Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
Community
645
dacorvo HF Staff commited on
Commit
4b701f0
·
verified ·
1 Parent(s): cde9e01

Synchronizing local compiler cache.

Browse files
Files changed (19) hide show
  1. .gitattributes +5 -0
  2. neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev6/qwen3_moe/Qwen/Qwen3-30B-A3B-Instruct-2507/99615150643a9e903722.json +66 -0
  3. neuronxcc-2.19.8089.0+8ab9f450/MODULE_8671664c95847cf8d9a7+cd3419b6/compile_flags.json +1 -0
  4. neuronxcc-2.19.8089.0+8ab9f450/MODULE_8671664c95847cf8d9a7+cd3419b6/model.done +0 -0
  5. neuronxcc-2.19.8089.0+8ab9f450/MODULE_8671664c95847cf8d9a7+cd3419b6/model.hlo_module.pb +3 -0
  6. neuronxcc-2.19.8089.0+8ab9f450/MODULE_8671664c95847cf8d9a7+cd3419b6/model.neff +3 -0
  7. neuronxcc-2.19.8089.0+8ab9f450/MODULE_8671664c95847cf8d9a7+cd3419b6/wrapped_neff.hlo +3 -0
  8. neuronxcc-2.19.8089.0+8ab9f450/MODULE_91ccf5a53294ba164897+cd3419b6/compile_flags.json +1 -0
  9. neuronxcc-2.19.8089.0+8ab9f450/MODULE_91ccf5a53294ba164897+cd3419b6/model.done +0 -0
  10. neuronxcc-2.19.8089.0+8ab9f450/MODULE_91ccf5a53294ba164897+cd3419b6/model.hlo_module.pb +3 -0
  11. neuronxcc-2.19.8089.0+8ab9f450/MODULE_91ccf5a53294ba164897+cd3419b6/model.neff +3 -0
  12. neuronxcc-2.19.8089.0+8ab9f450/MODULE_91ccf5a53294ba164897+cd3419b6/wrapped_neff.hlo +3 -0
  13. neuronxcc-2.19.8089.0+8ab9f450/MODULE_939df082e49d595a3158+253d6470/compile_flags.json +1 -0
  14. neuronxcc-2.19.8089.0+8ab9f450/MODULE_939df082e49d595a3158+253d6470/model.hlo_module.pb +3 -0
  15. neuronxcc-2.19.8089.0+8ab9f450/MODULE_939df082e49d595a3158+253d6470/model.log +3 -0
  16. neuronxcc-2.19.8089.0+8ab9f450/MODULE_dd67c7d76a4a3b7bc5ff+253d6470/compile_flags.json +1 -0
  17. neuronxcc-2.19.8089.0+8ab9f450/MODULE_dd67c7d76a4a3b7bc5ff+253d6470/model.done +0 -0
  18. neuronxcc-2.19.8089.0+8ab9f450/MODULE_dd67c7d76a4a3b7bc5ff+253d6470/model.hlo_module.pb +3 -0
  19. neuronxcc-2.19.8089.0+8ab9f450/MODULE_dd67c7d76a4a3b7bc5ff+253d6470/model.neff +3 -0
.gitattributes CHANGED
@@ -10952,3 +10952,8 @@ neuronxcc-2.19.8089.0+8ab9f450/MODULE_d0af99bdae3af4d95a22+a9d440f5/wrapped_neff
10952
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_e5e262df0f4573dfd367+253d6470/model.neff filter=lfs diff=lfs merge=lfs -text
10953
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_efa7fe965caa79aca41c+cd3419b6/model.neff filter=lfs diff=lfs merge=lfs -text
10954
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_efa7fe965caa79aca41c+cd3419b6/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
10952
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_e5e262df0f4573dfd367+253d6470/model.neff filter=lfs diff=lfs merge=lfs -text
10953
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_efa7fe965caa79aca41c+cd3419b6/model.neff filter=lfs diff=lfs merge=lfs -text
10954
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_efa7fe965caa79aca41c+cd3419b6/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
10955
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_8671664c95847cf8d9a7+cd3419b6/model.neff filter=lfs diff=lfs merge=lfs -text
10956
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_8671664c95847cf8d9a7+cd3419b6/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
10957
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_91ccf5a53294ba164897+cd3419b6/model.neff filter=lfs diff=lfs merge=lfs -text
10958
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_91ccf5a53294ba164897+cd3419b6/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
10959
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_dd67c7d76a4a3b7bc5ff+253d6470/model.neff filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev6/qwen3_moe/Qwen/Qwen3-30B-A3B-Instruct-2507/99615150643a9e903722.json ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen3-30B-A3B-Instruct-2507",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Qwen3MoeForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "decoder_sparse_step": 1,
11
+ "head_dim": 128,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 2048,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 6144,
16
+ "max_position_embeddings": 262144,
17
+ "max_window_layers": 48,
18
+ "mlp_only_layers": [],
19
+ "model_type": "qwen3_moe",
20
+ "moe_intermediate_size": 768,
21
+ "neuron": {
22
+ "_serialized_key": "NxDNeuronConfig",
23
+ "batch_size": 8,
24
+ "capacity_factor": null,
25
+ "checkpoint_id": "Qwen/Qwen3-30B-A3B-Instruct-2507",
26
+ "checkpoint_revision": "61082d4deaa4785f64943b443cbc2b5de7524fad",
27
+ "continuous_batching": true,
28
+ "enable_bucketing": false,
29
+ "ep_degree": 1,
30
+ "fused_qkv": false,
31
+ "glu_mlp": true,
32
+ "local_ranks_size": 16,
33
+ "logical_nc_config": 1,
34
+ "max_batch_size": 8,
35
+ "max_context_length": 4096,
36
+ "max_topk": 256,
37
+ "n_active_tokens": 4096,
38
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
39
+ "on_device_sampling": true,
40
+ "optimum_neuron_version": "0.3.1.dev6",
41
+ "output_logits": false,
42
+ "pp_degree": 1,
43
+ "sequence_length": 4096,
44
+ "speculation_length": 0,
45
+ "start_rank_id": 0,
46
+ "target": null,
47
+ "torch_dtype": "bfloat16",
48
+ "tp_degree": 16
49
+ },
50
+ "norm_topk_prob": true,
51
+ "num_attention_heads": 32,
52
+ "num_experts": 128,
53
+ "num_experts_per_tok": 8,
54
+ "num_hidden_layers": 48,
55
+ "num_key_value_heads": 4,
56
+ "output_router_logits": false,
57
+ "rms_norm_eps": 1e-06,
58
+ "rope_scaling": null,
59
+ "rope_theta": 10000000,
60
+ "router_aux_loss_coef": 0.001,
61
+ "sliding_window": null,
62
+ "tie_word_embeddings": false,
63
+ "use_cache": true,
64
+ "use_sliding_window": false,
65
+ "vocab_size": 151936
66
+ }
neuronxcc-2.19.8089.0+8ab9f450/MODULE_8671664c95847cf8d9a7+cd3419b6/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn1", "--enable-saturate-infinity", "--enable-mixed-precision-accumulation", "--model-type", "transformer", "-O1", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2", "--auto-cast=none", "--internal-enable-dge-levels", "vector_dynamic_offsets", "--internal-hlo2tensorizer-options=--verify-hlo=true", "--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.19.8089.0+8ab9f450/MODULE_8671664c95847cf8d9a7+cd3419b6/model.done ADDED
File without changes
neuronxcc-2.19.8089.0+8ab9f450/MODULE_8671664c95847cf8d9a7+cd3419b6/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:18e44bcca2d0392946b04d76ea0c08920b2977dd8ebbe62045de5b09475eb6ee
3
+ size 3545413
neuronxcc-2.19.8089.0+8ab9f450/MODULE_8671664c95847cf8d9a7+cd3419b6/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:256f25747a13b5e7d2d90243de1bed6f01d5792cd4a0597b5a4816d086a85af6
3
+ size 7035904
neuronxcc-2.19.8089.0+8ab9f450/MODULE_8671664c95847cf8d9a7+cd3419b6/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8e2c5a5f4491cfbd43ac527c4b0bd3a0b06581d2b6f4dc4cf9ab1700ec82f2f0
3
+ size 7287366
neuronxcc-2.19.8089.0+8ab9f450/MODULE_91ccf5a53294ba164897+cd3419b6/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn1", "--enable-saturate-infinity", "--enable-mixed-precision-accumulation", "--model-type", "transformer", "-O1", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2", "--auto-cast=none", "--internal-enable-dge-levels", "vector_dynamic_offsets", "--internal-hlo2tensorizer-options=--verify-hlo=true", "--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.19.8089.0+8ab9f450/MODULE_91ccf5a53294ba164897+cd3419b6/model.done ADDED
File without changes
neuronxcc-2.19.8089.0+8ab9f450/MODULE_91ccf5a53294ba164897+cd3419b6/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1baf13daf477f131f4a0605183ccd7855c12e7fad5122e3634807a697266aff3
3
+ size 3548477
neuronxcc-2.19.8089.0+8ab9f450/MODULE_91ccf5a53294ba164897+cd3419b6/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d430973d6a80a63feebf4521fa290aa1860acae23fb344a68c94f6390b9c13a5
3
+ size 6083584
neuronxcc-2.19.8089.0+8ab9f450/MODULE_91ccf5a53294ba164897+cd3419b6/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d3b708e8d335217684701b929f706c8d5c0c7a625f4027602d07a952a038d2dc
3
+ size 6333655
neuronxcc-2.19.8089.0+8ab9f450/MODULE_939df082e49d595a3158+253d6470/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn1", "--enable-saturate-infinity", "--enable-mixed-precision-accumulation", "--model-type", "transformer", "-O1", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2", "--auto-cast=none", "--internal-enable-dge-levels", "vector_dynamic_offsets", "--internal-hlo2tensorizer-options=--verify-hlo=true", "--logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.19.8089.0+8ab9f450/MODULE_939df082e49d595a3158+253d6470/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:76ac335630a721f2e3de37d63a053c4199b2c39ad751a52b3e7c05ee260d9efc
3
+ size 9665709
neuronxcc-2.19.8089.0+8ab9f450/MODULE_939df082e49d595a3158+253d6470/model.log ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ Failed compilation with ['neuronx-cc', 'compile', '--framework=XLA', '/tmp/nxd_model/context_encoding_model/_tp0_bk0/model.MODULE_939df082e49d595a3158+253d6470.hlo_module.pb', '--output', '/tmp/nxd_model/context_encoding_model/_tp0_bk0/model.MODULE_939df082e49d595a3158+253d6470.neff', '--target=trn1', '--enable-saturate-infinity', '--enable-mixed-precision-accumulation', '--model-type', 'transformer', '-O1', '--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2', '--auto-cast=none', '--internal-enable-dge-levels', 'vector_dynamic_offsets', '--internal-hlo2tensorizer-options=--verify-hlo=true', '--logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt', '--verbose=35']: [XCG815] Estimated peak HBM usage (16.329037) exceeds 16GB. Neff won't be able to load on chip - Please open a support ticket at https://github.com/aws-neuron/aws-neuron-sdk/issues/new. You may also be able to obtain more information using the 'XLA_IR_DEBUG' and 'XLA_HLO_DEBUG' environment variables.
2
+ 2025-09-11T15:00:22Z Non-signal exit. Backend exited with code 1 and stderr: [XCG815] Estimated peak HBM usage (16.329037) exceeds 16GB. Neff won't be able to load on chip - Please open a support ticket at https://github.com/aws-neuron/aws-neuron-sdk/issues/new. You may also be able to obtain more information using the 'XLA_IR_DEBUG' and 'XLA_HLO_DEBUG' environment variables.
3
+
neuronxcc-2.19.8089.0+8ab9f450/MODULE_dd67c7d76a4a3b7bc5ff+253d6470/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn1", "--enable-saturate-infinity", "--enable-mixed-precision-accumulation", "--model-type", "transformer", "-O1", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2", "--auto-cast=none", "--internal-enable-dge-levels", "vector_dynamic_offsets", "--internal-hlo2tensorizer-options=--verify-hlo=true", "--logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.19.8089.0+8ab9f450/MODULE_dd67c7d76a4a3b7bc5ff+253d6470/model.done ADDED
File without changes
neuronxcc-2.19.8089.0+8ab9f450/MODULE_dd67c7d76a4a3b7bc5ff+253d6470/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6fbd912d9b5ddd548a60cd6f06caf6fab6c360df519845d87f637cb6c1b6e724
3
+ size 9670898
neuronxcc-2.19.8089.0+8ab9f450/MODULE_dd67c7d76a4a3b7bc5ff+253d6470/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:abb3918c4f75e7e4dd978ff5a82d00e9547358f20ac608a6fd40853297acca6a
3
+ size 17439744