Hugging Face's logo

optimum-internal-testing
/
neuron-testing-cache

Model card Files
xet
 Community
dacorvo HF Staff commited on
Commit
fa163e9
·
verified ·
1 Parent(s): beb8835

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 +18 -0
  2. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev1/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/de98fc65f348a2d8c295.json +59 -0
  3. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev1/llama/llamafactory/tiny-random-Llama-3/dad125ca8a11952d0a3f.json +63 -0
  4. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev1/llama4_text/tiny-random/llama-4/cf5cee8c9de5d6faf940.json +82 -0
  5. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev1/mixtral/dacorvo/Mixtral-tiny/3779219bc59a2ae3e045.json +59 -0
  6. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev1/phi3/yujiepan/phi-4-tiny-random/2ebdbeae8a00090a231f.json +60 -0
  7. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev1/qwen2/yujiepan/qwen2.5-128k-tiny-random/e6f05187d051ac3c3e46.json +65 -0
  8. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev1/qwen3/Qwen/Qwen3-Embedding-0.6B/18af48e9de9305ddddd5.json +88 -0
  9. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev1/qwen3/Qwen/Qwen3-Embedding-0.6B/80594a5958040f8b1ebe.json +88 -0
  10. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev1/qwen3_moe/optimum-internal-testing/tiny-random-qwen3_moe/a5cf4ee087cd9567449f.json +66 -0
  11. neuronxcc-2.21.33363.0+82129205/MODULE_01328268a19023ed0f24+f8e6d902/model.hlo_module.pb +1 -1
  12. neuronxcc-2.21.33363.0+82129205/MODULE_01328268a19023ed0f24+f8e6d902/model.neff +1 -1
  13. neuronxcc-2.21.33363.0+82129205/MODULE_0efb0afa4ef56e94d498+a02c3a36/compile_flags.json +1 -0
  14. neuronxcc-2.21.33363.0+82129205/MODULE_0efb0afa4ef56e94d498+a02c3a36/model.done +0 -0
  15. neuronxcc-2.21.33363.0+82129205/MODULE_0efb0afa4ef56e94d498+a02c3a36/model.hlo_module.pb +3 -0
  16. neuronxcc-2.21.33363.0+82129205/MODULE_0efb0afa4ef56e94d498+a02c3a36/model.neff +3 -0
  17. neuronxcc-2.21.33363.0+82129205/MODULE_0efb0afa4ef56e94d498+a02c3a36/wrapped_neff.hlo +3 -0
  18. neuronxcc-2.21.33363.0+82129205/MODULE_14cda09d048ccfb252f8+a02c3a36/compile_flags.json +1 -0
  19. neuronxcc-2.21.33363.0+82129205/MODULE_14cda09d048ccfb252f8+a02c3a36/model.done +0 -0
  20. neuronxcc-2.21.33363.0+82129205/MODULE_14cda09d048ccfb252f8+a02c3a36/model.hlo_module.pb +3 -0
  21. neuronxcc-2.21.33363.0+82129205/MODULE_14cda09d048ccfb252f8+a02c3a36/model.neff +3 -0
  22. neuronxcc-2.21.33363.0+82129205/MODULE_14cda09d048ccfb252f8+a02c3a36/wrapped_neff.hlo +3 -0
  23. neuronxcc-2.21.33363.0+82129205/MODULE_245d388b7b8c64a4d846+a02c3a36/compile_flags.json +1 -0
  24. neuronxcc-2.21.33363.0+82129205/MODULE_245d388b7b8c64a4d846+a02c3a36/model.done +0 -0
  25. neuronxcc-2.21.33363.0+82129205/MODULE_245d388b7b8c64a4d846+a02c3a36/model.hlo_module.pb +3 -0
  26. neuronxcc-2.21.33363.0+82129205/MODULE_245d388b7b8c64a4d846+a02c3a36/model.neff +3 -0
  27. neuronxcc-2.21.33363.0+82129205/MODULE_245d388b7b8c64a4d846+a02c3a36/wrapped_neff.hlo +3 -0
  28. neuronxcc-2.21.33363.0+82129205/MODULE_2672b897bf6327bea9e4+24129607/model.hlo_module.pb +1 -1
  29. neuronxcc-2.21.33363.0+82129205/MODULE_2672b897bf6327bea9e4+24129607/model.neff +1 -1
  30. neuronxcc-2.21.33363.0+82129205/MODULE_3a758a7122ca68a1243a+a02c3a36/compile_flags.json +1 -0
  31. neuronxcc-2.21.33363.0+82129205/MODULE_3a758a7122ca68a1243a+a02c3a36/model.done +0 -0
  32. neuronxcc-2.21.33363.0+82129205/MODULE_3a758a7122ca68a1243a+a02c3a36/model.hlo_module.pb +3 -0
  33. neuronxcc-2.21.33363.0+82129205/MODULE_3a758a7122ca68a1243a+a02c3a36/model.neff +3 -0
  34. neuronxcc-2.21.33363.0+82129205/MODULE_3a758a7122ca68a1243a+a02c3a36/wrapped_neff.hlo +3 -0
  35. neuronxcc-2.21.33363.0+82129205/MODULE_5960bdedd7549bfacf7a+24129607/model.hlo_module.pb +1 -1
  36. neuronxcc-2.21.33363.0+82129205/MODULE_5960bdedd7549bfacf7a+24129607/model.neff +1 -1
  37. neuronxcc-2.21.33363.0+82129205/MODULE_5ee5dade1050cf89f316+a02c3a36/compile_flags.json +1 -0
  38. neuronxcc-2.21.33363.0+82129205/MODULE_5ee5dade1050cf89f316+a02c3a36/model.done +0 -0
  39. neuronxcc-2.21.33363.0+82129205/MODULE_5ee5dade1050cf89f316+a02c3a36/model.hlo_module.pb +3 -0
  40. neuronxcc-2.21.33363.0+82129205/MODULE_5ee5dade1050cf89f316+a02c3a36/model.neff +3 -0
  41. neuronxcc-2.21.33363.0+82129205/MODULE_5ee5dade1050cf89f316+a02c3a36/wrapped_neff.hlo +3 -0
  42. neuronxcc-2.21.33363.0+82129205/MODULE_864a283be77e0814947c+bafdbdde/compile_flags.json +1 -0
  43. neuronxcc-2.21.33363.0+82129205/MODULE_864a283be77e0814947c+bafdbdde/model.done +0 -0
  44. neuronxcc-2.21.33363.0+82129205/MODULE_864a283be77e0814947c+bafdbdde/model.hlo_module.pb +3 -0
  45. neuronxcc-2.21.33363.0+82129205/MODULE_864a283be77e0814947c+bafdbdde/model.neff +3 -0
  46. neuronxcc-2.21.33363.0+82129205/MODULE_864a283be77e0814947c+bafdbdde/wrapped_neff.hlo +3 -0
  47. neuronxcc-2.21.33363.0+82129205/MODULE_8b2505504a54575836b0+a02c3a36/compile_flags.json +1 -0
  48. neuronxcc-2.21.33363.0+82129205/MODULE_8b2505504a54575836b0+a02c3a36/model.done +0 -0
  49. neuronxcc-2.21.33363.0+82129205/MODULE_8b2505504a54575836b0+a02c3a36/model.hlo_module.pb +3 -0
  50. neuronxcc-2.21.33363.0+82129205/MODULE_8b2505504a54575836b0+a02c3a36/model.neff +3 -0
.gitattributes CHANGED
@@ -5827,3 +5827,21 @@ neuronxcc-2.21.33363.0+82129205/MODULE_cb05698d60b06f387ed6+a02c3a36/wrapped_nef
5827
  neuronxcc-2.21.33363.0+82129205/MODULE_4cfdf541f39ee43e568e+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
5828
  neuronxcc-2.21.33363.0+82129205/MODULE_4cfdf541f39ee43e568e+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
5829
  neuronxcc-2.21.33363.0+82129205/MODULE_8c6b204d1bccf64885d0+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
5827
  neuronxcc-2.21.33363.0+82129205/MODULE_4cfdf541f39ee43e568e+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
5828
  neuronxcc-2.21.33363.0+82129205/MODULE_4cfdf541f39ee43e568e+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
5829
  neuronxcc-2.21.33363.0+82129205/MODULE_8c6b204d1bccf64885d0+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
5830
+ neuronxcc-2.21.33363.0+82129205/MODULE_0efb0afa4ef56e94d498+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
5831
+ neuronxcc-2.21.33363.0+82129205/MODULE_0efb0afa4ef56e94d498+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
5832
+ neuronxcc-2.21.33363.0+82129205/MODULE_14cda09d048ccfb252f8+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
5833
+ neuronxcc-2.21.33363.0+82129205/MODULE_14cda09d048ccfb252f8+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
5834
+ neuronxcc-2.21.33363.0+82129205/MODULE_245d388b7b8c64a4d846+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
5835
+ neuronxcc-2.21.33363.0+82129205/MODULE_245d388b7b8c64a4d846+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
5836
+ neuronxcc-2.21.33363.0+82129205/MODULE_3a758a7122ca68a1243a+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
5837
+ neuronxcc-2.21.33363.0+82129205/MODULE_3a758a7122ca68a1243a+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
5838
+ neuronxcc-2.21.33363.0+82129205/MODULE_5ee5dade1050cf89f316+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
5839
+ neuronxcc-2.21.33363.0+82129205/MODULE_5ee5dade1050cf89f316+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
5840
+ neuronxcc-2.21.33363.0+82129205/MODULE_864a283be77e0814947c+bafdbdde/model.neff filter=lfs diff=lfs merge=lfs -text
5841
+ neuronxcc-2.21.33363.0+82129205/MODULE_864a283be77e0814947c+bafdbdde/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
5842
+ neuronxcc-2.21.33363.0+82129205/MODULE_8b2505504a54575836b0+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
5843
+ neuronxcc-2.21.33363.0+82129205/MODULE_8b2505504a54575836b0+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
5844
+ neuronxcc-2.21.33363.0+82129205/MODULE_c4913590384814c92cb6+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
5845
+ neuronxcc-2.21.33363.0+82129205/MODULE_c4913590384814c92cb6+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
5846
+ neuronxcc-2.21.33363.0+82129205/MODULE_e0ccc071060e441622c2+bafdbdde/model.neff filter=lfs diff=lfs merge=lfs -text
5847
+ neuronxcc-2.21.33363.0+82129205/MODULE_e0ccc071060e441622c2+bafdbdde/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev1/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/de98fc65f348a2d8c295.json ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "hf-internal-testing/tiny-random-GraniteForCausalLM",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "GraniteForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "attention_multiplier": 1.0,
11
+ "dtype": "float32",
12
+ "embedding_multiplier": 1.0,
13
+ "hidden_act": "silu",
14
+ "hidden_size": 32,
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 64,
17
+ "logits_scaling": 1.0,
18
+ "max_position_embeddings": 2048,
19
+ "mlp_bias": false,
20
+ "model_type": "granite",
21
+ "neuron": {
22
+ "_serialized_key": "NxDNeuronConfig",
23
+ "batch_size": 1,
24
+ "capacity_factor": null,
25
+ "checkpoint_id": "hf-internal-testing/tiny-random-GraniteForCausalLM",
26
+ "checkpoint_revision": "c3074ebc0ac2fe545305f5e5f6cce2cc9b2aa0c5",
27
+ "continuous_batching": false,
28
+ "ep_degree": 1,
29
+ "fused_qkv": true,
30
+ "glu_mlp": true,
31
+ "local_ranks_size": 2,
32
+ "max_batch_size": 1,
33
+ "max_context_length": 1024,
34
+ "max_topk": 256,
35
+ "n_active_tokens": 1024,
36
+ "neuronxcc_version": "2.21.33363.0+82129205",
37
+ "on_device_sampling": true,
38
+ "optimum_neuron_version": "0.4.4.dev1",
39
+ "output_logits": false,
40
+ "pp_degree": 1,
41
+ "sequence_length": 1024,
42
+ "sequence_parallel_enabled": false,
43
+ "speculation_length": 0,
44
+ "start_rank_id": 0,
45
+ "target": "trn1",
46
+ "torch_dtype": "float32",
47
+ "tp_degree": 2
48
+ },
49
+ "num_attention_heads": 4,
50
+ "num_hidden_layers": 2,
51
+ "num_key_value_heads": 4,
52
+ "residual_multiplier": 1.0,
53
+ "rms_norm_eps": 1e-06,
54
+ "rope_scaling": null,
55
+ "rope_theta": 10000.0,
56
+ "tie_word_embeddings": false,
57
+ "use_cache": true,
58
+ "vocab_size": 49152
59
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev1/llama/llamafactory/tiny-random-Llama-3/dad125ca8a11952d0a3f.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "llamafactory/tiny-random-Llama-3",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "LlamaForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "dtype": "float16",
11
+ "head_dim": 4,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 16,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 64,
16
+ "max_position_embeddings": 131072,
17
+ "mlp_bias": false,
18
+ "model_type": "llama",
19
+ "neuron": {
20
+ "_serialized_key": "NxDNeuronConfig",
21
+ "batch_size": 1,
22
+ "capacity_factor": null,
23
+ "checkpoint_id": "llamafactory/tiny-random-Llama-3",
24
+ "checkpoint_revision": "bf2a2e3bf199ad2ee96f02a3c00246c608db22a8",
25
+ "continuous_batching": false,
26
+ "ep_degree": 1,
27
+ "fused_qkv": true,
28
+ "glu_mlp": true,
29
+ "local_ranks_size": 2,
30
+ "max_batch_size": 1,
31
+ "max_context_length": 1024,
32
+ "max_topk": 256,
33
+ "n_active_tokens": 1024,
34
+ "neuronxcc_version": "2.21.33363.0+82129205",
35
+ "on_device_sampling": true,
36
+ "optimum_neuron_version": "0.4.4.dev1",
37
+ "output_logits": false,
38
+ "pp_degree": 1,
39
+ "sequence_length": 1024,
40
+ "sequence_parallel_enabled": false,
41
+ "speculation_length": 0,
42
+ "start_rank_id": 0,
43
+ "target": "trn1",
44
+ "torch_dtype": "float16",
45
+ "tp_degree": 2
46
+ },
47
+ "num_attention_heads": 4,
48
+ "num_hidden_layers": 2,
49
+ "num_key_value_heads": 4,
50
+ "pretraining_tp": 1,
51
+ "rms_norm_eps": 1e-05,
52
+ "rope_scaling": {
53
+ "factor": 8.0,
54
+ "high_freq_factor": 4.0,
55
+ "low_freq_factor": 1.0,
56
+ "original_max_position_embeddings": 8192,
57
+ "rope_type": "llama3"
58
+ },
59
+ "rope_theta": 500000.0,
60
+ "tie_word_embeddings": false,
61
+ "use_cache": true,
62
+ "vocab_size": 128256
63
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev1/llama4_text/tiny-random/llama-4/cf5cee8c9de5d6faf940.json ADDED
@@ -0,0 +1,82 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "tiny-random/llama-4",
4
+ "_task": "text-generation",
5
+ "attention_bias": false,
6
+ "attention_chunk_size": 128,
7
+ "attention_dropout": 0.0,
8
+ "attn_scale": 0.1,
9
+ "attn_temperature_tuning": 4,
10
+ "cache_implementation": "hybrid",
11
+ "dtype": "bfloat16",
12
+ "floor_scale": 8192,
13
+ "for_llm_compressor": false,
14
+ "head_dim": 32,
15
+ "hidden_act": "silu",
16
+ "hidden_size": 32,
17
+ "initializer_range": 0.02,
18
+ "interleave_moe_layer_step": 2,
19
+ "intermediate_size": 64,
20
+ "intermediate_size_mlp": 128,
21
+ "layer_types": [
22
+ "chunked_attention",
23
+ "chunked_attention",
24
+ "chunked_attention",
25
+ "full_attention"
26
+ ],
27
+ "max_position_embeddings": 1048576,
28
+ "model_type": "llama4_text",
29
+ "moe_layers": [
30
+ 1,
31
+ 3
32
+ ],
33
+ "neuron": {
34
+ "_serialized_key": "NxDNeuronConfig",
35
+ "batch_size": 1,
36
+ "capacity_factor": null,
37
+ "checkpoint_id": "tiny-random/llama-4",
38
+ "checkpoint_revision": "9e716f5d4d1ffe0a44a15f46f4a12b840439aba4",
39
+ "continuous_batching": false,
40
+ "ep_degree": 1,
41
+ "fused_qkv": false,
42
+ "glu_mlp": true,
43
+ "local_ranks_size": 2,
44
+ "max_batch_size": 1,
45
+ "max_context_length": 1024,
46
+ "max_topk": 256,
47
+ "n_active_tokens": 1024,
48
+ "neuronxcc_version": "2.21.33363.0+82129205",
49
+ "on_device_sampling": true,
50
+ "optimum_neuron_version": "0.4.4.dev1",
51
+ "output_logits": false,
52
+ "pp_degree": 1,
53
+ "sequence_length": 1024,
54
+ "sequence_parallel_enabled": false,
55
+ "speculation_length": 0,
56
+ "start_rank_id": 0,
57
+ "target": "trn1",
58
+ "torch_dtype": "bfloat16",
59
+ "tp_degree": 2
60
+ },
61
+ "no_rope_layers": [
62
+ 1,
63
+ 1,
64
+ 1,
65
+ 0
66
+ ],
67
+ "num_attention_heads": 1,
68
+ "num_experts_per_tok": 1,
69
+ "num_hidden_layers": 4,
70
+ "num_key_value_heads": 1,
71
+ "num_local_experts": 8,
72
+ "output_router_logits": false,
73
+ "rms_norm_eps": 1e-05,
74
+ "rope_scaling": null,
75
+ "rope_theta": 500000.0,
76
+ "router_aux_loss_coef": 0.001,
77
+ "router_jitter_noise": 0.0,
78
+ "tie_word_embeddings": true,
79
+ "use_cache": true,
80
+ "use_qk_norm": true,
81
+ "vocab_size": 202048
82
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev1/mixtral/dacorvo/Mixtral-tiny/3779219bc59a2ae3e045.json ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "dacorvo/Mixtral-tiny",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "MixtralForCausalLM"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "dtype": "float16",
10
+ "head_dim": 32,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 1024,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 3584,
15
+ "max_position_embeddings": 1024,
16
+ "model_type": "mixtral",
17
+ "neuron": {
18
+ "_serialized_key": "NxDNeuronConfig",
19
+ "batch_size": 1,
20
+ "capacity_factor": null,
21
+ "checkpoint_id": "dacorvo/Mixtral-tiny",
22
+ "checkpoint_revision": "c557ba205ddff6ea911f4719e0d543d6c08356b6",
23
+ "continuous_batching": false,
24
+ "ep_degree": 1,
25
+ "fused_qkv": false,
26
+ "glu_mlp": true,
27
+ "local_ranks_size": 2,
28
+ "max_batch_size": 1,
29
+ "max_context_length": 1024,
30
+ "max_topk": 256,
31
+ "n_active_tokens": 1024,
32
+ "neuronxcc_version": "2.21.33363.0+82129205",
33
+ "on_device_sampling": false,
34
+ "optimum_neuron_version": "0.4.4.dev1",
35
+ "output_logits": false,
36
+ "pp_degree": 1,
37
+ "sequence_length": 1024,
38
+ "sequence_parallel_enabled": false,
39
+ "speculation_length": 0,
40
+ "start_rank_id": 0,
41
+ "target": "trn1",
42
+ "torch_dtype": "float16",
43
+ "tp_degree": 2
44
+ },
45
+ "num_attention_heads": 32,
46
+ "num_experts_per_tok": 2,
47
+ "num_hidden_layers": 2,
48
+ "num_key_value_heads": 8,
49
+ "num_local_experts": 8,
50
+ "output_router_logits": false,
51
+ "rms_norm_eps": 1e-05,
52
+ "rope_theta": 10000.0,
53
+ "router_aux_loss_coef": 0.001,
54
+ "router_jitter_noise": 0.0,
55
+ "sliding_window": 4096,
56
+ "tie_word_embeddings": false,
57
+ "use_cache": true,
58
+ "vocab_size": 32000
59
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev1/phi3/yujiepan/phi-4-tiny-random/2ebdbeae8a00090a231f.json ADDED
@@ -0,0 +1,60 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "yujiepan/phi-4-tiny-random",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Phi3ForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "auto_map": {},
11
+ "dtype": "bfloat16",
12
+ "embd_pdrop": 0.0,
13
+ "hidden_act": "silu",
14
+ "hidden_size": 16,
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 32,
17
+ "max_position_embeddings": 16384,
18
+ "model_type": "phi3",
19
+ "neuron": {
20
+ "_serialized_key": "NxDNeuronConfig",
21
+ "batch_size": 1,
22
+ "capacity_factor": null,
23
+ "checkpoint_id": "yujiepan/phi-4-tiny-random",
24
+ "checkpoint_revision": "18a9a1168dc97ac6d128f811925670c275610f5a",
25
+ "continuous_batching": false,
26
+ "ep_degree": 1,
27
+ "fused_qkv": true,
28
+ "glu_mlp": true,
29
+ "local_ranks_size": 2,
30
+ "max_batch_size": 1,
31
+ "max_context_length": 1024,
32
+ "max_topk": 256,
33
+ "n_active_tokens": 1024,
34
+ "neuronxcc_version": "2.21.33363.0+82129205",
35
+ "on_device_sampling": true,
36
+ "optimum_neuron_version": "0.4.4.dev1",
37
+ "output_logits": false,
38
+ "pp_degree": 1,
39
+ "sequence_length": 1024,
40
+ "sequence_parallel_enabled": false,
41
+ "speculation_length": 0,
42
+ "start_rank_id": 0,
43
+ "target": "trn1",
44
+ "torch_dtype": "bfloat16",
45
+ "tp_degree": 2
46
+ },
47
+ "num_attention_heads": 2,
48
+ "num_hidden_layers": 2,
49
+ "num_key_value_heads": 1,
50
+ "original_max_position_embeddings": 16384,
51
+ "partial_rotary_factor": 1.0,
52
+ "resid_pdrop": 0.0,
53
+ "rms_norm_eps": 1e-05,
54
+ "rope_scaling": null,
55
+ "rope_theta": 250000,
56
+ "sliding_window": null,
57
+ "tie_word_embeddings": false,
58
+ "use_cache": true,
59
+ "vocab_size": 100352
60
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev1/qwen2/yujiepan/qwen2.5-128k-tiny-random/e6f05187d051ac3c3e46.json ADDED
@@ -0,0 +1,65 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "yujiepan/qwen2.5-128k-tiny-random",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Qwen2ForCausalLM"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "dtype": "bfloat16",
10
+ "hidden_act": "silu",
11
+ "hidden_size": 8,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 16,
14
+ "layer_types": [
15
+ "full_attention",
16
+ "full_attention"
17
+ ],
18
+ "max_position_embeddings": 32768,
19
+ "max_window_layers": 1,
20
+ "model_type": "qwen2",
21
+ "neuron": {
22
+ "_serialized_key": "NxDNeuronConfig",
23
+ "batch_size": 1,
24
+ "capacity_factor": null,
25
+ "checkpoint_id": "yujiepan/qwen2.5-128k-tiny-random",
26
+ "checkpoint_revision": "c8296d4ca3f87782876d2382fbb6481d1beb8ef0",
27
+ "continuous_batching": false,
28
+ "ep_degree": 1,
29
+ "fused_qkv": false,
30
+ "glu_mlp": true,
31
+ "local_ranks_size": 2,
32
+ "max_batch_size": 1,
33
+ "max_context_length": 1024,
34
+ "max_topk": 256,
35
+ "n_active_tokens": 1024,
36
+ "neuronxcc_version": "2.21.33363.0+82129205",
37
+ "on_device_sampling": true,
38
+ "optimum_neuron_version": "0.4.4.dev1",
39
+ "output_logits": false,
40
+ "pp_degree": 1,
41
+ "sequence_length": 1024,
42
+ "sequence_parallel_enabled": false,
43
+ "speculation_length": 0,
44
+ "start_rank_id": 0,
45
+ "target": "trn1",
46
+ "torch_dtype": "bfloat16",
47
+ "tp_degree": 2
48
+ },
49
+ "num_attention_heads": 4,
50
+ "num_hidden_layers": 2,
51
+ "num_key_value_heads": 2,
52
+ "rms_norm_eps": 1e-06,
53
+ "rope_scaling": {
54
+ "factor": 4.0,
55
+ "original_max_position_embeddings": 32768,
56
+ "rope_type": "yarn",
57
+ "type": "yarn"
58
+ },
59
+ "rope_theta": 1000000.0,
60
+ "sliding_window": null,
61
+ "tie_word_embeddings": false,
62
+ "use_cache": true,
63
+ "use_sliding_window": false,
64
+ "vocab_size": 152064
65
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev1/qwen3/Qwen/Qwen3-Embedding-0.6B/18af48e9de9305ddddd5.json ADDED
@@ -0,0 +1,88 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen3-Embedding-0.6B",
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": 1024,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 3072,
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
+ ],
46
+ "max_position_embeddings": 32768,
47
+ "max_window_layers": 28,
48
+ "model_type": "qwen3",
49
+ "neuron": {
50
+ "_serialized_key": "NxDNeuronConfig",
51
+ "batch_size": 4,
52
+ "capacity_factor": null,
53
+ "checkpoint_id": "Qwen/Qwen3-Embedding-0.6B",
54
+ "checkpoint_revision": "c54f2e6e80b2d7b7de06f51cec4959f6b3e03418",
55
+ "continuous_batching": false,
56
+ "ep_degree": 1,
57
+ "fused_qkv": true,
58
+ "glu_mlp": true,
59
+ "local_ranks_size": 2,
60
+ "max_batch_size": 4,
61
+ "max_context_length": 1024,
62
+ "max_topk": 256,
63
+ "n_active_tokens": 1024,
64
+ "neuronxcc_version": "2.21.33363.0+82129205",
65
+ "on_device_sampling": false,
66
+ "optimum_neuron_version": "0.4.4.dev1",
67
+ "output_logits": false,
68
+ "pp_degree": 1,
69
+ "sequence_length": 1024,
70
+ "sequence_parallel_enabled": false,
71
+ "speculation_length": 0,
72
+ "start_rank_id": 0,
73
+ "target": "trn1",
74
+ "torch_dtype": "bfloat16",
75
+ "tp_degree": 2
76
+ },
77
+ "num_attention_heads": 16,
78
+ "num_hidden_layers": 28,
79
+ "num_key_value_heads": 8,
80
+ "rms_norm_eps": 1e-06,
81
+ "rope_scaling": null,
82
+ "rope_theta": 1000000,
83
+ "sliding_window": null,
84
+ "tie_word_embeddings": true,
85
+ "use_cache": true,
86
+ "use_sliding_window": false,
87
+ "vocab_size": 151669
88
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev1/qwen3/Qwen/Qwen3-Embedding-0.6B/80594a5958040f8b1ebe.json ADDED
@@ -0,0 +1,88 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen3-Embedding-0.6B",
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": 1024,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 3072,
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
+ ],
46
+ "max_position_embeddings": 32768,
47
+ "max_window_layers": 28,
48
+ "model_type": "qwen3",
49
+ "neuron": {
50
+ "_serialized_key": "NxDNeuronConfig",
51
+ "batch_size": 6,
52
+ "capacity_factor": null,
53
+ "checkpoint_id": "Qwen/Qwen3-Embedding-0.6B",
54
+ "checkpoint_revision": "c54f2e6e80b2d7b7de06f51cec4959f6b3e03418",
55
+ "continuous_batching": false,
56
+ "ep_degree": 1,
57
+ "fused_qkv": true,
58
+ "glu_mlp": true,
59
+ "local_ranks_size": 2,
60
+ "max_batch_size": 6,
61
+ "max_context_length": 1024,
62
+ "max_topk": 256,
63
+ "n_active_tokens": 1024,
64
+ "neuronxcc_version": "2.21.33363.0+82129205",
65
+ "on_device_sampling": false,
66
+ "optimum_neuron_version": "0.4.4.dev1",
67
+ "output_logits": false,
68
+ "pp_degree": 1,
69
+ "sequence_length": 1024,
70
+ "sequence_parallel_enabled": false,
71
+ "speculation_length": 0,
72
+ "start_rank_id": 0,
73
+ "target": "trn1",
74
+ "torch_dtype": "bfloat16",
75
+ "tp_degree": 2
76
+ },
77
+ "num_attention_heads": 16,
78
+ "num_hidden_layers": 28,
79
+ "num_key_value_heads": 8,
80
+ "rms_norm_eps": 1e-06,
81
+ "rope_scaling": null,
82
+ "rope_theta": 1000000,
83
+ "sliding_window": null,
84
+ "tie_word_embeddings": true,
85
+ "use_cache": true,
86
+ "use_sliding_window": false,
87
+ "vocab_size": 151669
88
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4.dev1/qwen3_moe/optimum-internal-testing/tiny-random-qwen3_moe/a5cf4ee087cd9567449f.json ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "optimum-internal-testing/tiny-random-qwen3_moe",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Qwen3MoeForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "decoder_sparse_step": 2,
11
+ "dtype": "float32",
12
+ "head_dim": 32,
13
+ "hidden_act": "silu",
14
+ "hidden_size": 64,
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 128,
17
+ "max_position_embeddings": 40960,
18
+ "max_window_layers": 1,
19
+ "mlp_only_layers": [],
20
+ "model_type": "qwen3_moe",
21
+ "moe_intermediate_size": 128,
22
+ "neuron": {
23
+ "_serialized_key": "NxDNeuronConfig",
24
+ "batch_size": 1,
25
+ "capacity_factor": null,
26
+ "checkpoint_id": "optimum-internal-testing/tiny-random-qwen3_moe",
27
+ "checkpoint_revision": "e0230be2839556b44b7400a233c73c74b4abb7af",
28
+ "continuous_batching": false,
29
+ "ep_degree": 1,
30
+ "fused_qkv": false,
31
+ "glu_mlp": true,
32
+ "local_ranks_size": 2,
33
+ "max_batch_size": 1,
34
+ "max_context_length": 1024,
35
+ "max_topk": 256,
36
+ "n_active_tokens": 1024,
37
+ "neuronxcc_version": "2.21.33363.0+82129205",
38
+ "on_device_sampling": true,
39
+ "optimum_neuron_version": "0.4.4.dev1",
40
+ "output_logits": false,
41
+ "pp_degree": 1,
42
+ "sequence_length": 1024,
43
+ "sequence_parallel_enabled": false,
44
+ "speculation_length": 0,
45
+ "start_rank_id": 0,
46
+ "target": "trn1",
47
+ "torch_dtype": "float32",
48
+ "tp_degree": 2
49
+ },
50
+ "norm_topk_prob": true,
51
+ "num_attention_heads": 2,
52
+ "num_experts": 8,
53
+ "num_experts_per_tok": 2,
54
+ "num_hidden_layers": 2,
55
+ "num_key_value_heads": 1,
56
+ "output_router_logits": false,
57
+ "rms_norm_eps": 1e-06,
58
+ "rope_scaling": null,
59
+ "rope_theta": 1000000.0,
60
+ "router_aux_loss_coef": 0.001,
61
+ "sliding_window": null,
62
+ "tie_word_embeddings": true,
63
+ "use_cache": true,
64
+ "use_sliding_window": false,
65
+ "vocab_size": 151936
66
+ }
neuronxcc-2.21.33363.0+82129205/MODULE_01328268a19023ed0f24+f8e6d902/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5a7be97d2012aa41b5f56bf5bf86357a0dae77fd27821fa3e0737aba2fe5521e
3
  size 97794
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8ef50e479badb0312e208336533316393fc797857eb5a2cde744753c266d7f3f
3
  size 97794
neuronxcc-2.21.33363.0+82129205/MODULE_01328268a19023ed0f24+f8e6d902/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7ff6c1f2a8f76328f3ffd5cd9cbb7ad62552ac54af96be0b68b754f3e79f2c6e
3
  size 410624
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:671f2968ee09229cb96bb22be152df8bd197829d6ce7ad77ef854a3963b716b9
3
  size 410624
neuronxcc-2.21.33363.0+82129205/MODULE_0efb0afa4ef56e94d498+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_0efb0afa4ef56e94d498+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_0efb0afa4ef56e94d498+a02c3a36/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4aeec7d218e22beefd303645f302d7a805df96146ecddac686eb936f8d682209
3
+ size 84114
neuronxcc-2.21.33363.0+82129205/MODULE_0efb0afa4ef56e94d498+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7f7732c3ad2ac92d6eca8c6efd83baeefa27be9061021e79a069e2496d6faa00
3
+ size 246784
neuronxcc-2.21.33363.0+82129205/MODULE_0efb0afa4ef56e94d498+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b9411f69d1d13a85dfcf64804e552e540bfc24470b7f2b52f419c5eceb8f6357
3
+ size 254967
neuronxcc-2.21.33363.0+82129205/MODULE_14cda09d048ccfb252f8+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_14cda09d048ccfb252f8+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_14cda09d048ccfb252f8+a02c3a36/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:948a51e7715bd8232b0ce82f8cca432d73c2a709e6a794d3a228525868c057f7
3
+ size 93425
neuronxcc-2.21.33363.0+82129205/MODULE_14cda09d048ccfb252f8+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f2c30438f983884de1b30e5866ed44b55b755f839caf513e2312cbdfefc8eda3
3
+ size 277504
neuronxcc-2.21.33363.0+82129205/MODULE_14cda09d048ccfb252f8+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dbbb4dba625a46bae0587ff94ea66196c0ebfd2bfebd5b2d8680fa00fcf7292a
3
+ size 288898
neuronxcc-2.21.33363.0+82129205/MODULE_245d388b7b8c64a4d846+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_245d388b7b8c64a4d846+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_245d388b7b8c64a4d846+a02c3a36/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dc5d22607282b38e6ccc318f3d4d6f232ae382dbbf78a4b7ff65a37c4cdfbad8
3
+ size 82653
neuronxcc-2.21.33363.0+82129205/MODULE_245d388b7b8c64a4d846+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7bb1c68e78267670445d5a05741b98754ca8194705a9ab1bdfb3ce7d44956568
3
+ size 216064
neuronxcc-2.21.33363.0+82129205/MODULE_245d388b7b8c64a4d846+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:780a4a6a0d57688dd4bc494548c72ecaebbfb862d6d3e880442cc279fd2d9437
3
+ size 224275
neuronxcc-2.21.33363.0+82129205/MODULE_2672b897bf6327bea9e4+24129607/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b4b6486a12378501e2484cf81b9f6885eabca76a7d817e3e2041c1736bac16d7
3
  size 82772
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6b56b0d331077542e72e5da19741dfcd8d5d6bdd01d35c1cdbe74227327956a6
3
  size 82772
neuronxcc-2.21.33363.0+82129205/MODULE_2672b897bf6327bea9e4+24129607/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1c65b6b3a5f7c34ae231694981cf76347fb89393c25990695cd4f921f3d243e3
3
  size 267264
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d1cd6ca36926fad065cd480b6e705f247170aeb533426465d808212c1b7a5123
3
  size 267264
neuronxcc-2.21.33363.0+82129205/MODULE_3a758a7122ca68a1243a+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_3a758a7122ca68a1243a+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_3a758a7122ca68a1243a+a02c3a36/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d6d63d7e965d9319351949e66cbad9a3c799e1687254fc9e9b2a96c87adb32fe
3
+ size 865291
neuronxcc-2.21.33363.0+82129205/MODULE_3a758a7122ca68a1243a+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:94b4f8ca1cd6dceeea9e1f3cab2cd6d38961323966034736ce0190d3011373fd
3
+ size 4967424
neuronxcc-2.21.33363.0+82129205/MODULE_3a758a7122ca68a1243a+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c379355b24f769c6ef1b9dae63ff12937987e7a12f79d81471854f83c88f565c
3
+ size 5133948
neuronxcc-2.21.33363.0+82129205/MODULE_5960bdedd7549bfacf7a+24129607/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:76c10df9f9fadac3ce5fee74c4470b2cdade440cf97718b4545a4e3de7fc54aa
3
  size 694128
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5231390fb7b7b3c9416cee6d063cf72f57c3e66de2333a27302fd7480aad917a
3
  size 694128
neuronxcc-2.21.33363.0+82129205/MODULE_5960bdedd7549bfacf7a+24129607/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5f816d25661f62f7f838c0800cb426bb2a2a552c974bc9c93181f83780ef36d8
3
  size 625664
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1d609c20314399dc90d7d3efc2f573991a8799ceb9c8c13a0528387d15454663
3
  size 625664
neuronxcc-2.21.33363.0+82129205/MODULE_5ee5dade1050cf89f316+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_5ee5dade1050cf89f316+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_5ee5dade1050cf89f316+a02c3a36/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:948a51e7715bd8232b0ce82f8cca432d73c2a709e6a794d3a228525868c057f7
3
+ size 93425
neuronxcc-2.21.33363.0+82129205/MODULE_5ee5dade1050cf89f316+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aae99c7c352a3bf0003fc6a1d3cf17bbbd7360829224f73be9a92a36fca32245
3
+ size 277504
neuronxcc-2.21.33363.0+82129205/MODULE_5ee5dade1050cf89f316+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:02fc3c1e2ef02e67ce1ef2daf774a482f12c32bcea4cf8e8c91ce3505d1ee588
3
+ size 288898
neuronxcc-2.21.33363.0+82129205/MODULE_864a283be77e0814947c+bafdbdde/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", "-O1", "--lnc=1", "--enable-saturate-infinity", "--enable-mixed-precision-accumulation", "--internal-enable-dge-levels=vector_dynamic_offsets", "--internal-hlo2tensorizer-options=--verify-hlo=true", "--logfile=/tmp/nxd_model/token_generation/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.21.33363.0+82129205/MODULE_864a283be77e0814947c+bafdbdde/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_864a283be77e0814947c+bafdbdde/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d1c364e560eed61f02284b32afbec9ee92735a7fc75f5da0fdebf4e6f62981de
3
+ size 91833
neuronxcc-2.21.33363.0+82129205/MODULE_864a283be77e0814947c+bafdbdde/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ac0a5f8c623e69c32df0f65a514ae1f29f63c1e4629f5fd16842c7f8630970cf
3
+ size 369664
neuronxcc-2.21.33363.0+82129205/MODULE_864a283be77e0814947c+bafdbdde/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ad55b6aa538df835c51877539b04aa7ec79969d2cfbbc39edc3ecc7feb5ee267
3
+ size 379225
neuronxcc-2.21.33363.0+82129205/MODULE_8b2505504a54575836b0+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_8b2505504a54575836b0+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_8b2505504a54575836b0+a02c3a36/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:04fa3d25bccfae87c411912a1318f7b09223d1a267d2024c09aef48a926aec16
3
+ size 85024
neuronxcc-2.21.33363.0+82129205/MODULE_8b2505504a54575836b0+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1060286be5259ba3f3a1e6e70edf309b3a5f9c2b9edfa4b7f049ca9554af484f
3
+ size 277504