Hugging Face's logo

optimum-internal-testing
/
neuron-testing-cache

Model card Files
xet
 Community
tengomucho HF Staff commited on
Commit
53049e1
·
verified ·
1 Parent(s): fa4e01e

Synchronizing local compiler cache.

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. .gitattributes +33 -0
  2. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev4/bd8b5aaeb4693c199f97f6eebd1bbca77255bea6e0ed042c0c592ee5adccec97/0538515e2eab8cf1a81f.json +81 -0
  3. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev4/bd8b5aaeb4693c199f97f6eebd1bbca77255bea6e0ed042c0c592ee5adccec97/18421eedc4d9b7b7fb51.json +81 -0
  4. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev4/bd8b5aaeb4693c199f97f6eebd1bbca77255bea6e0ed042c0c592ee5adccec97/5756ac9be9333f9c8d82.json +81 -0
  5. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev4/bd8b5aaeb4693c199f97f6eebd1bbca77255bea6e0ed042c0c592ee5adccec97/d3503138cf92a3bb6ba7.json +81 -0
  6. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev4/bd8b5aaeb4693c199f97f6eebd1bbca77255bea6e0ed042c0c592ee5adccec97/f7dfefb13d09e6187d80.json +81 -0
  7. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev4/d02c07e1054a0d66fe47bc2bd8603a512b513ad501882b6d49d36b9747c64a47/00688423de1428d98e68.json +91 -0
  8. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev4/d02c07e1054a0d66fe47bc2bd8603a512b513ad501882b6d49d36b9747c64a47/87e6309b5c6fe024caaa.json +91 -0
  9. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev4/d02c07e1054a0d66fe47bc2bd8603a512b513ad501882b6d49d36b9747c64a47/8eff71913f842fa8f404.json +91 -0
  10. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev4/d02c07e1054a0d66fe47bc2bd8603a512b513ad501882b6d49d36b9747c64a47/b8db5a05f548780ebfae.json +91 -0
  11. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev4/d02c07e1054a0d66fe47bc2bd8603a512b513ad501882b6d49d36b9747c64a47/dfcd91ed17670ed71f9d.json +91 -0
  12. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev4/d02c07e1054a0d66fe47bc2bd8603a512b513ad501882b6d49d36b9747c64a47/ed4c5924c8e305088a85.json +91 -0
  13. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev4/gemma3_text/google/gemma-3-1b-it/b8db5a05f548780ebfae.json +91 -0
  14. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev4/gemma3_text/google/gemma-3-270m-it/18421eedc4d9b7b7fb51.json +81 -0
  15. neuronxcc-2.21.33363.0+82129205/MODULE_083c75747563fca496d7+24129607/compile_flags.json +1 -0
  16. neuronxcc-2.21.33363.0+82129205/MODULE_083c75747563fca496d7+24129607/model.done +0 -0
  17. neuronxcc-2.21.33363.0+82129205/MODULE_083c75747563fca496d7+24129607/model.hlo_module.pb +3 -0
  18. neuronxcc-2.21.33363.0+82129205/MODULE_083c75747563fca496d7+24129607/model.neff +3 -0
  19. neuronxcc-2.21.33363.0+82129205/MODULE_134003a684b11c49131d+24129607/compile_flags.json +1 -0
  20. neuronxcc-2.21.33363.0+82129205/MODULE_134003a684b11c49131d+24129607/model.done +0 -0
  21. neuronxcc-2.21.33363.0+82129205/MODULE_134003a684b11c49131d+24129607/model.hlo_module.pb +3 -0
  22. neuronxcc-2.21.33363.0+82129205/MODULE_134003a684b11c49131d+24129607/model.neff +3 -0
  23. neuronxcc-2.21.33363.0+82129205/MODULE_1b9cad3a2eb3c406661d+a02c3a36/compile_flags.json +1 -0
  24. neuronxcc-2.21.33363.0+82129205/MODULE_1b9cad3a2eb3c406661d+a02c3a36/model.done +0 -0
  25. neuronxcc-2.21.33363.0+82129205/MODULE_1b9cad3a2eb3c406661d+a02c3a36/model.hlo_module.pb +3 -0
  26. neuronxcc-2.21.33363.0+82129205/MODULE_1b9cad3a2eb3c406661d+a02c3a36/model.neff +3 -0
  27. neuronxcc-2.21.33363.0+82129205/MODULE_1b9cad3a2eb3c406661d+a02c3a36/wrapped_neff.hlo +3 -0
  28. neuronxcc-2.21.33363.0+82129205/MODULE_33dcb1875eda19b3ebb5+24129607/compile_flags.json +1 -0
  29. neuronxcc-2.21.33363.0+82129205/MODULE_33dcb1875eda19b3ebb5+24129607/model.done +0 -0
  30. neuronxcc-2.21.33363.0+82129205/MODULE_33dcb1875eda19b3ebb5+24129607/model.hlo_module.pb +3 -0
  31. neuronxcc-2.21.33363.0+82129205/MODULE_33dcb1875eda19b3ebb5+24129607/model.neff +3 -0
  32. neuronxcc-2.21.33363.0+82129205/MODULE_4978e0dffee300919f4f+24129607/compile_flags.json +1 -0
  33. neuronxcc-2.21.33363.0+82129205/MODULE_4978e0dffee300919f4f+24129607/model.done +0 -0
  34. neuronxcc-2.21.33363.0+82129205/MODULE_4978e0dffee300919f4f+24129607/model.hlo_module.pb +3 -0
  35. neuronxcc-2.21.33363.0+82129205/MODULE_4978e0dffee300919f4f+24129607/model.neff +3 -0
  36. neuronxcc-2.21.33363.0+82129205/MODULE_4dd21482986783a76fee+24129607/compile_flags.json +1 -0
  37. neuronxcc-2.21.33363.0+82129205/MODULE_4dd21482986783a76fee+24129607/model.done +0 -0
  38. neuronxcc-2.21.33363.0+82129205/MODULE_4dd21482986783a76fee+24129607/model.hlo_module.pb +3 -0
  39. neuronxcc-2.21.33363.0+82129205/MODULE_4dd21482986783a76fee+24129607/model.neff +3 -0
  40. neuronxcc-2.21.33363.0+82129205/MODULE_61998e972adde9b6f5d2+24129607/compile_flags.json +1 -0
  41. neuronxcc-2.21.33363.0+82129205/MODULE_61998e972adde9b6f5d2+24129607/model.done +0 -0
  42. neuronxcc-2.21.33363.0+82129205/MODULE_61998e972adde9b6f5d2+24129607/model.hlo_module.pb +3 -0
  43. neuronxcc-2.21.33363.0+82129205/MODULE_61998e972adde9b6f5d2+24129607/model.neff +3 -0
  44. neuronxcc-2.21.33363.0+82129205/MODULE_7701466c22adfdd959a6+a02c3a36/compile_flags.json +1 -0
  45. neuronxcc-2.21.33363.0+82129205/MODULE_7701466c22adfdd959a6+a02c3a36/model.done +0 -0
  46. neuronxcc-2.21.33363.0+82129205/MODULE_7701466c22adfdd959a6+a02c3a36/model.hlo_module.pb +3 -0
  47. neuronxcc-2.21.33363.0+82129205/MODULE_7701466c22adfdd959a6+a02c3a36/model.neff +3 -0
  48. neuronxcc-2.21.33363.0+82129205/MODULE_7701466c22adfdd959a6+a02c3a36/wrapped_neff.hlo +3 -0
  49. neuronxcc-2.21.33363.0+82129205/MODULE_8b7c51b3642e17a79f64+24129607/compile_flags.json +1 -0
  50. neuronxcc-2.21.33363.0+82129205/MODULE_8b7c51b3642e17a79f64+24129607/model.done +0 -0
.gitattributes CHANGED
@@ -7824,3 +7824,36 @@ neuronxcc-2.23.6484.0+3b612583/MODULE_c0822879a9404f208f02+84337dd9/model.neff f
7824
  neuronxcc-2.23.6484.0+3b612583/MODULE_c0822879a9404f208f02+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
7825
  neuronxcc-2.23.6484.0+3b612583/MODULE_b202b8e3d4558ac1c782+97496b47/model.neff filter=lfs diff=lfs merge=lfs -text
7826
  neuronxcc-2.23.6484.0+3b612583/MODULE_2268c0503ccbf4859036+97496b47/model.neff filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
7824
  neuronxcc-2.23.6484.0+3b612583/MODULE_c0822879a9404f208f02+84337dd9/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
7825
  neuronxcc-2.23.6484.0+3b612583/MODULE_b202b8e3d4558ac1c782+97496b47/model.neff filter=lfs diff=lfs merge=lfs -text
7826
  neuronxcc-2.23.6484.0+3b612583/MODULE_2268c0503ccbf4859036+97496b47/model.neff filter=lfs diff=lfs merge=lfs -text
7827
+ neuronxcc-2.21.33363.0+82129205/MODULE_083c75747563fca496d7+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
7828
+ neuronxcc-2.21.33363.0+82129205/MODULE_134003a684b11c49131d+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
7829
+ neuronxcc-2.21.33363.0+82129205/MODULE_1b9cad3a2eb3c406661d+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
7830
+ neuronxcc-2.21.33363.0+82129205/MODULE_1b9cad3a2eb3c406661d+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
7831
+ neuronxcc-2.21.33363.0+82129205/MODULE_33dcb1875eda19b3ebb5+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
7832
+ neuronxcc-2.21.33363.0+82129205/MODULE_4978e0dffee300919f4f+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
7833
+ neuronxcc-2.21.33363.0+82129205/MODULE_4dd21482986783a76fee+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
7834
+ neuronxcc-2.21.33363.0+82129205/MODULE_61998e972adde9b6f5d2+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
7835
+ neuronxcc-2.21.33363.0+82129205/MODULE_7701466c22adfdd959a6+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
7836
+ neuronxcc-2.21.33363.0+82129205/MODULE_7701466c22adfdd959a6+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
7837
+ neuronxcc-2.21.33363.0+82129205/MODULE_8b7c51b3642e17a79f64+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
7838
+ neuronxcc-2.21.33363.0+82129205/MODULE_8ff80109ec96db204c79+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
7839
+ neuronxcc-2.21.33363.0+82129205/MODULE_8ff80109ec96db204c79+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
7840
+ neuronxcc-2.21.33363.0+82129205/MODULE_a36fd4fef78c745e1416+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
7841
+ neuronxcc-2.21.33363.0+82129205/MODULE_a36fd4fef78c745e1416+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
7842
+ neuronxcc-2.21.33363.0+82129205/MODULE_ac3c7cc1cd56727823f0+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
7843
+ neuronxcc-2.21.33363.0+82129205/MODULE_ac3c7cc1cd56727823f0+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
7844
+ neuronxcc-2.21.33363.0+82129205/MODULE_b1639d901305de0ccbb9+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
7845
+ neuronxcc-2.21.33363.0+82129205/MODULE_b406585baf2c99ea743d+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
7846
+ neuronxcc-2.21.33363.0+82129205/MODULE_b406585baf2c99ea743d+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
7847
+ neuronxcc-2.21.33363.0+82129205/MODULE_b96a1f42beeb09ee40ca+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
7848
+ neuronxcc-2.21.33363.0+82129205/MODULE_c49ef4b16e6a8513ee7c+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
7849
+ neuronxcc-2.21.33363.0+82129205/MODULE_c49ef4b16e6a8513ee7c+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
7850
+ neuronxcc-2.21.33363.0+82129205/MODULE_d986d392e2b714a772e0+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
7851
+ neuronxcc-2.21.33363.0+82129205/MODULE_da330f1b1aad44160016+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
7852
+ neuronxcc-2.21.33363.0+82129205/MODULE_dba668a28cef64f8b275+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
7853
+ neuronxcc-2.21.33363.0+82129205/MODULE_dba668a28cef64f8b275+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
7854
+ neuronxcc-2.21.33363.0+82129205/MODULE_e21228aada91cc0cee76+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
7855
+ neuronxcc-2.21.33363.0+82129205/MODULE_e21228aada91cc0cee76+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
7856
+ neuronxcc-2.21.33363.0+82129205/MODULE_eaa8316a0cd427639f95+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
7857
+ neuronxcc-2.21.33363.0+82129205/MODULE_eaa8316a0cd427639f95+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
7858
+ neuronxcc-2.21.33363.0+82129205/MODULE_ee2fac66603ba635ef0e+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
7859
+ neuronxcc-2.21.33363.0+82129205/MODULE_ee2fac66603ba635ef0e+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev4/bd8b5aaeb4693c199f97f6eebd1bbca77255bea6e0ed042c0c592ee5adccec97/0538515e2eab8cf1a81f.json ADDED
@@ -0,0 +1,81 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "google/gemma-3-270m-it",
4
+ "_sliding_window_pattern": 6,
5
+ "_task": "text-generation",
6
+ "architectures": [
7
+ "Gemma3ForCausalLM"
8
+ ],
9
+ "attention_bias": false,
10
+ "attention_dropout": 0.0,
11
+ "attn_logit_softcapping": null,
12
+ "dtype": "bfloat16",
13
+ "final_logit_softcapping": null,
14
+ "head_dim": 256,
15
+ "hidden_activation": "gelu_pytorch_tanh",
16
+ "hidden_size": 640,
17
+ "initializer_range": 0.02,
18
+ "intermediate_size": 2048,
19
+ "layer_types": [
20
+ "sliding_attention",
21
+ "sliding_attention",
22
+ "sliding_attention",
23
+ "sliding_attention",
24
+ "sliding_attention",
25
+ "full_attention",
26
+ "sliding_attention",
27
+ "sliding_attention",
28
+ "sliding_attention",
29
+ "sliding_attention",
30
+ "sliding_attention",
31
+ "full_attention",
32
+ "sliding_attention",
33
+ "sliding_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "sliding_attention",
37
+ "full_attention"
38
+ ],
39
+ "max_position_embeddings": 32768,
40
+ "model_type": "gemma3_text",
41
+ "neuron": {
42
+ "_serialized_key": "NxDNeuronConfig",
43
+ "batch_size": 16,
44
+ "capacity_factor": null,
45
+ "checkpoint_id": "google/gemma-3-270m-it",
46
+ "checkpoint_revision": "ac82b4e820549b854eebf28ce6dedaf9fdfa17b3",
47
+ "continuous_batching": true,
48
+ "ep_degree": 1,
49
+ "fused_qkv": true,
50
+ "glu_mlp": true,
51
+ "local_ranks_size": 2,
52
+ "max_batch_size": 16,
53
+ "max_context_length": 32768,
54
+ "max_topk": 256,
55
+ "n_active_tokens": 32768,
56
+ "neuronxcc_version": "2.21.33363.0+82129205",
57
+ "on_device_sampling": true,
58
+ "optimum_neuron_version": "0.4.6.dev4",
59
+ "output_logits": false,
60
+ "pp_degree": 1,
61
+ "prefill_chunk_size": 0,
62
+ "sequence_length": 32768,
63
+ "speculation_length": 0,
64
+ "start_rank_id": 0,
65
+ "target": "trn1",
66
+ "torch_dtype": "bfloat16",
67
+ "tp_degree": 2
68
+ },
69
+ "num_attention_heads": 4,
70
+ "num_hidden_layers": 18,
71
+ "num_key_value_heads": 1,
72
+ "query_pre_attn_scalar": 256,
73
+ "rms_norm_eps": 1e-06,
74
+ "rope_local_base_freq": 10000.0,
75
+ "rope_scaling": null,
76
+ "rope_theta": 1000000.0,
77
+ "sliding_window": 512,
78
+ "use_bidirectional_attention": false,
79
+ "use_cache": true,
80
+ "vocab_size": 262144
81
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev4/bd8b5aaeb4693c199f97f6eebd1bbca77255bea6e0ed042c0c592ee5adccec97/18421eedc4d9b7b7fb51.json ADDED
@@ -0,0 +1,81 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "google/gemma-3-270m-it",
4
+ "_sliding_window_pattern": 6,
5
+ "_task": "text-generation",
6
+ "architectures": [
7
+ "Gemma3ForCausalLM"
8
+ ],
9
+ "attention_bias": false,
10
+ "attention_dropout": 0.0,
11
+ "attn_logit_softcapping": null,
12
+ "dtype": "bfloat16",
13
+ "final_logit_softcapping": null,
14
+ "head_dim": 256,
15
+ "hidden_activation": "gelu_pytorch_tanh",
16
+ "hidden_size": 640,
17
+ "initializer_range": 0.02,
18
+ "intermediate_size": 2048,
19
+ "layer_types": [
20
+ "sliding_attention",
21
+ "sliding_attention",
22
+ "sliding_attention",
23
+ "sliding_attention",
24
+ "sliding_attention",
25
+ "full_attention",
26
+ "sliding_attention",
27
+ "sliding_attention",
28
+ "sliding_attention",
29
+ "sliding_attention",
30
+ "sliding_attention",
31
+ "full_attention",
32
+ "sliding_attention",
33
+ "sliding_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "sliding_attention",
37
+ "full_attention"
38
+ ],
39
+ "max_position_embeddings": 32768,
40
+ "model_type": "gemma3_text",
41
+ "neuron": {
42
+ "_serialized_key": "NxDNeuronConfig",
43
+ "batch_size": 32,
44
+ "capacity_factor": null,
45
+ "checkpoint_id": "google/gemma-3-270m-it",
46
+ "checkpoint_revision": "ac82b4e820549b854eebf28ce6dedaf9fdfa17b3",
47
+ "continuous_batching": true,
48
+ "ep_degree": 1,
49
+ "fused_qkv": true,
50
+ "glu_mlp": true,
51
+ "local_ranks_size": 2,
52
+ "max_batch_size": 32,
53
+ "max_context_length": 16384,
54
+ "max_topk": 256,
55
+ "n_active_tokens": 16384,
56
+ "neuronxcc_version": "2.21.33363.0+82129205",
57
+ "on_device_sampling": true,
58
+ "optimum_neuron_version": "0.4.6.dev4",
59
+ "output_logits": false,
60
+ "pp_degree": 1,
61
+ "prefill_chunk_size": 0,
62
+ "sequence_length": 16384,
63
+ "speculation_length": 0,
64
+ "start_rank_id": 0,
65
+ "target": "trn1",
66
+ "torch_dtype": "bfloat16",
67
+ "tp_degree": 2
68
+ },
69
+ "num_attention_heads": 4,
70
+ "num_hidden_layers": 18,
71
+ "num_key_value_heads": 1,
72
+ "query_pre_attn_scalar": 256,
73
+ "rms_norm_eps": 1e-06,
74
+ "rope_local_base_freq": 10000.0,
75
+ "rope_scaling": null,
76
+ "rope_theta": 1000000.0,
77
+ "sliding_window": 512,
78
+ "use_bidirectional_attention": false,
79
+ "use_cache": true,
80
+ "vocab_size": 262144
81
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev4/bd8b5aaeb4693c199f97f6eebd1bbca77255bea6e0ed042c0c592ee5adccec97/5756ac9be9333f9c8d82.json ADDED
@@ -0,0 +1,81 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "google/gemma-3-270m-it",
4
+ "_sliding_window_pattern": 6,
5
+ "_task": "text-generation",
6
+ "architectures": [
7
+ "Gemma3ForCausalLM"
8
+ ],
9
+ "attention_bias": false,
10
+ "attention_dropout": 0.0,
11
+ "attn_logit_softcapping": null,
12
+ "dtype": "bfloat16",
13
+ "final_logit_softcapping": null,
14
+ "head_dim": 256,
15
+ "hidden_activation": "gelu_pytorch_tanh",
16
+ "hidden_size": 640,
17
+ "initializer_range": 0.02,
18
+ "intermediate_size": 2048,
19
+ "layer_types": [
20
+ "sliding_attention",
21
+ "sliding_attention",
22
+ "sliding_attention",
23
+ "sliding_attention",
24
+ "sliding_attention",
25
+ "full_attention",
26
+ "sliding_attention",
27
+ "sliding_attention",
28
+ "sliding_attention",
29
+ "sliding_attention",
30
+ "sliding_attention",
31
+ "full_attention",
32
+ "sliding_attention",
33
+ "sliding_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "sliding_attention",
37
+ "full_attention"
38
+ ],
39
+ "max_position_embeddings": 32768,
40
+ "model_type": "gemma3_text",
41
+ "neuron": {
42
+ "_serialized_key": "NxDNeuronConfig",
43
+ "batch_size": 4,
44
+ "capacity_factor": null,
45
+ "checkpoint_id": "google/gemma-3-270m-it",
46
+ "checkpoint_revision": "ac82b4e820549b854eebf28ce6dedaf9fdfa17b3",
47
+ "continuous_batching": true,
48
+ "ep_degree": 1,
49
+ "fused_qkv": true,
50
+ "glu_mlp": true,
51
+ "local_ranks_size": 2,
52
+ "max_batch_size": 4,
53
+ "max_context_length": 32768,
54
+ "max_topk": 256,
55
+ "n_active_tokens": 32768,
56
+ "neuronxcc_version": "2.21.33363.0+82129205",
57
+ "on_device_sampling": true,
58
+ "optimum_neuron_version": "0.4.6.dev4",
59
+ "output_logits": false,
60
+ "pp_degree": 1,
61
+ "prefill_chunk_size": 0,
62
+ "sequence_length": 32768,
63
+ "speculation_length": 0,
64
+ "start_rank_id": 0,
65
+ "target": "trn1",
66
+ "torch_dtype": "bfloat16",
67
+ "tp_degree": 2
68
+ },
69
+ "num_attention_heads": 4,
70
+ "num_hidden_layers": 18,
71
+ "num_key_value_heads": 1,
72
+ "query_pre_attn_scalar": 256,
73
+ "rms_norm_eps": 1e-06,
74
+ "rope_local_base_freq": 10000.0,
75
+ "rope_scaling": null,
76
+ "rope_theta": 1000000.0,
77
+ "sliding_window": 512,
78
+ "use_bidirectional_attention": false,
79
+ "use_cache": true,
80
+ "vocab_size": 262144
81
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev4/bd8b5aaeb4693c199f97f6eebd1bbca77255bea6e0ed042c0c592ee5adccec97/d3503138cf92a3bb6ba7.json ADDED
@@ -0,0 +1,81 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "google/gemma-3-270m-it",
4
+ "_sliding_window_pattern": 6,
5
+ "_task": "text-generation",
6
+ "architectures": [
7
+ "Gemma3ForCausalLM"
8
+ ],
9
+ "attention_bias": false,
10
+ "attention_dropout": 0.0,
11
+ "attn_logit_softcapping": null,
12
+ "dtype": "bfloat16",
13
+ "final_logit_softcapping": null,
14
+ "head_dim": 256,
15
+ "hidden_activation": "gelu_pytorch_tanh",
16
+ "hidden_size": 640,
17
+ "initializer_range": 0.02,
18
+ "intermediate_size": 2048,
19
+ "layer_types": [
20
+ "sliding_attention",
21
+ "sliding_attention",
22
+ "sliding_attention",
23
+ "sliding_attention",
24
+ "sliding_attention",
25
+ "full_attention",
26
+ "sliding_attention",
27
+ "sliding_attention",
28
+ "sliding_attention",
29
+ "sliding_attention",
30
+ "sliding_attention",
31
+ "full_attention",
32
+ "sliding_attention",
33
+ "sliding_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "sliding_attention",
37
+ "full_attention"
38
+ ],
39
+ "max_position_embeddings": 32768,
40
+ "model_type": "gemma3_text",
41
+ "neuron": {
42
+ "_serialized_key": "NxDNeuronConfig",
43
+ "batch_size": 8,
44
+ "capacity_factor": null,
45
+ "checkpoint_id": "google/gemma-3-270m-it",
46
+ "checkpoint_revision": "ac82b4e820549b854eebf28ce6dedaf9fdfa17b3",
47
+ "continuous_batching": true,
48
+ "ep_degree": 1,
49
+ "fused_qkv": true,
50
+ "glu_mlp": true,
51
+ "local_ranks_size": 2,
52
+ "max_batch_size": 8,
53
+ "max_context_length": 32768,
54
+ "max_topk": 256,
55
+ "n_active_tokens": 32768,
56
+ "neuronxcc_version": "2.21.33363.0+82129205",
57
+ "on_device_sampling": true,
58
+ "optimum_neuron_version": "0.4.6.dev4",
59
+ "output_logits": false,
60
+ "pp_degree": 1,
61
+ "prefill_chunk_size": 0,
62
+ "sequence_length": 32768,
63
+ "speculation_length": 0,
64
+ "start_rank_id": 0,
65
+ "target": "trn1",
66
+ "torch_dtype": "bfloat16",
67
+ "tp_degree": 2
68
+ },
69
+ "num_attention_heads": 4,
70
+ "num_hidden_layers": 18,
71
+ "num_key_value_heads": 1,
72
+ "query_pre_attn_scalar": 256,
73
+ "rms_norm_eps": 1e-06,
74
+ "rope_local_base_freq": 10000.0,
75
+ "rope_scaling": null,
76
+ "rope_theta": 1000000.0,
77
+ "sliding_window": 512,
78
+ "use_bidirectional_attention": false,
79
+ "use_cache": true,
80
+ "vocab_size": 262144
81
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev4/bd8b5aaeb4693c199f97f6eebd1bbca77255bea6e0ed042c0c592ee5adccec97/f7dfefb13d09e6187d80.json ADDED
@@ -0,0 +1,81 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "google/gemma-3-270m-it",
4
+ "_sliding_window_pattern": 6,
5
+ "_task": "text-generation",
6
+ "architectures": [
7
+ "Gemma3ForCausalLM"
8
+ ],
9
+ "attention_bias": false,
10
+ "attention_dropout": 0.0,
11
+ "attn_logit_softcapping": null,
12
+ "dtype": "bfloat16",
13
+ "final_logit_softcapping": null,
14
+ "head_dim": 256,
15
+ "hidden_activation": "gelu_pytorch_tanh",
16
+ "hidden_size": 640,
17
+ "initializer_range": 0.02,
18
+ "intermediate_size": 2048,
19
+ "layer_types": [
20
+ "sliding_attention",
21
+ "sliding_attention",
22
+ "sliding_attention",
23
+ "sliding_attention",
24
+ "sliding_attention",
25
+ "full_attention",
26
+ "sliding_attention",
27
+ "sliding_attention",
28
+ "sliding_attention",
29
+ "sliding_attention",
30
+ "sliding_attention",
31
+ "full_attention",
32
+ "sliding_attention",
33
+ "sliding_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "sliding_attention",
37
+ "full_attention"
38
+ ],
39
+ "max_position_embeddings": 32768,
40
+ "model_type": "gemma3_text",
41
+ "neuron": {
42
+ "_serialized_key": "NxDNeuronConfig",
43
+ "batch_size": 1,
44
+ "capacity_factor": null,
45
+ "checkpoint_id": "google/gemma-3-270m-it",
46
+ "checkpoint_revision": "ac82b4e820549b854eebf28ce6dedaf9fdfa17b3",
47
+ "continuous_batching": false,
48
+ "ep_degree": 1,
49
+ "fused_qkv": true,
50
+ "glu_mlp": true,
51
+ "local_ranks_size": 2,
52
+ "max_batch_size": 1,
53
+ "max_context_length": 32768,
54
+ "max_topk": 256,
55
+ "n_active_tokens": 32768,
56
+ "neuronxcc_version": "2.21.33363.0+82129205",
57
+ "on_device_sampling": true,
58
+ "optimum_neuron_version": "0.4.6.dev4",
59
+ "output_logits": false,
60
+ "pp_degree": 1,
61
+ "prefill_chunk_size": 0,
62
+ "sequence_length": 32768,
63
+ "speculation_length": 0,
64
+ "start_rank_id": 0,
65
+ "target": "trn1",
66
+ "torch_dtype": "bfloat16",
67
+ "tp_degree": 2
68
+ },
69
+ "num_attention_heads": 4,
70
+ "num_hidden_layers": 18,
71
+ "num_key_value_heads": 1,
72
+ "query_pre_attn_scalar": 256,
73
+ "rms_norm_eps": 1e-06,
74
+ "rope_local_base_freq": 10000.0,
75
+ "rope_scaling": null,
76
+ "rope_theta": 1000000.0,
77
+ "sliding_window": 512,
78
+ "use_bidirectional_attention": false,
79
+ "use_cache": true,
80
+ "vocab_size": 262144
81
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev4/d02c07e1054a0d66fe47bc2bd8603a512b513ad501882b6d49d36b9747c64a47/00688423de1428d98e68.json ADDED
@@ -0,0 +1,91 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "google/gemma-3-1b-it",
4
+ "_sliding_window_pattern": 6,
5
+ "_task": "text-generation",
6
+ "architectures": [
7
+ "Gemma3ForCausalLM"
8
+ ],
9
+ "attention_bias": false,
10
+ "attention_dropout": 0.0,
11
+ "attn_logit_softcapping": null,
12
+ "cache_implementation": "hybrid",
13
+ "dtype": "bfloat16",
14
+ "final_logit_softcapping": null,
15
+ "head_dim": 256,
16
+ "hidden_activation": "gelu_pytorch_tanh",
17
+ "hidden_size": 1152,
18
+ "initializer_range": 0.02,
19
+ "intermediate_size": 6912,
20
+ "layer_types": [
21
+ "sliding_attention",
22
+ "sliding_attention",
23
+ "sliding_attention",
24
+ "sliding_attention",
25
+ "sliding_attention",
26
+ "full_attention",
27
+ "sliding_attention",
28
+ "sliding_attention",
29
+ "sliding_attention",
30
+ "sliding_attention",
31
+ "sliding_attention",
32
+ "full_attention",
33
+ "sliding_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "sliding_attention",
37
+ "sliding_attention",
38
+ "full_attention",
39
+ "sliding_attention",
40
+ "sliding_attention",
41
+ "sliding_attention",
42
+ "sliding_attention",
43
+ "sliding_attention",
44
+ "full_attention",
45
+ "sliding_attention",
46
+ "sliding_attention"
47
+ ],
48
+ "max_position_embeddings": 32768,
49
+ "model_type": "gemma3_text",
50
+ "neuron": {
51
+ "_serialized_key": "NxDNeuronConfig",
52
+ "batch_size": 1,
53
+ "capacity_factor": null,
54
+ "checkpoint_id": "google/gemma-3-1b-it",
55
+ "checkpoint_revision": "dcc83ea841ab6100d6b47a070329e1ba4cf78752",
56
+ "continuous_batching": false,
57
+ "ep_degree": 1,
58
+ "fused_qkv": true,
59
+ "glu_mlp": true,
60
+ "local_ranks_size": 2,
61
+ "max_batch_size": 1,
62
+ "max_context_length": 32768,
63
+ "max_topk": 256,
64
+ "n_active_tokens": 32768,
65
+ "neuronxcc_version": "2.21.33363.0+82129205",
66
+ "on_device_sampling": true,
67
+ "optimum_neuron_version": "0.4.6.dev4",
68
+ "output_logits": false,
69
+ "pp_degree": 1,
70
+ "prefill_chunk_size": 0,
71
+ "sequence_length": 32768,
72
+ "speculation_length": 0,
73
+ "start_rank_id": 0,
74
+ "target": "trn1",
75
+ "torch_dtype": "bfloat16",
76
+ "tp_degree": 2
77
+ },
78
+ "num_attention_heads": 4,
79
+ "num_hidden_layers": 26,
80
+ "num_key_value_heads": 1,
81
+ "query_pre_attn_scalar": 256,
82
+ "rms_norm_eps": 1e-06,
83
+ "rope_local_base_freq": 10000,
84
+ "rope_scaling": null,
85
+ "rope_theta": 1000000,
86
+ "sliding_window": 512,
87
+ "sliding_window_pattern": 6,
88
+ "use_bidirectional_attention": false,
89
+ "use_cache": true,
90
+ "vocab_size": 262144
91
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev4/d02c07e1054a0d66fe47bc2bd8603a512b513ad501882b6d49d36b9747c64a47/87e6309b5c6fe024caaa.json ADDED
@@ -0,0 +1,91 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "google/gemma-3-1b-it",
4
+ "_sliding_window_pattern": 6,
5
+ "_task": "text-generation",
6
+ "architectures": [
7
+ "Gemma3ForCausalLM"
8
+ ],
9
+ "attention_bias": false,
10
+ "attention_dropout": 0.0,
11
+ "attn_logit_softcapping": null,
12
+ "cache_implementation": "hybrid",
13
+ "dtype": "bfloat16",
14
+ "final_logit_softcapping": null,
15
+ "head_dim": 256,
16
+ "hidden_activation": "gelu_pytorch_tanh",
17
+ "hidden_size": 1152,
18
+ "initializer_range": 0.02,
19
+ "intermediate_size": 6912,
20
+ "layer_types": [
21
+ "sliding_attention",
22
+ "sliding_attention",
23
+ "sliding_attention",
24
+ "sliding_attention",
25
+ "sliding_attention",
26
+ "full_attention",
27
+ "sliding_attention",
28
+ "sliding_attention",
29
+ "sliding_attention",
30
+ "sliding_attention",
31
+ "sliding_attention",
32
+ "full_attention",
33
+ "sliding_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "sliding_attention",
37
+ "sliding_attention",
38
+ "full_attention",
39
+ "sliding_attention",
40
+ "sliding_attention",
41
+ "sliding_attention",
42
+ "sliding_attention",
43
+ "sliding_attention",
44
+ "full_attention",
45
+ "sliding_attention",
46
+ "sliding_attention"
47
+ ],
48
+ "max_position_embeddings": 32768,
49
+ "model_type": "gemma3_text",
50
+ "neuron": {
51
+ "_serialized_key": "NxDNeuronConfig",
52
+ "batch_size": 16,
53
+ "capacity_factor": null,
54
+ "checkpoint_id": "google/gemma-3-1b-it",
55
+ "checkpoint_revision": "dcc83ea841ab6100d6b47a070329e1ba4cf78752",
56
+ "continuous_batching": true,
57
+ "ep_degree": 1,
58
+ "fused_qkv": true,
59
+ "glu_mlp": true,
60
+ "local_ranks_size": 2,
61
+ "max_batch_size": 16,
62
+ "max_context_length": 16384,
63
+ "max_topk": 256,
64
+ "n_active_tokens": 16384,
65
+ "neuronxcc_version": "2.21.33363.0+82129205",
66
+ "on_device_sampling": true,
67
+ "optimum_neuron_version": "0.4.6.dev4",
68
+ "output_logits": false,
69
+ "pp_degree": 1,
70
+ "prefill_chunk_size": 0,
71
+ "sequence_length": 16384,
72
+ "speculation_length": 0,
73
+ "start_rank_id": 0,
74
+ "target": "trn1",
75
+ "torch_dtype": "bfloat16",
76
+ "tp_degree": 2
77
+ },
78
+ "num_attention_heads": 4,
79
+ "num_hidden_layers": 26,
80
+ "num_key_value_heads": 1,
81
+ "query_pre_attn_scalar": 256,
82
+ "rms_norm_eps": 1e-06,
83
+ "rope_local_base_freq": 10000,
84
+ "rope_scaling": null,
85
+ "rope_theta": 1000000,
86
+ "sliding_window": 512,
87
+ "sliding_window_pattern": 6,
88
+ "use_bidirectional_attention": false,
89
+ "use_cache": true,
90
+ "vocab_size": 262144
91
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev4/d02c07e1054a0d66fe47bc2bd8603a512b513ad501882b6d49d36b9747c64a47/8eff71913f842fa8f404.json ADDED
@@ -0,0 +1,91 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "google/gemma-3-1b-it",
4
+ "_sliding_window_pattern": 6,
5
+ "_task": "text-generation",
6
+ "architectures": [
7
+ "Gemma3ForCausalLM"
8
+ ],
9
+ "attention_bias": false,
10
+ "attention_dropout": 0.0,
11
+ "attn_logit_softcapping": null,
12
+ "cache_implementation": "hybrid",
13
+ "dtype": "bfloat16",
14
+ "final_logit_softcapping": null,
15
+ "head_dim": 256,
16
+ "hidden_activation": "gelu_pytorch_tanh",
17
+ "hidden_size": 1152,
18
+ "initializer_range": 0.02,
19
+ "intermediate_size": 6912,
20
+ "layer_types": [
21
+ "sliding_attention",
22
+ "sliding_attention",
23
+ "sliding_attention",
24
+ "sliding_attention",
25
+ "sliding_attention",
26
+ "full_attention",
27
+ "sliding_attention",
28
+ "sliding_attention",
29
+ "sliding_attention",
30
+ "sliding_attention",
31
+ "sliding_attention",
32
+ "full_attention",
33
+ "sliding_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "sliding_attention",
37
+ "sliding_attention",
38
+ "full_attention",
39
+ "sliding_attention",
40
+ "sliding_attention",
41
+ "sliding_attention",
42
+ "sliding_attention",
43
+ "sliding_attention",
44
+ "full_attention",
45
+ "sliding_attention",
46
+ "sliding_attention"
47
+ ],
48
+ "max_position_embeddings": 32768,
49
+ "model_type": "gemma3_text",
50
+ "neuron": {
51
+ "_serialized_key": "NxDNeuronConfig",
52
+ "batch_size": 8,
53
+ "capacity_factor": null,
54
+ "checkpoint_id": "google/gemma-3-1b-it",
55
+ "checkpoint_revision": "dcc83ea841ab6100d6b47a070329e1ba4cf78752",
56
+ "continuous_batching": true,
57
+ "ep_degree": 1,
58
+ "fused_qkv": true,
59
+ "glu_mlp": true,
60
+ "local_ranks_size": 2,
61
+ "max_batch_size": 8,
62
+ "max_context_length": 32768,
63
+ "max_topk": 256,
64
+ "n_active_tokens": 32768,
65
+ "neuronxcc_version": "2.21.33363.0+82129205",
66
+ "on_device_sampling": true,
67
+ "optimum_neuron_version": "0.4.6.dev4",
68
+ "output_logits": false,
69
+ "pp_degree": 1,
70
+ "prefill_chunk_size": 0,
71
+ "sequence_length": 32768,
72
+ "speculation_length": 0,
73
+ "start_rank_id": 0,
74
+ "target": "trn1",
75
+ "torch_dtype": "bfloat16",
76
+ "tp_degree": 2
77
+ },
78
+ "num_attention_heads": 4,
79
+ "num_hidden_layers": 26,
80
+ "num_key_value_heads": 1,
81
+ "query_pre_attn_scalar": 256,
82
+ "rms_norm_eps": 1e-06,
83
+ "rope_local_base_freq": 10000,
84
+ "rope_scaling": null,
85
+ "rope_theta": 1000000,
86
+ "sliding_window": 512,
87
+ "sliding_window_pattern": 6,
88
+ "use_bidirectional_attention": false,
89
+ "use_cache": true,
90
+ "vocab_size": 262144
91
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev4/d02c07e1054a0d66fe47bc2bd8603a512b513ad501882b6d49d36b9747c64a47/b8db5a05f548780ebfae.json ADDED
@@ -0,0 +1,91 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "google/gemma-3-1b-it",
4
+ "_sliding_window_pattern": 6,
5
+ "_task": "text-generation",
6
+ "architectures": [
7
+ "Gemma3ForCausalLM"
8
+ ],
9
+ "attention_bias": false,
10
+ "attention_dropout": 0.0,
11
+ "attn_logit_softcapping": null,
12
+ "cache_implementation": "hybrid",
13
+ "dtype": "bfloat16",
14
+ "final_logit_softcapping": null,
15
+ "head_dim": 256,
16
+ "hidden_activation": "gelu_pytorch_tanh",
17
+ "hidden_size": 1152,
18
+ "initializer_range": 0.02,
19
+ "intermediate_size": 6912,
20
+ "layer_types": [
21
+ "sliding_attention",
22
+ "sliding_attention",
23
+ "sliding_attention",
24
+ "sliding_attention",
25
+ "sliding_attention",
26
+ "full_attention",
27
+ "sliding_attention",
28
+ "sliding_attention",
29
+ "sliding_attention",
30
+ "sliding_attention",
31
+ "sliding_attention",
32
+ "full_attention",
33
+ "sliding_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "sliding_attention",
37
+ "sliding_attention",
38
+ "full_attention",
39
+ "sliding_attention",
40
+ "sliding_attention",
41
+ "sliding_attention",
42
+ "sliding_attention",
43
+ "sliding_attention",
44
+ "full_attention",
45
+ "sliding_attention",
46
+ "sliding_attention"
47
+ ],
48
+ "max_position_embeddings": 32768,
49
+ "model_type": "gemma3_text",
50
+ "neuron": {
51
+ "_serialized_key": "NxDNeuronConfig",
52
+ "batch_size": 32,
53
+ "capacity_factor": null,
54
+ "checkpoint_id": "google/gemma-3-1b-it",
55
+ "checkpoint_revision": "dcc83ea841ab6100d6b47a070329e1ba4cf78752",
56
+ "continuous_batching": true,
57
+ "ep_degree": 1,
58
+ "fused_qkv": true,
59
+ "glu_mlp": true,
60
+ "local_ranks_size": 2,
61
+ "max_batch_size": 32,
62
+ "max_context_length": 16384,
63
+ "max_topk": 256,
64
+ "n_active_tokens": 16384,
65
+ "neuronxcc_version": "2.21.33363.0+82129205",
66
+ "on_device_sampling": true,
67
+ "optimum_neuron_version": "0.4.6.dev4",
68
+ "output_logits": false,
69
+ "pp_degree": 1,
70
+ "prefill_chunk_size": 0,
71
+ "sequence_length": 16384,
72
+ "speculation_length": 0,
73
+ "start_rank_id": 0,
74
+ "target": "trn1",
75
+ "torch_dtype": "bfloat16",
76
+ "tp_degree": 2
77
+ },
78
+ "num_attention_heads": 4,
79
+ "num_hidden_layers": 26,
80
+ "num_key_value_heads": 1,
81
+ "query_pre_attn_scalar": 256,
82
+ "rms_norm_eps": 1e-06,
83
+ "rope_local_base_freq": 10000,
84
+ "rope_scaling": null,
85
+ "rope_theta": 1000000,
86
+ "sliding_window": 512,
87
+ "sliding_window_pattern": 6,
88
+ "use_bidirectional_attention": false,
89
+ "use_cache": true,
90
+ "vocab_size": 262144
91
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev4/d02c07e1054a0d66fe47bc2bd8603a512b513ad501882b6d49d36b9747c64a47/dfcd91ed17670ed71f9d.json ADDED
@@ -0,0 +1,91 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "google/gemma-3-1b-it",
4
+ "_sliding_window_pattern": 6,
5
+ "_task": "text-generation",
6
+ "architectures": [
7
+ "Gemma3ForCausalLM"
8
+ ],
9
+ "attention_bias": false,
10
+ "attention_dropout": 0.0,
11
+ "attn_logit_softcapping": null,
12
+ "cache_implementation": "hybrid",
13
+ "dtype": "bfloat16",
14
+ "final_logit_softcapping": null,
15
+ "head_dim": 256,
16
+ "hidden_activation": "gelu_pytorch_tanh",
17
+ "hidden_size": 1152,
18
+ "initializer_range": 0.02,
19
+ "intermediate_size": 6912,
20
+ "layer_types": [
21
+ "sliding_attention",
22
+ "sliding_attention",
23
+ "sliding_attention",
24
+ "sliding_attention",
25
+ "sliding_attention",
26
+ "full_attention",
27
+ "sliding_attention",
28
+ "sliding_attention",
29
+ "sliding_attention",
30
+ "sliding_attention",
31
+ "sliding_attention",
32
+ "full_attention",
33
+ "sliding_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "sliding_attention",
37
+ "sliding_attention",
38
+ "full_attention",
39
+ "sliding_attention",
40
+ "sliding_attention",
41
+ "sliding_attention",
42
+ "sliding_attention",
43
+ "sliding_attention",
44
+ "full_attention",
45
+ "sliding_attention",
46
+ "sliding_attention"
47
+ ],
48
+ "max_position_embeddings": 32768,
49
+ "model_type": "gemma3_text",
50
+ "neuron": {
51
+ "_serialized_key": "NxDNeuronConfig",
52
+ "batch_size": 16,
53
+ "capacity_factor": null,
54
+ "checkpoint_id": "google/gemma-3-1b-it",
55
+ "checkpoint_revision": "dcc83ea841ab6100d6b47a070329e1ba4cf78752",
56
+ "continuous_batching": true,
57
+ "ep_degree": 1,
58
+ "fused_qkv": true,
59
+ "glu_mlp": true,
60
+ "local_ranks_size": 2,
61
+ "max_batch_size": 16,
62
+ "max_context_length": 32768,
63
+ "max_topk": 256,
64
+ "n_active_tokens": 32768,
65
+ "neuronxcc_version": "2.21.33363.0+82129205",
66
+ "on_device_sampling": true,
67
+ "optimum_neuron_version": "0.4.6.dev4",
68
+ "output_logits": false,
69
+ "pp_degree": 1,
70
+ "prefill_chunk_size": 0,
71
+ "sequence_length": 32768,
72
+ "speculation_length": 0,
73
+ "start_rank_id": 0,
74
+ "target": "trn1",
75
+ "torch_dtype": "bfloat16",
76
+ "tp_degree": 2
77
+ },
78
+ "num_attention_heads": 4,
79
+ "num_hidden_layers": 26,
80
+ "num_key_value_heads": 1,
81
+ "query_pre_attn_scalar": 256,
82
+ "rms_norm_eps": 1e-06,
83
+ "rope_local_base_freq": 10000,
84
+ "rope_scaling": null,
85
+ "rope_theta": 1000000,
86
+ "sliding_window": 512,
87
+ "sliding_window_pattern": 6,
88
+ "use_bidirectional_attention": false,
89
+ "use_cache": true,
90
+ "vocab_size": 262144
91
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev4/d02c07e1054a0d66fe47bc2bd8603a512b513ad501882b6d49d36b9747c64a47/ed4c5924c8e305088a85.json ADDED
@@ -0,0 +1,91 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "google/gemma-3-1b-it",
4
+ "_sliding_window_pattern": 6,
5
+ "_task": "text-generation",
6
+ "architectures": [
7
+ "Gemma3ForCausalLM"
8
+ ],
9
+ "attention_bias": false,
10
+ "attention_dropout": 0.0,
11
+ "attn_logit_softcapping": null,
12
+ "cache_implementation": "hybrid",
13
+ "dtype": "bfloat16",
14
+ "final_logit_softcapping": null,
15
+ "head_dim": 256,
16
+ "hidden_activation": "gelu_pytorch_tanh",
17
+ "hidden_size": 1152,
18
+ "initializer_range": 0.02,
19
+ "intermediate_size": 6912,
20
+ "layer_types": [
21
+ "sliding_attention",
22
+ "sliding_attention",
23
+ "sliding_attention",
24
+ "sliding_attention",
25
+ "sliding_attention",
26
+ "full_attention",
27
+ "sliding_attention",
28
+ "sliding_attention",
29
+ "sliding_attention",
30
+ "sliding_attention",
31
+ "sliding_attention",
32
+ "full_attention",
33
+ "sliding_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "sliding_attention",
37
+ "sliding_attention",
38
+ "full_attention",
39
+ "sliding_attention",
40
+ "sliding_attention",
41
+ "sliding_attention",
42
+ "sliding_attention",
43
+ "sliding_attention",
44
+ "full_attention",
45
+ "sliding_attention",
46
+ "sliding_attention"
47
+ ],
48
+ "max_position_embeddings": 32768,
49
+ "model_type": "gemma3_text",
50
+ "neuron": {
51
+ "_serialized_key": "NxDNeuronConfig",
52
+ "batch_size": 4,
53
+ "capacity_factor": null,
54
+ "checkpoint_id": "google/gemma-3-1b-it",
55
+ "checkpoint_revision": "dcc83ea841ab6100d6b47a070329e1ba4cf78752",
56
+ "continuous_batching": true,
57
+ "ep_degree": 1,
58
+ "fused_qkv": true,
59
+ "glu_mlp": true,
60
+ "local_ranks_size": 2,
61
+ "max_batch_size": 4,
62
+ "max_context_length": 32768,
63
+ "max_topk": 256,
64
+ "n_active_tokens": 32768,
65
+ "neuronxcc_version": "2.21.33363.0+82129205",
66
+ "on_device_sampling": true,
67
+ "optimum_neuron_version": "0.4.6.dev4",
68
+ "output_logits": false,
69
+ "pp_degree": 1,
70
+ "prefill_chunk_size": 0,
71
+ "sequence_length": 32768,
72
+ "speculation_length": 0,
73
+ "start_rank_id": 0,
74
+ "target": "trn1",
75
+ "torch_dtype": "bfloat16",
76
+ "tp_degree": 2
77
+ },
78
+ "num_attention_heads": 4,
79
+ "num_hidden_layers": 26,
80
+ "num_key_value_heads": 1,
81
+ "query_pre_attn_scalar": 256,
82
+ "rms_norm_eps": 1e-06,
83
+ "rope_local_base_freq": 10000,
84
+ "rope_scaling": null,
85
+ "rope_theta": 1000000,
86
+ "sliding_window": 512,
87
+ "sliding_window_pattern": 6,
88
+ "use_bidirectional_attention": false,
89
+ "use_cache": true,
90
+ "vocab_size": 262144
91
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev4/gemma3_text/google/gemma-3-1b-it/b8db5a05f548780ebfae.json ADDED
@@ -0,0 +1,91 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "google/gemma-3-1b-it",
4
+ "_sliding_window_pattern": 6,
5
+ "_task": "text-generation",
6
+ "architectures": [
7
+ "Gemma3ForCausalLM"
8
+ ],
9
+ "attention_bias": false,
10
+ "attention_dropout": 0.0,
11
+ "attn_logit_softcapping": null,
12
+ "cache_implementation": "hybrid",
13
+ "dtype": "bfloat16",
14
+ "final_logit_softcapping": null,
15
+ "head_dim": 256,
16
+ "hidden_activation": "gelu_pytorch_tanh",
17
+ "hidden_size": 1152,
18
+ "initializer_range": 0.02,
19
+ "intermediate_size": 6912,
20
+ "layer_types": [
21
+ "sliding_attention",
22
+ "sliding_attention",
23
+ "sliding_attention",
24
+ "sliding_attention",
25
+ "sliding_attention",
26
+ "full_attention",
27
+ "sliding_attention",
28
+ "sliding_attention",
29
+ "sliding_attention",
30
+ "sliding_attention",
31
+ "sliding_attention",
32
+ "full_attention",
33
+ "sliding_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "sliding_attention",
37
+ "sliding_attention",
38
+ "full_attention",
39
+ "sliding_attention",
40
+ "sliding_attention",
41
+ "sliding_attention",
42
+ "sliding_attention",
43
+ "sliding_attention",
44
+ "full_attention",
45
+ "sliding_attention",
46
+ "sliding_attention"
47
+ ],
48
+ "max_position_embeddings": 32768,
49
+ "model_type": "gemma3_text",
50
+ "neuron": {
51
+ "_serialized_key": "NxDNeuronConfig",
52
+ "batch_size": 32,
53
+ "capacity_factor": null,
54
+ "checkpoint_id": "google/gemma-3-1b-it",
55
+ "checkpoint_revision": "dcc83ea841ab6100d6b47a070329e1ba4cf78752",
56
+ "continuous_batching": true,
57
+ "ep_degree": 1,
58
+ "fused_qkv": true,
59
+ "glu_mlp": true,
60
+ "local_ranks_size": 2,
61
+ "max_batch_size": 32,
62
+ "max_context_length": 16384,
63
+ "max_topk": 256,
64
+ "n_active_tokens": 16384,
65
+ "neuronxcc_version": "2.21.33363.0+82129205",
66
+ "on_device_sampling": true,
67
+ "optimum_neuron_version": "0.4.6.dev4",
68
+ "output_logits": false,
69
+ "pp_degree": 1,
70
+ "prefill_chunk_size": 0,
71
+ "sequence_length": 16384,
72
+ "speculation_length": 0,
73
+ "start_rank_id": 0,
74
+ "target": "trn1",
75
+ "torch_dtype": "bfloat16",
76
+ "tp_degree": 2
77
+ },
78
+ "num_attention_heads": 4,
79
+ "num_hidden_layers": 26,
80
+ "num_key_value_heads": 1,
81
+ "query_pre_attn_scalar": 256,
82
+ "rms_norm_eps": 1e-06,
83
+ "rope_local_base_freq": 10000,
84
+ "rope_scaling": null,
85
+ "rope_theta": 1000000,
86
+ "sliding_window": 512,
87
+ "sliding_window_pattern": 6,
88
+ "use_bidirectional_attention": false,
89
+ "use_cache": true,
90
+ "vocab_size": 262144
91
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.6.dev4/gemma3_text/google/gemma-3-270m-it/18421eedc4d9b7b7fb51.json ADDED
@@ -0,0 +1,81 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "google/gemma-3-270m-it",
4
+ "_sliding_window_pattern": 6,
5
+ "_task": "text-generation",
6
+ "architectures": [
7
+ "Gemma3ForCausalLM"
8
+ ],
9
+ "attention_bias": false,
10
+ "attention_dropout": 0.0,
11
+ "attn_logit_softcapping": null,
12
+ "dtype": "bfloat16",
13
+ "final_logit_softcapping": null,
14
+ "head_dim": 256,
15
+ "hidden_activation": "gelu_pytorch_tanh",
16
+ "hidden_size": 640,
17
+ "initializer_range": 0.02,
18
+ "intermediate_size": 2048,
19
+ "layer_types": [
20
+ "sliding_attention",
21
+ "sliding_attention",
22
+ "sliding_attention",
23
+ "sliding_attention",
24
+ "sliding_attention",
25
+ "full_attention",
26
+ "sliding_attention",
27
+ "sliding_attention",
28
+ "sliding_attention",
29
+ "sliding_attention",
30
+ "sliding_attention",
31
+ "full_attention",
32
+ "sliding_attention",
33
+ "sliding_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "sliding_attention",
37
+ "full_attention"
38
+ ],
39
+ "max_position_embeddings": 32768,
40
+ "model_type": "gemma3_text",
41
+ "neuron": {
42
+ "_serialized_key": "NxDNeuronConfig",
43
+ "batch_size": 32,
44
+ "capacity_factor": null,
45
+ "checkpoint_id": "google/gemma-3-270m-it",
46
+ "checkpoint_revision": "ac82b4e820549b854eebf28ce6dedaf9fdfa17b3",
47
+ "continuous_batching": true,
48
+ "ep_degree": 1,
49
+ "fused_qkv": true,
50
+ "glu_mlp": true,
51
+ "local_ranks_size": 2,
52
+ "max_batch_size": 32,
53
+ "max_context_length": 16384,
54
+ "max_topk": 256,
55
+ "n_active_tokens": 16384,
56
+ "neuronxcc_version": "2.21.33363.0+82129205",
57
+ "on_device_sampling": true,
58
+ "optimum_neuron_version": "0.4.6.dev4",
59
+ "output_logits": false,
60
+ "pp_degree": 1,
61
+ "prefill_chunk_size": 0,
62
+ "sequence_length": 16384,
63
+ "speculation_length": 0,
64
+ "start_rank_id": 0,
65
+ "target": "trn1",
66
+ "torch_dtype": "bfloat16",
67
+ "tp_degree": 2
68
+ },
69
+ "num_attention_heads": 4,
70
+ "num_hidden_layers": 18,
71
+ "num_key_value_heads": 1,
72
+ "query_pre_attn_scalar": 256,
73
+ "rms_norm_eps": 1e-06,
74
+ "rope_local_base_freq": 10000.0,
75
+ "rope_scaling": null,
76
+ "rope_theta": 1000000.0,
77
+ "sliding_window": 512,
78
+ "use_bidirectional_attention": false,
79
+ "use_cache": true,
80
+ "vocab_size": 262144
81
+ }
neuronxcc-2.21.33363.0+82129205/MODULE_083c75747563fca496d7+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_083c75747563fca496d7+24129607/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_083c75747563fca496d7+24129607/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:03adffb1ac29009d5dc73c5a1cb3238a85f92058a525ad350f72cac0708fce18
3
+ size 1779146
neuronxcc-2.21.33363.0+82129205/MODULE_083c75747563fca496d7+24129607/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:597f4f3cdcd094d9b61867edb5be9b2772bccfdc5109041406729cae7091bb0f
3
+ size 32287744
neuronxcc-2.21.33363.0+82129205/MODULE_134003a684b11c49131d+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_134003a684b11c49131d+24129607/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_134003a684b11c49131d+24129607/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:570017663a6900b5493f75c47921e938e0e1744a44c1f04d5f4536bfa44d644c
3
+ size 1779218
neuronxcc-2.21.33363.0+82129205/MODULE_134003a684b11c49131d+24129607/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9544338723eb399adde799a747e2c438ff69fdfd1deeec0e1ce187b38edab180
3
+ size 83313664
neuronxcc-2.21.33363.0+82129205/MODULE_1b9cad3a2eb3c406661d+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_1b9cad3a2eb3c406661d+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_1b9cad3a2eb3c406661d+a02c3a36/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4cffabd2edca1fa4e310a911a7f1ee3949022a4d512d81ab71073566d4e75b82
3
+ size 842788
neuronxcc-2.21.33363.0+82129205/MODULE_1b9cad3a2eb3c406661d+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b397f20205d28dac763ae3e1b0b2d6cc82391a85343d2e59a61b09c83f0b41a2
3
+ size 15289344
neuronxcc-2.21.33363.0+82129205/MODULE_1b9cad3a2eb3c406661d+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d7789b9e434dece65641d8f5f98d2b54d005c9a140d377677724c8cef055f366
3
+ size 15466469
neuronxcc-2.21.33363.0+82129205/MODULE_33dcb1875eda19b3ebb5+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_33dcb1875eda19b3ebb5+24129607/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_33dcb1875eda19b3ebb5+24129607/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3cb39e4ee43fef9e4ab34ba0c65d0e5d40a495386b6e2b33b16ade366f630493
3
+ size 2549448
neuronxcc-2.21.33363.0+82129205/MODULE_33dcb1875eda19b3ebb5+24129607/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4a41555756fc779961639134f6e2566d8faf2f2e441379455fa9d097f0232525
3
+ size 7711744
neuronxcc-2.21.33363.0+82129205/MODULE_4978e0dffee300919f4f+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_4978e0dffee300919f4f+24129607/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_4978e0dffee300919f4f+24129607/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:86f1aa49dbddfc33c2f59d85ddec5a2e3d26274e2bd24dc9b44645a626a9acf0
3
+ size 2549552
neuronxcc-2.21.33363.0+82129205/MODULE_4978e0dffee300919f4f+24129607/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8427cba9f0cc79b11f2a4905885b74a5ee87deac370e3845d8b65da6f6250a03
3
+ size 17900544
neuronxcc-2.21.33363.0+82129205/MODULE_4dd21482986783a76fee+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_4dd21482986783a76fee+24129607/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_4dd21482986783a76fee+24129607/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:61d0ec3e537d1edb6bd63452260cc0127e02213437bfd0e40bd7c37a039b2993
3
+ size 1739896
neuronxcc-2.21.33363.0+82129205/MODULE_4dd21482986783a76fee+24129607/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e6cf965cb2fdea318aa68997cd8cca53e7fb87c41a68cf7e0c7acca4bdcefa06
3
+ size 82566144
neuronxcc-2.21.33363.0+82129205/MODULE_61998e972adde9b6f5d2+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_61998e972adde9b6f5d2+24129607/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_61998e972adde9b6f5d2+24129607/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5a014b74ceda95eade982c79b8f8c1b429e66ab6a36b6a6a7fed94bb0e453bf6
3
+ size 2549448
neuronxcc-2.21.33363.0+82129205/MODULE_61998e972adde9b6f5d2+24129607/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e406af060c688d0d78087538230afc828390f3b763072b3a181d212d56a4ce59
3
+ size 7711744
neuronxcc-2.21.33363.0+82129205/MODULE_7701466c22adfdd959a6+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_7701466c22adfdd959a6+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_7701466c22adfdd959a6+a02c3a36/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:123441c0c01d741960d9edf16df88dad93f3426a0a1a940c2348c8e6d3add05a
3
+ size 891117
neuronxcc-2.21.33363.0+82129205/MODULE_7701466c22adfdd959a6+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:936df18dc8957bed20db8622f6d7b72bc72e61832b07efa4c8d925146b024023
3
+ size 5530624
neuronxcc-2.21.33363.0+82129205/MODULE_7701466c22adfdd959a6+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aaa2ade0e387063ba57d08563e9714f731604a9747d8a4a39ef642f0dbed3402
3
+ size 5707749
neuronxcc-2.21.33363.0+82129205/MODULE_8b7c51b3642e17a79f64+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_8b7c51b3642e17a79f64+24129607/model.done ADDED
File without changes