Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
 Community
666
dacorvo HF Staff commited on
Commit
fd740ce
·
verified ·
1 Parent(s): 2467782

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 +9 -0
  2. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev2/0cbac419f8839d90c6bafb19e17441f1a052e93e227f0fa62918ebe7d882e225/88fa555c8659e3f7fc5a.json +95 -0
  3. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev2/af58eb15d8e02338dc2f2e880e9c6ec803a98278914b3606acdcc252e7e18429/49052c1d22e88f219887.json +95 -0
  4. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev2/af58eb15d8e02338dc2f2e880e9c6ec803a98278914b3606acdcc252e7e18429/67070f0a4d500338d5aa.json +95 -0
  5. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev2/af58eb15d8e02338dc2f2e880e9c6ec803a98278914b3606acdcc252e7e18429/99d672cbaa2e018da4e4.json +95 -0
  6. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev2/af58eb15d8e02338dc2f2e880e9c6ec803a98278914b3606acdcc252e7e18429/ad354086b250f133c9c6.json +95 -0
  7. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev2/af58eb15d8e02338dc2f2e880e9c6ec803a98278914b3606acdcc252e7e18429/b8a28a3ac7bcba98b595.json +95 -0
  8. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev2/af58eb15d8e02338dc2f2e880e9c6ec803a98278914b3606acdcc252e7e18429/fa7d033eb1dca73fe9c3.json +95 -0
  9. neuronxcc-2.21.33363.0+82129205/MODULE_1a3939beb88e18a29e1e+fb4cc044/model.hlo_module.pb +1 -1
  10. neuronxcc-2.21.33363.0+82129205/MODULE_1a41189de75fc0ed15bb+fb4cc044/model.hlo_module.pb +1 -1
  11. neuronxcc-2.21.33363.0+82129205/MODULE_1afb27ccf10708e03947+fb4cc044/model.hlo_module.pb +1 -1
  12. neuronxcc-2.21.33363.0+82129205/MODULE_269a21b3e3cb06fc787d+fb4cc044/model.hlo_module.pb +1 -1
  13. neuronxcc-2.21.33363.0+82129205/MODULE_2831e4c199ca2002f484+24129607/compile_flags.json +1 -0
  14. neuronxcc-2.21.33363.0+82129205/MODULE_2831e4c199ca2002f484+24129607/model.done +0 -0
  15. neuronxcc-2.21.33363.0+82129205/MODULE_2831e4c199ca2002f484+24129607/model.hlo_module.pb +3 -0
  16. neuronxcc-2.21.33363.0+82129205/MODULE_2831e4c199ca2002f484+24129607/model.neff +3 -0
  17. neuronxcc-2.21.33363.0+82129205/MODULE_35958e2f8cf19e5eddc6+fb4cc044/compile_flags.json +1 -0
  18. neuronxcc-2.21.33363.0+82129205/MODULE_35958e2f8cf19e5eddc6+fb4cc044/model.done +0 -0
  19. neuronxcc-2.21.33363.0+82129205/MODULE_35958e2f8cf19e5eddc6+fb4cc044/model.hlo_module.pb +3 -0
  20. neuronxcc-2.21.33363.0+82129205/MODULE_35958e2f8cf19e5eddc6+fb4cc044/model.neff +3 -0
  21. neuronxcc-2.21.33363.0+82129205/MODULE_54baea067a19d8c74866+fb4cc044/model.hlo_module.pb +1 -1
  22. neuronxcc-2.21.33363.0+82129205/MODULE_5c4a6aceb0aa544053e8+fb4cc044/model.hlo_module.pb +1 -1
  23. neuronxcc-2.21.33363.0+82129205/MODULE_5dafb05b2fa15b606ad4+a02c3a36/compile_flags.json +1 -0
  24. neuronxcc-2.21.33363.0+82129205/MODULE_5dafb05b2fa15b606ad4+a02c3a36/model.done +0 -0
  25. neuronxcc-2.21.33363.0+82129205/MODULE_5dafb05b2fa15b606ad4+a02c3a36/model.hlo_module.pb +3 -0
  26. neuronxcc-2.21.33363.0+82129205/MODULE_5dafb05b2fa15b606ad4+a02c3a36/model.neff +3 -0
  27. neuronxcc-2.21.33363.0+82129205/MODULE_5dafb05b2fa15b606ad4+a02c3a36/wrapped_neff.hlo +3 -0
  28. neuronxcc-2.21.33363.0+82129205/MODULE_63b228a1aedd2e24d261+fb4cc044/model.hlo_module.pb +1 -1
  29. neuronxcc-2.21.33363.0+82129205/MODULE_64886f8b7709113eb14b+fb4cc044/compile_flags.json +1 -0
  30. neuronxcc-2.21.33363.0+82129205/MODULE_64886f8b7709113eb14b+fb4cc044/model.done +0 -0
  31. neuronxcc-2.21.33363.0+82129205/MODULE_64886f8b7709113eb14b+fb4cc044/model.hlo_module.pb +3 -0
  32. neuronxcc-2.21.33363.0+82129205/MODULE_64886f8b7709113eb14b+fb4cc044/model.neff +3 -0
  33. neuronxcc-2.21.33363.0+82129205/MODULE_758311bdb6e777cb53a2+fb4cc044/compile_flags.json +1 -0
  34. neuronxcc-2.21.33363.0+82129205/MODULE_758311bdb6e777cb53a2+fb4cc044/model.done +0 -0
  35. neuronxcc-2.21.33363.0+82129205/MODULE_758311bdb6e777cb53a2+fb4cc044/model.hlo_module.pb +3 -0
  36. neuronxcc-2.21.33363.0+82129205/MODULE_758311bdb6e777cb53a2+fb4cc044/model.neff +3 -0
  37. neuronxcc-2.21.33363.0+82129205/MODULE_81503bf195cd5d2e22b3+fb4cc044/model.hlo_module.pb +1 -1
  38. neuronxcc-2.21.33363.0+82129205/MODULE_87ac40e3234811a067b9+fb4cc044/model.hlo_module.pb +1 -1
  39. neuronxcc-2.21.33363.0+82129205/MODULE_8d8d38359e9e9f052a77+fb4cc044/model.hlo_module.pb +1 -1
  40. neuronxcc-2.21.33363.0+82129205/MODULE_9a666844f4acef43578b+fb4cc044/model.hlo_module.pb +1 -1
  41. neuronxcc-2.21.33363.0+82129205/MODULE_aaefb5ede5ee0772566c+fb4cc044/compile_flags.json +1 -0
  42. neuronxcc-2.21.33363.0+82129205/MODULE_aaefb5ede5ee0772566c+fb4cc044/model.done +0 -0
  43. neuronxcc-2.21.33363.0+82129205/MODULE_aaefb5ede5ee0772566c+fb4cc044/model.hlo_module.pb +3 -0
  44. neuronxcc-2.21.33363.0+82129205/MODULE_aaefb5ede5ee0772566c+fb4cc044/model.neff +3 -0
  45. neuronxcc-2.21.33363.0+82129205/MODULE_b0cd0a9c4d91cfdefe87+fb4cc044/compile_flags.json +1 -0
  46. neuronxcc-2.21.33363.0+82129205/MODULE_b0cd0a9c4d91cfdefe87+fb4cc044/model.done +0 -0
  47. neuronxcc-2.21.33363.0+82129205/MODULE_b0cd0a9c4d91cfdefe87+fb4cc044/model.hlo_module.pb +3 -0
  48. neuronxcc-2.21.33363.0+82129205/MODULE_b0cd0a9c4d91cfdefe87+fb4cc044/model.neff +3 -0
  49. neuronxcc-2.21.33363.0+82129205/MODULE_b234477c449ee671d3c1+fb4cc044/compile_flags.json +1 -0
  50. neuronxcc-2.21.33363.0+82129205/MODULE_b234477c449ee671d3c1+fb4cc044/model.hlo_module.pb +3 -0
.gitattributes CHANGED
@@ -16456,3 +16456,12 @@ neuronxcc-2.21.33363.0+82129205/MODULE_c0300a21335f2118973b+8bc117fd/model.neff
16456
  neuronxcc-2.21.33363.0+82129205/MODULE_c0300a21335f2118973b+8bc117fd/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
16457
  neuronxcc-2.21.33363.0+82129205/MODULE_fa79bce2a7462b20ad61+c899ece2/model.neff filter=lfs diff=lfs merge=lfs -text
16458
  neuronxcc-2.21.33363.0+82129205/MODULE_fa79bce2a7462b20ad61+c899ece2/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
 
 
 
16456
  neuronxcc-2.21.33363.0+82129205/MODULE_c0300a21335f2118973b+8bc117fd/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
16457
  neuronxcc-2.21.33363.0+82129205/MODULE_fa79bce2a7462b20ad61+c899ece2/model.neff filter=lfs diff=lfs merge=lfs -text
16458
  neuronxcc-2.21.33363.0+82129205/MODULE_fa79bce2a7462b20ad61+c899ece2/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
16459
+ neuronxcc-2.21.33363.0+82129205/MODULE_2831e4c199ca2002f484+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
16460
+ neuronxcc-2.21.33363.0+82129205/MODULE_35958e2f8cf19e5eddc6+fb4cc044/model.neff filter=lfs diff=lfs merge=lfs -text
16461
+ neuronxcc-2.21.33363.0+82129205/MODULE_5dafb05b2fa15b606ad4+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
16462
+ neuronxcc-2.21.33363.0+82129205/MODULE_5dafb05b2fa15b606ad4+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
16463
+ neuronxcc-2.21.33363.0+82129205/MODULE_64886f8b7709113eb14b+fb4cc044/model.neff filter=lfs diff=lfs merge=lfs -text
16464
+ neuronxcc-2.21.33363.0+82129205/MODULE_758311bdb6e777cb53a2+fb4cc044/model.neff filter=lfs diff=lfs merge=lfs -text
16465
+ neuronxcc-2.21.33363.0+82129205/MODULE_aaefb5ede5ee0772566c+fb4cc044/model.neff filter=lfs diff=lfs merge=lfs -text
16466
+ neuronxcc-2.21.33363.0+82129205/MODULE_b0cd0a9c4d91cfdefe87+fb4cc044/model.neff filter=lfs diff=lfs merge=lfs -text
16467
+ neuronxcc-2.21.33363.0+82129205/MODULE_f81dd9a0c4854a438c30+fb4cc044/model.neff filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev2/0cbac419f8839d90c6bafb19e17441f1a052e93e227f0fa62918ebe7d882e225/88fa555c8659e3f7fc5a.json ADDED
@@ -0,0 +1,95 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen3-Embedding-8B",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Qwen3ForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "dtype": "bfloat16",
11
+ "head_dim": 128,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 4096,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 12288,
16
+ "layer_types": [
17
+ "full_attention",
18
+ "full_attention",
19
+ "full_attention",
20
+ "full_attention",
21
+ "full_attention",
22
+ "full_attention",
23
+ "full_attention",
24
+ "full_attention",
25
+ "full_attention",
26
+ "full_attention",
27
+ "full_attention",
28
+ "full_attention",
29
+ "full_attention",
30
+ "full_attention",
31
+ "full_attention",
32
+ "full_attention",
33
+ "full_attention",
34
+ "full_attention",
35
+ "full_attention",
36
+ "full_attention",
37
+ "full_attention",
38
+ "full_attention",
39
+ "full_attention",
40
+ "full_attention",
41
+ "full_attention",
42
+ "full_attention",
43
+ "full_attention",
44
+ "full_attention",
45
+ "full_attention",
46
+ "full_attention",
47
+ "full_attention",
48
+ "full_attention",
49
+ "full_attention",
50
+ "full_attention",
51
+ "full_attention",
52
+ "full_attention"
53
+ ],
54
+ "max_position_embeddings": 40960,
55
+ "max_window_layers": 36,
56
+ "model_type": "qwen3",
57
+ "neuron": {
58
+ "_serialized_key": "NxDNeuronConfig",
59
+ "batch_size": 32,
60
+ "capacity_factor": null,
61
+ "checkpoint_id": "Qwen/Qwen3-Embedding-8B",
62
+ "checkpoint_revision": "1d8ad4ca9b3dd8059ad90a75d4983776a23d44af",
63
+ "continuous_batching": true,
64
+ "ep_degree": 1,
65
+ "fused_qkv": true,
66
+ "glu_mlp": true,
67
+ "local_ranks_size": 8,
68
+ "max_batch_size": 32,
69
+ "max_context_length": 1024,
70
+ "max_topk": 256,
71
+ "n_active_tokens": 1024,
72
+ "neuronxcc_version": "2.21.33363.0+82129205",
73
+ "on_device_sampling": true,
74
+ "optimum_neuron_version": "0.4.5.dev2",
75
+ "output_logits": false,
76
+ "pp_degree": 1,
77
+ "sequence_length": 1024,
78
+ "speculation_length": 0,
79
+ "start_rank_id": 0,
80
+ "target": "trn1",
81
+ "torch_dtype": "bfloat16",
82
+ "tp_degree": 8
83
+ },
84
+ "num_attention_heads": 32,
85
+ "num_hidden_layers": 36,
86
+ "num_key_value_heads": 8,
87
+ "rms_norm_eps": 1e-06,
88
+ "rope_scaling": null,
89
+ "rope_theta": 1000000,
90
+ "sliding_window": null,
91
+ "tie_word_embeddings": false,
92
+ "use_cache": true,
93
+ "use_sliding_window": false,
94
+ "vocab_size": 151665
95
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev2/af58eb15d8e02338dc2f2e880e9c6ec803a98278914b3606acdcc252e7e18429/49052c1d22e88f219887.json ADDED
@@ -0,0 +1,95 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen3-Embedding-8B",
4
+ "_task": "feature-extraction",
5
+ "architectures": [
6
+ "Qwen3ForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "dtype": "bfloat16",
11
+ "head_dim": 128,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 4096,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 12288,
16
+ "layer_types": [
17
+ "full_attention",
18
+ "full_attention",
19
+ "full_attention",
20
+ "full_attention",
21
+ "full_attention",
22
+ "full_attention",
23
+ "full_attention",
24
+ "full_attention",
25
+ "full_attention",
26
+ "full_attention",
27
+ "full_attention",
28
+ "full_attention",
29
+ "full_attention",
30
+ "full_attention",
31
+ "full_attention",
32
+ "full_attention",
33
+ "full_attention",
34
+ "full_attention",
35
+ "full_attention",
36
+ "full_attention",
37
+ "full_attention",
38
+ "full_attention",
39
+ "full_attention",
40
+ "full_attention",
41
+ "full_attention",
42
+ "full_attention",
43
+ "full_attention",
44
+ "full_attention",
45
+ "full_attention",
46
+ "full_attention",
47
+ "full_attention",
48
+ "full_attention",
49
+ "full_attention",
50
+ "full_attention",
51
+ "full_attention",
52
+ "full_attention"
53
+ ],
54
+ "max_position_embeddings": 40960,
55
+ "max_window_layers": 36,
56
+ "model_type": "qwen3",
57
+ "neuron": {
58
+ "_serialized_key": "NxDNeuronConfig",
59
+ "batch_size": 1,
60
+ "capacity_factor": null,
61
+ "checkpoint_id": "Qwen/Qwen3-Embedding-8B",
62
+ "checkpoint_revision": "1d8ad4ca9b3dd8059ad90a75d4983776a23d44af",
63
+ "continuous_batching": false,
64
+ "ep_degree": 1,
65
+ "fused_qkv": true,
66
+ "glu_mlp": true,
67
+ "local_ranks_size": 16,
68
+ "max_batch_size": 1,
69
+ "max_context_length": 32768,
70
+ "max_topk": 256,
71
+ "n_active_tokens": 32768,
72
+ "neuronxcc_version": "2.21.33363.0+82129205",
73
+ "on_device_sampling": false,
74
+ "optimum_neuron_version": "0.4.5.dev2",
75
+ "output_logits": false,
76
+ "pp_degree": 1,
77
+ "sequence_length": 32768,
78
+ "speculation_length": 0,
79
+ "start_rank_id": 0,
80
+ "target": "trn1",
81
+ "torch_dtype": "bfloat16",
82
+ "tp_degree": 16
83
+ },
84
+ "num_attention_heads": 32,
85
+ "num_hidden_layers": 36,
86
+ "num_key_value_heads": 8,
87
+ "rms_norm_eps": 1e-06,
88
+ "rope_scaling": null,
89
+ "rope_theta": 1000000,
90
+ "sliding_window": null,
91
+ "tie_word_embeddings": false,
92
+ "use_cache": true,
93
+ "use_sliding_window": false,
94
+ "vocab_size": 151665
95
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev2/af58eb15d8e02338dc2f2e880e9c6ec803a98278914b3606acdcc252e7e18429/67070f0a4d500338d5aa.json ADDED
@@ -0,0 +1,95 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen3-Embedding-8B",
4
+ "_task": "feature-extraction",
5
+ "architectures": [
6
+ "Qwen3ForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "dtype": "bfloat16",
11
+ "head_dim": 128,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 4096,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 12288,
16
+ "layer_types": [
17
+ "full_attention",
18
+ "full_attention",
19
+ "full_attention",
20
+ "full_attention",
21
+ "full_attention",
22
+ "full_attention",
23
+ "full_attention",
24
+ "full_attention",
25
+ "full_attention",
26
+ "full_attention",
27
+ "full_attention",
28
+ "full_attention",
29
+ "full_attention",
30
+ "full_attention",
31
+ "full_attention",
32
+ "full_attention",
33
+ "full_attention",
34
+ "full_attention",
35
+ "full_attention",
36
+ "full_attention",
37
+ "full_attention",
38
+ "full_attention",
39
+ "full_attention",
40
+ "full_attention",
41
+ "full_attention",
42
+ "full_attention",
43
+ "full_attention",
44
+ "full_attention",
45
+ "full_attention",
46
+ "full_attention",
47
+ "full_attention",
48
+ "full_attention",
49
+ "full_attention",
50
+ "full_attention",
51
+ "full_attention",
52
+ "full_attention"
53
+ ],
54
+ "max_position_embeddings": 40960,
55
+ "max_window_layers": 36,
56
+ "model_type": "qwen3",
57
+ "neuron": {
58
+ "_serialized_key": "NxDNeuronConfig",
59
+ "batch_size": 4,
60
+ "capacity_factor": null,
61
+ "checkpoint_id": "Qwen/Qwen3-Embedding-8B",
62
+ "checkpoint_revision": "1d8ad4ca9b3dd8059ad90a75d4983776a23d44af",
63
+ "continuous_batching": false,
64
+ "ep_degree": 1,
65
+ "fused_qkv": true,
66
+ "glu_mlp": true,
67
+ "local_ranks_size": 16,
68
+ "max_batch_size": 4,
69
+ "max_context_length": 32768,
70
+ "max_topk": 256,
71
+ "n_active_tokens": 32768,
72
+ "neuronxcc_version": "2.21.33363.0+82129205",
73
+ "on_device_sampling": false,
74
+ "optimum_neuron_version": "0.4.5.dev2",
75
+ "output_logits": false,
76
+ "pp_degree": 1,
77
+ "sequence_length": 32768,
78
+ "speculation_length": 0,
79
+ "start_rank_id": 0,
80
+ "target": "trn1",
81
+ "torch_dtype": "bfloat16",
82
+ "tp_degree": 16
83
+ },
84
+ "num_attention_heads": 32,
85
+ "num_hidden_layers": 36,
86
+ "num_key_value_heads": 8,
87
+ "rms_norm_eps": 1e-06,
88
+ "rope_scaling": null,
89
+ "rope_theta": 1000000,
90
+ "sliding_window": null,
91
+ "tie_word_embeddings": false,
92
+ "use_cache": true,
93
+ "use_sliding_window": false,
94
+ "vocab_size": 151665
95
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev2/af58eb15d8e02338dc2f2e880e9c6ec803a98278914b3606acdcc252e7e18429/99d672cbaa2e018da4e4.json ADDED
@@ -0,0 +1,95 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen3-Embedding-8B",
4
+ "_task": "feature-extraction",
5
+ "architectures": [
6
+ "Qwen3ForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "dtype": "bfloat16",
11
+ "head_dim": 128,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 4096,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 12288,
16
+ "layer_types": [
17
+ "full_attention",
18
+ "full_attention",
19
+ "full_attention",
20
+ "full_attention",
21
+ "full_attention",
22
+ "full_attention",
23
+ "full_attention",
24
+ "full_attention",
25
+ "full_attention",
26
+ "full_attention",
27
+ "full_attention",
28
+ "full_attention",
29
+ "full_attention",
30
+ "full_attention",
31
+ "full_attention",
32
+ "full_attention",
33
+ "full_attention",
34
+ "full_attention",
35
+ "full_attention",
36
+ "full_attention",
37
+ "full_attention",
38
+ "full_attention",
39
+ "full_attention",
40
+ "full_attention",
41
+ "full_attention",
42
+ "full_attention",
43
+ "full_attention",
44
+ "full_attention",
45
+ "full_attention",
46
+ "full_attention",
47
+ "full_attention",
48
+ "full_attention",
49
+ "full_attention",
50
+ "full_attention",
51
+ "full_attention",
52
+ "full_attention"
53
+ ],
54
+ "max_position_embeddings": 40960,
55
+ "max_window_layers": 36,
56
+ "model_type": "qwen3",
57
+ "neuron": {
58
+ "_serialized_key": "NxDNeuronConfig",
59
+ "batch_size": 8,
60
+ "capacity_factor": null,
61
+ "checkpoint_id": "Qwen/Qwen3-Embedding-8B",
62
+ "checkpoint_revision": "1d8ad4ca9b3dd8059ad90a75d4983776a23d44af",
63
+ "continuous_batching": false,
64
+ "ep_degree": 1,
65
+ "fused_qkv": true,
66
+ "glu_mlp": true,
67
+ "local_ranks_size": 16,
68
+ "max_batch_size": 8,
69
+ "max_context_length": 16384,
70
+ "max_topk": 256,
71
+ "n_active_tokens": 16384,
72
+ "neuronxcc_version": "2.21.33363.0+82129205",
73
+ "on_device_sampling": false,
74
+ "optimum_neuron_version": "0.4.5.dev2",
75
+ "output_logits": false,
76
+ "pp_degree": 1,
77
+ "sequence_length": 16384,
78
+ "speculation_length": 0,
79
+ "start_rank_id": 0,
80
+ "target": "trn1",
81
+ "torch_dtype": "bfloat16",
82
+ "tp_degree": 16
83
+ },
84
+ "num_attention_heads": 32,
85
+ "num_hidden_layers": 36,
86
+ "num_key_value_heads": 8,
87
+ "rms_norm_eps": 1e-06,
88
+ "rope_scaling": null,
89
+ "rope_theta": 1000000,
90
+ "sliding_window": null,
91
+ "tie_word_embeddings": false,
92
+ "use_cache": true,
93
+ "use_sliding_window": false,
94
+ "vocab_size": 151665
95
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev2/af58eb15d8e02338dc2f2e880e9c6ec803a98278914b3606acdcc252e7e18429/ad354086b250f133c9c6.json ADDED
@@ -0,0 +1,95 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen3-Embedding-8B",
4
+ "_task": "feature-extraction",
5
+ "architectures": [
6
+ "Qwen3ForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "dtype": "bfloat16",
11
+ "head_dim": 128,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 4096,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 12288,
16
+ "layer_types": [
17
+ "full_attention",
18
+ "full_attention",
19
+ "full_attention",
20
+ "full_attention",
21
+ "full_attention",
22
+ "full_attention",
23
+ "full_attention",
24
+ "full_attention",
25
+ "full_attention",
26
+ "full_attention",
27
+ "full_attention",
28
+ "full_attention",
29
+ "full_attention",
30
+ "full_attention",
31
+ "full_attention",
32
+ "full_attention",
33
+ "full_attention",
34
+ "full_attention",
35
+ "full_attention",
36
+ "full_attention",
37
+ "full_attention",
38
+ "full_attention",
39
+ "full_attention",
40
+ "full_attention",
41
+ "full_attention",
42
+ "full_attention",
43
+ "full_attention",
44
+ "full_attention",
45
+ "full_attention",
46
+ "full_attention",
47
+ "full_attention",
48
+ "full_attention",
49
+ "full_attention",
50
+ "full_attention",
51
+ "full_attention",
52
+ "full_attention"
53
+ ],
54
+ "max_position_embeddings": 40960,
55
+ "max_window_layers": 36,
56
+ "model_type": "qwen3",
57
+ "neuron": {
58
+ "_serialized_key": "NxDNeuronConfig",
59
+ "batch_size": 4,
60
+ "capacity_factor": null,
61
+ "checkpoint_id": "Qwen/Qwen3-Embedding-8B",
62
+ "checkpoint_revision": "1d8ad4ca9b3dd8059ad90a75d4983776a23d44af",
63
+ "continuous_batching": false,
64
+ "ep_degree": 1,
65
+ "fused_qkv": true,
66
+ "glu_mlp": true,
67
+ "local_ranks_size": 16,
68
+ "max_batch_size": 4,
69
+ "max_context_length": 16384,
70
+ "max_topk": 256,
71
+ "n_active_tokens": 16384,
72
+ "neuronxcc_version": "2.21.33363.0+82129205",
73
+ "on_device_sampling": false,
74
+ "optimum_neuron_version": "0.4.5.dev2",
75
+ "output_logits": false,
76
+ "pp_degree": 1,
77
+ "sequence_length": 16384,
78
+ "speculation_length": 0,
79
+ "start_rank_id": 0,
80
+ "target": "trn1",
81
+ "torch_dtype": "bfloat16",
82
+ "tp_degree": 16
83
+ },
84
+ "num_attention_heads": 32,
85
+ "num_hidden_layers": 36,
86
+ "num_key_value_heads": 8,
87
+ "rms_norm_eps": 1e-06,
88
+ "rope_scaling": null,
89
+ "rope_theta": 1000000,
90
+ "sliding_window": null,
91
+ "tie_word_embeddings": false,
92
+ "use_cache": true,
93
+ "use_sliding_window": false,
94
+ "vocab_size": 151665
95
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev2/af58eb15d8e02338dc2f2e880e9c6ec803a98278914b3606acdcc252e7e18429/b8a28a3ac7bcba98b595.json ADDED
@@ -0,0 +1,95 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen3-Embedding-8B",
4
+ "_task": "feature-extraction",
5
+ "architectures": [
6
+ "Qwen3ForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "dtype": "bfloat16",
11
+ "head_dim": 128,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 4096,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 12288,
16
+ "layer_types": [
17
+ "full_attention",
18
+ "full_attention",
19
+ "full_attention",
20
+ "full_attention",
21
+ "full_attention",
22
+ "full_attention",
23
+ "full_attention",
24
+ "full_attention",
25
+ "full_attention",
26
+ "full_attention",
27
+ "full_attention",
28
+ "full_attention",
29
+ "full_attention",
30
+ "full_attention",
31
+ "full_attention",
32
+ "full_attention",
33
+ "full_attention",
34
+ "full_attention",
35
+ "full_attention",
36
+ "full_attention",
37
+ "full_attention",
38
+ "full_attention",
39
+ "full_attention",
40
+ "full_attention",
41
+ "full_attention",
42
+ "full_attention",
43
+ "full_attention",
44
+ "full_attention",
45
+ "full_attention",
46
+ "full_attention",
47
+ "full_attention",
48
+ "full_attention",
49
+ "full_attention",
50
+ "full_attention",
51
+ "full_attention",
52
+ "full_attention"
53
+ ],
54
+ "max_position_embeddings": 40960,
55
+ "max_window_layers": 36,
56
+ "model_type": "qwen3",
57
+ "neuron": {
58
+ "_serialized_key": "NxDNeuronConfig",
59
+ "batch_size": 8,
60
+ "capacity_factor": null,
61
+ "checkpoint_id": "Qwen/Qwen3-Embedding-8B",
62
+ "checkpoint_revision": "1d8ad4ca9b3dd8059ad90a75d4983776a23d44af",
63
+ "continuous_batching": false,
64
+ "ep_degree": 1,
65
+ "fused_qkv": true,
66
+ "glu_mlp": true,
67
+ "local_ranks_size": 16,
68
+ "max_batch_size": 8,
69
+ "max_context_length": 8192,
70
+ "max_topk": 256,
71
+ "n_active_tokens": 8192,
72
+ "neuronxcc_version": "2.21.33363.0+82129205",
73
+ "on_device_sampling": false,
74
+ "optimum_neuron_version": "0.4.5.dev2",
75
+ "output_logits": false,
76
+ "pp_degree": 1,
77
+ "sequence_length": 8192,
78
+ "speculation_length": 0,
79
+ "start_rank_id": 0,
80
+ "target": "trn1",
81
+ "torch_dtype": "bfloat16",
82
+ "tp_degree": 16
83
+ },
84
+ "num_attention_heads": 32,
85
+ "num_hidden_layers": 36,
86
+ "num_key_value_heads": 8,
87
+ "rms_norm_eps": 1e-06,
88
+ "rope_scaling": null,
89
+ "rope_theta": 1000000,
90
+ "sliding_window": null,
91
+ "tie_word_embeddings": false,
92
+ "use_cache": true,
93
+ "use_sliding_window": false,
94
+ "vocab_size": 151665
95
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev2/af58eb15d8e02338dc2f2e880e9c6ec803a98278914b3606acdcc252e7e18429/fa7d033eb1dca73fe9c3.json ADDED
@@ -0,0 +1,95 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen3-Embedding-8B",
4
+ "_task": "feature-extraction",
5
+ "architectures": [
6
+ "Qwen3ForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "dtype": "bfloat16",
11
+ "head_dim": 128,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 4096,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 12288,
16
+ "layer_types": [
17
+ "full_attention",
18
+ "full_attention",
19
+ "full_attention",
20
+ "full_attention",
21
+ "full_attention",
22
+ "full_attention",
23
+ "full_attention",
24
+ "full_attention",
25
+ "full_attention",
26
+ "full_attention",
27
+ "full_attention",
28
+ "full_attention",
29
+ "full_attention",
30
+ "full_attention",
31
+ "full_attention",
32
+ "full_attention",
33
+ "full_attention",
34
+ "full_attention",
35
+ "full_attention",
36
+ "full_attention",
37
+ "full_attention",
38
+ "full_attention",
39
+ "full_attention",
40
+ "full_attention",
41
+ "full_attention",
42
+ "full_attention",
43
+ "full_attention",
44
+ "full_attention",
45
+ "full_attention",
46
+ "full_attention",
47
+ "full_attention",
48
+ "full_attention",
49
+ "full_attention",
50
+ "full_attention",
51
+ "full_attention",
52
+ "full_attention"
53
+ ],
54
+ "max_position_embeddings": 40960,
55
+ "max_window_layers": 36,
56
+ "model_type": "qwen3",
57
+ "neuron": {
58
+ "_serialized_key": "NxDNeuronConfig",
59
+ "batch_size": 16,
60
+ "capacity_factor": null,
61
+ "checkpoint_id": "Qwen/Qwen3-Embedding-8B",
62
+ "checkpoint_revision": "1d8ad4ca9b3dd8059ad90a75d4983776a23d44af",
63
+ "continuous_batching": false,
64
+ "ep_degree": 1,
65
+ "fused_qkv": true,
66
+ "glu_mlp": true,
67
+ "local_ranks_size": 16,
68
+ "max_batch_size": 16,
69
+ "max_context_length": 4096,
70
+ "max_topk": 256,
71
+ "n_active_tokens": 4096,
72
+ "neuronxcc_version": "2.21.33363.0+82129205",
73
+ "on_device_sampling": false,
74
+ "optimum_neuron_version": "0.4.5.dev2",
75
+ "output_logits": false,
76
+ "pp_degree": 1,
77
+ "sequence_length": 4096,
78
+ "speculation_length": 0,
79
+ "start_rank_id": 0,
80
+ "target": "trn1",
81
+ "torch_dtype": "bfloat16",
82
+ "tp_degree": 16
83
+ },
84
+ "num_attention_heads": 32,
85
+ "num_hidden_layers": 36,
86
+ "num_key_value_heads": 8,
87
+ "rms_norm_eps": 1e-06,
88
+ "rope_scaling": null,
89
+ "rope_theta": 1000000,
90
+ "sliding_window": null,
91
+ "tie_word_embeddings": false,
92
+ "use_cache": true,
93
+ "use_sliding_window": false,
94
+ "vocab_size": 151665
95
+ }
neuronxcc-2.21.33363.0+82129205/MODULE_1a3939beb88e18a29e1e+fb4cc044/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:49dae5bb2894dc0c7ea22a320b4b2de1774e3f5b385da2dfd4bee2d6f0ab9d0e
3
  size 855517
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f4962df721f1ff99db7ed9282a0413c045e2ed26f72534f5bac1b8e92cc90ab6
3
  size 855517
neuronxcc-2.21.33363.0+82129205/MODULE_1a41189de75fc0ed15bb+fb4cc044/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3ae3632a5072d91624b76d190b5435e08c58ad0d694bbb070627e66714c00d34
3
  size 854933
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:30d9f50b25c8ca835fdfe8f59ab1803840811b018c6c0ca09e8cd8e73ad1cc23
3
  size 854933
neuronxcc-2.21.33363.0+82129205/MODULE_1afb27ccf10708e03947+fb4cc044/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c6c082016fb78c0694a6b26131ba3769d0366bdd1be75fcf052dc0599f9da811
3
  size 847335
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe09b160df7c75422ed8c70794cba8f700532ea5e055a888f3218d4ddc133cc2
3
  size 847335
neuronxcc-2.21.33363.0+82129205/MODULE_269a21b3e3cb06fc787d+fb4cc044/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ce7965a675031fb6ab8179087e299eef62fde401cc47e3276351e8e7ff0f2022
3
  size 848819
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:510d77f70ca853cb4b1c6a5c212deeac8665a2247e19d253089cc6baad1bfe5e
3
  size 848819
neuronxcc-2.21.33363.0+82129205/MODULE_2831e4c199ca2002f484+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_2831e4c199ca2002f484+24129607/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_2831e4c199ca2002f484+24129607/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d6586f1806f42425bfaa3fdfabde1ed3e812e471c3fb402ff55d086a77b2fd77
3
+ size 964631
neuronxcc-2.21.33363.0+82129205/MODULE_2831e4c199ca2002f484+24129607/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:872e6e30a502e8a2b1b521ead33019131caf567e16724954de06860e9920423a
3
+ size 9585664
neuronxcc-2.21.33363.0+82129205/MODULE_35958e2f8cf19e5eddc6+fb4cc044/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/encoding/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.21.33363.0+82129205/MODULE_35958e2f8cf19e5eddc6+fb4cc044/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_35958e2f8cf19e5eddc6+fb4cc044/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:332badac6d5178793a73d00fb629bea96371d302c0a2c06dd5f7b839fb48ca69
3
+ size 846674
neuronxcc-2.21.33363.0+82129205/MODULE_35958e2f8cf19e5eddc6+fb4cc044/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f4d30fbd63a17bc6b510ec148961ccb47b7572e3d23faa33b5bedc8ccb1257ef
3
+ size 18801664
neuronxcc-2.21.33363.0+82129205/MODULE_54baea067a19d8c74866+fb4cc044/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1ab1648fdab97bf68365e1aa5973e5b6031c2705415d4fd1b323d85ddf5087ee
3
  size 859976
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:294ed9b3de40b9ff7fc577e30fca96ed67140aaf4a582099c8372ca196a5d0a8
3
  size 859976
neuronxcc-2.21.33363.0+82129205/MODULE_5c4a6aceb0aa544053e8+fb4cc044/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a9afd9b922494f37910fd123b49a349abe77e409cd972f1235a17fd3f9528a78
3
  size 857165
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:052b3e66105967fa30b2f55fcb561b2c9006d27f850cb49cf74c4bdbfef09d5c
3
  size 857165
neuronxcc-2.21.33363.0+82129205/MODULE_5dafb05b2fa15b606ad4+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_5dafb05b2fa15b606ad4+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_5dafb05b2fa15b606ad4+a02c3a36/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3be6595e498a428334b25dd2f1c128bc34c04dfd8b3f7ddb7b08c1124ebce3b3
3
+ size 978680
neuronxcc-2.21.33363.0+82129205/MODULE_5dafb05b2fa15b606ad4+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:52670f09ed33bad9c7f72e08060928342d842d55c0cb3e24b2976fa91c926b98
3
+ size 4680704
neuronxcc-2.21.33363.0+82129205/MODULE_5dafb05b2fa15b606ad4+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:836aba59ceabd1cd7e53d2b75579736fd1bbf20405f14c62fdf85c56cbcb312c
3
+ size 4855395
neuronxcc-2.21.33363.0+82129205/MODULE_63b228a1aedd2e24d261+fb4cc044/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f5a83949de28fde124a6106707f8c43c8434bb6c9497076254c57018052a7a49
3
  size 635925
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:64cf5020409415c8e7ac78ea6885a166b6fedefb10771a68c14ba7704c559b0e
3
  size 635925
neuronxcc-2.21.33363.0+82129205/MODULE_64886f8b7709113eb14b+fb4cc044/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/encoding/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.21.33363.0+82129205/MODULE_64886f8b7709113eb14b+fb4cc044/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_64886f8b7709113eb14b+fb4cc044/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cfe53cc824521d9307b6a7bd6d11b686fd303c8fde5836aaf2486c21598a2108
3
+ size 846739
neuronxcc-2.21.33363.0+82129205/MODULE_64886f8b7709113eb14b+fb4cc044/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:83f93438c0e69e302e132b757c2d8cd6cdab76bf163ab872cf7493f09172ccb1
3
+ size 26665984
neuronxcc-2.21.33363.0+82129205/MODULE_758311bdb6e777cb53a2+fb4cc044/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/encoding/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.21.33363.0+82129205/MODULE_758311bdb6e777cb53a2+fb4cc044/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_758311bdb6e777cb53a2+fb4cc044/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:843db1ac96310937ce94dd8aa7867f17bd07233477d0bd666d88c861ab033274
3
+ size 839280
neuronxcc-2.21.33363.0+82129205/MODULE_758311bdb6e777cb53a2+fb4cc044/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b99e257a60373b0a0fad68c538aab65902f9519537fec23a7e65fc72d329da42
3
+ size 17705984
neuronxcc-2.21.33363.0+82129205/MODULE_81503bf195cd5d2e22b3+fb4cc044/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:155c6e081c7eb1c61ece71fa77412b3525882a84d8a82f0347908930458bf305
3
  size 847459
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7dbf1d0c8b099ae011b2f2c6c9098ccad692b88500c77060395fc45117bb57d6
3
  size 847459
neuronxcc-2.21.33363.0+82129205/MODULE_87ac40e3234811a067b9+fb4cc044/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3e45c384167132d812fe4c7c2ee3cae54466a96795a354027643ec4f3d9650f8
3
  size 840640
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d88d7a8112d0c699bc222e2840023091b282cc19dbc8ec4db56c2f16b3ee423d
3
  size 840640
neuronxcc-2.21.33363.0+82129205/MODULE_8d8d38359e9e9f052a77+fb4cc044/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:193b34e4fbfb73b0e96c027774b7518fc7978c624a2c8d92e76a7b6b110bec34
3
  size 859109
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5a2529d821ddf64b20d4049d1f86abfc2fb1243068aac24799f77eb0ecc109e1
3
  size 859109
neuronxcc-2.21.33363.0+82129205/MODULE_9a666844f4acef43578b+fb4cc044/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e07f46d157553a8a739c6e9f5624d2082b1a1a0d09b34b31b5e1f658670c74f6
3
  size 840640
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:22b506d19399abfcc0502e27c2eed2a9e9f03b475ab85185faa2d3cacbfc5b81
3
  size 840640
neuronxcc-2.21.33363.0+82129205/MODULE_aaefb5ede5ee0772566c+fb4cc044/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/encoding/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.21.33363.0+82129205/MODULE_aaefb5ede5ee0772566c+fb4cc044/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_aaefb5ede5ee0772566c+fb4cc044/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ad0b9c4ee4906c651bfa256315106235b617658068447866ee0587cfd3435969
3
+ size 846739
neuronxcc-2.21.33363.0+82129205/MODULE_aaefb5ede5ee0772566c+fb4cc044/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e1947873efe9a373e8016e55285025c7e1dc5af4d006a4b885a1927e4e1e416f
3
+ size 75367424
neuronxcc-2.21.33363.0+82129205/MODULE_b0cd0a9c4d91cfdefe87+fb4cc044/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/encoding/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.21.33363.0+82129205/MODULE_b0cd0a9c4d91cfdefe87+fb4cc044/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_b0cd0a9c4d91cfdefe87+fb4cc044/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a796666ba49f04593eaab9a45bcc19a3fefde4129103779a7c65887c9ede250b
3
+ size 847459
neuronxcc-2.21.33363.0+82129205/MODULE_b0cd0a9c4d91cfdefe87+fb4cc044/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:88cef71738055a8afd22c9bcb71943a1bcb314cdca7c2bfab354c8e197455852
3
+ size 53105664
neuronxcc-2.21.33363.0+82129205/MODULE_b234477c449ee671d3c1+fb4cc044/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/encoding/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.21.33363.0+82129205/MODULE_b234477c449ee671d3c1+fb4cc044/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9f55bce31e2660b08a6b612b7aec51341271e6818bb861f8b484603c45b6cebc
3
+ size 847459