5505/model.done +0 -0
  7. neuronxcc-2.17.194.0+d312836f/MODULE_4a2733ad0c48499b3cb1+431f5505/model.hlo_module.pb +3 -0
  8. neuronxcc-2.17.194.0+d312836f/MODULE_4a2733ad0c48499b3cb1+431f5505/model.neff +3 -0
  9. neuronxcc-2.17.194.0+d312836f/MODULE_5e3af8e039ed38dc1451+bfe5714b/compile_flags.json +1 -0
  10. neuronxcc-2.17.194.0+d312836f/MODULE_5e3af8e039ed38dc1451+bfe5714b/model.done +0 -0
  11. neuronxcc-2.17.194.0+d312836f/MODULE_5e3af8e039ed38dc1451+bfe5714b/model.hlo_module.pb +3 -0
  12. neuronxcc-2.17.194.0+d312836f/MODULE_5e3af8e039ed38dc1451+bfe5714b/model.neff +3 -0
  13. neuronxcc-2.17.194.0+d312836f/MODULE_5ee958efa9d808fa7242+165e9558/model.hlo_module.pb +1 -1
  14. neuronxcc-2.17.194.0+d312836f/MODULE_7d97d3b9bee047397c87+165e9558/compile_flags.json +1 -0
  15. neuronxcc-2.17.194.0+d312836f/MODULE_7d97d3b9bee047397c87+165e9558/model.done +0 -0
  16. neuronxcc-2.17.194.0+d312836f/MODULE_7d97d3b9bee047397c87+165e9558/model.hlo_module.pb +3 -0
  17. neuronxcc-2.17.194.0+d312836f/MODULE_7d97d3b9bee047397c87+165e9558/model.neff +3 -0
  18. neuronxcc-2.17.194.0+d312836f/MODULE_7d97d3b9bee047397c87+165e9558/wrapped_neff.hlo +3 -0
  19. neuronxcc-2.17.194.0+d312836f/MODULE_b04ad5c157e55310df86+165e9558/compile_flags.json +1 -0
  20. neuronxcc-2.17.194.0+d312836f/MODULE_b04ad5c157e55310df86+165e9558/model.hlo_module.pb +3 -0
  21. neuronxcc-2.17.194.0+d312836f/MODULE_d96567e523bd6df1ddcb+165e9558/compile_flags.json +1 -0
  22. neuronxcc-2.17.194.0+d312836f/MODULE_d96567e523bd6df1ddcb+165e9558/model.hlo_module.pb +3 -0
  23. neuronxcc-2.17.194.0+d312836f/MODULE_e67996812ad4b0713c1f+165e9558/model.hlo_module.pb +1 -1
.gitattributes CHANGED
@@ -2998,3 +2998,7 @@ neuronxcc-2.17.194.0+d312836f/MODULE_389d56c4ece62e7f27bf+431f5505/model.neff fi
2998
  neuronxcc-2.17.194.0+d312836f/MODULE_6eaa88a4e11b815b0091+bfe5714b/model.neff filter=lfs diff=lfs merge=lfs -text
2999
  neuronxcc-2.17.194.0+d312836f/MODULE_801af15b5001c885c608+165e9558/model.neff filter=lfs diff=lfs merge=lfs -text
3000
  neuronxcc-2.17.194.0+d312836f/MODULE_801af15b5001c885c608+165e9558/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
2998
  neuronxcc-2.17.194.0+d312836f/MODULE_6eaa88a4e11b815b0091+bfe5714b/model.neff filter=lfs diff=lfs merge=lfs -text
2999
  neuronxcc-2.17.194.0+d312836f/MODULE_801af15b5001c885c608+165e9558/model.neff filter=lfs diff=lfs merge=lfs -text
3000
  neuronxcc-2.17.194.0+d312836f/MODULE_801af15b5001c885c608+165e9558/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
3001
+ neuronxcc-2.17.194.0+d312836f/MODULE_4a2733ad0c48499b3cb1+431f5505/model.neff filter=lfs diff=lfs merge=lfs -text
3002
+ neuronxcc-2.17.194.0+d312836f/MODULE_5e3af8e039ed38dc1451+bfe5714b/model.neff filter=lfs diff=lfs merge=lfs -text
3003
+ neuronxcc-2.17.194.0+d312836f/MODULE_7d97d3b9bee047397c87+165e9558/model.neff filter=lfs diff=lfs merge=lfs -text
3004
+ neuronxcc-2.17.194.0+d312836f/MODULE_7d97d3b9bee047397c87+165e9558/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.3.0.dev1/granite/ibm-granite/granite-3.1-2b-instruct/6c659b1c4f864a345f17.json ADDED
@@ -0,0 +1,73 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "embedding_multiplier": 12.0,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 2048,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 8192,
16
+ "logits_scaling": 8.0,
17
+ "max_position_embeddings": 131072,
18
+ "mlp_bias": false,
19
+ "model_type": "granite",
20
+ "neuron": {
21
+ "_serialized_key": "NxDNeuronConfig",
22
+ "async_mode": false,
23
+ "attn_kernel_enabled": false,
24
+ "batch_size": 4,
25
+ "capacity_factor": null,
26
+ "cc_pipeline_tiling_factor": 2,
27
+ "checkpoint_id": "ibm-granite/granite-3.1-2b-instruct",
28
+ "checkpoint_revision": "bbc2aed595bd38bd770263dc3ab831db9794441d",
29
+ "continuous_batching": true,
30
+ "enable_bucketing": false,
31
+ "ep_degree": 1,
32
+ "flash_decoding_enabled": false,
33
+ "fused_qkv": true,
34
+ "glu_mlp": true,
35
+ "is_chunked_prefill": false,
36
+ "local_ranks_size": 2,
37
+ "logical_nc_config": 1,
38
+ "max_batch_size": 4,
39
+ "max_context_length": 4096,
40
+ "max_topk": 256,
41
+ "mlp_kernel_enabled": false,
42
+ "mlp_kernel_fuse_residual_add": false,
43
+ "n_active_tokens": 4096,
44
+ "neuronxcc_version": "2.17.194.0+d312836f",
45
+ "num_cores_per_group": 1,
46
+ "on_device_sampling": false,
47
+ "optimum_neuron_version": "0.3.0.dev1",
48
+ "output_logits": false,
49
+ "padding_side": "right",
50
+ "pp_degree": 1,
51
+ "qk_layernorm": false,
52
+ "qkv_kernel_enabled": false,
53
+ "rpl_reduce_dtype": "bfloat16",
54
+ "sequence_length": 4096,
55
+ "sequence_parallel_enabled": false,
56
+ "speculation_length": 0,
57
+ "start_rank_id": 0,
58
+ "target": null,
59
+ "torch_dtype": "bfloat16",
60
+ "tp_degree": 2,
61
+ "vocab_parallel": false
62
+ },
63
+ "num_attention_heads": 32,
64
+ "num_hidden_layers": 40,
65
+ "num_key_value_heads": 8,
66
+ "residual_multiplier": 0.22,
67
+ "rms_norm_eps": 1e-05,
68
+ "rope_scaling": null,
69
+ "rope_theta": 5000000.0,
70
+ "tie_word_embeddings": true,
71
+ "use_cache": true,
72
+ "vocab_size": 49155
73
+ }
neuronxcc-2.17.194.0+d312836f/MODULE_3fd5107b3bbf73dca906+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_3fd5107b3bbf73dca906+165e9558/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4da71cd8c2d73b196d80edfd3a059991fe59320506d544bab5f7c789c0ce46b3
3
+ size 2550812
neuronxcc-2.17.194.0+d312836f/MODULE_4a2733ad0c48499b3cb1+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_4a2733ad0c48499b3cb1+431f5505/model.done ADDED
File without changes
neuronxcc-2.17.194.0+d312836f/MODULE_4a2733ad0c48499b3cb1+431f5505/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c6c331526b8f81328de18d402f353bb1407e0e23fd156d22c02bfb034a6e206b
3
+ size 172368
neuronxcc-2.17.194.0+d312836f/MODULE_4a2733ad0c48499b3cb1+431f5505/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c003282b2e3e4ae95ae4b5cd9d54e32380fc65dce9bfe986514fe59cef7835c3
3
+ size 2243584
neuronxcc-2.17.194.0+d312836f/MODULE_5e3af8e039ed38dc1451+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_5e3af8e039ed38dc1451+bfe5714b/model.done ADDED
File without changes
neuronxcc-2.17.194.0+d312836f/MODULE_5e3af8e039ed38dc1451+bfe5714b/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:603218278a42c788949131e26b851876358cc9ea52800c5c25e2ac44f5c507c3
3
+ size 2210522
neuronxcc-2.17.194.0+d312836f/MODULE_5e3af8e039ed38dc1451+bfe5714b/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:495da33df4a2e3e0917e8956f17b6483ae15866e56a37dfac502145856c9eef4
3
+ size 6319104
neuronxcc-2.17.194.0+d312836f/MODULE_5ee958efa9d808fa7242+165e9558/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:141fc96d1818343609c0bf818dd0a83270912fd4483455868a9bab88d2d6ae7e
3
  size 966389
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dd1fcf86d544f20a6b87ccb6a53143af5126e05c2711a0b999a3e53a94fba9c5
3
  size 966389
neuronxcc-2.17.194.0+d312836f/MODULE_7d97d3b9bee047397c87+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_7d97d3b9bee047397c87+165e9558/model.done ADDED
File without changes
neuronxcc-2.17.194.0+d312836f/MODULE_7d97d3b9bee047397c87+165e9558/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cab022b86f9fcc17f6f5257ee73c050070878f0ff17329d5f8568376524b746b
3
+ size 2547878
neuronxcc-2.17.194.0+d312836f/MODULE_7d97d3b9bee047397c87+165e9558/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7c030b0cc1aa42b6fcce4a6b7d27ff80fca41e934c4bcb5d1711edcb917a8eb0
3
+ size 6974464
neuronxcc-2.17.194.0+d312836f/MODULE_7d97d3b9bee047397c87+165e9558/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:78ecc5b043b769824c2ba802e802cca42019fec94cfc027d6923eeaa7aebf5a3
3
+ size 7147636
neuronxcc-2.17.194.0+d312836f/MODULE_b04ad5c157e55310df86+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_b04ad5c157e55310df86+165e9558/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:85d6e3af8f28f0c6f204be1ab259693c263e8f6a81d3ea4e5c5fe980bb5a6d8e
3
+ size 2494947
neuronxcc-2.17.194.0+d312836f/MODULE_d96567e523bd6df1ddcb+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_d96567e523bd6df1ddcb+165e9558/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6992afc1a115ca412a632149c9f09bea1fb5b3e20c4993eff598624602ca2551
3
+ size 2403893
neuronxcc-2.17.194.0+d312836f/MODULE_e67996812ad4b0713c1f+165e9558/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2ebac139feddf2b2fe04c45c67d1fe0dd9bc01bbd9f2fa32f72e13ca22cf3b72
3
  size 2413541
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8e5ec3336a02acfb3c655412a507d92dd3d62c984858fbb2b345b7d0687450e5
3
  size 2413541