Hugging Face's logo

optimum-internal-testing
/
neuron-testing-cache

Model card Files
xet
 Community
tengomucho HF Staff commited on
Commit
534c36f
·
verified ·
1 Parent(s): 5e55a3e

Synchronizing local compiler cache.

Browse files
Files changed (45) hide show
  1. .gitattributes +12 -0
  2. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4/llama/llamafactory/tiny-random-Llama-3/c18a26167d3d544bfcff.json +62 -0
  3. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4/llama4_text/tiny-random/llama-4/fc0515836b8f5d1f3ae2.json +81 -0
  4. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev1/qwen3/Qwen/Qwen3-0.6B/97503f4e2dca15bd6721.json +87 -0
  5. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev2/04e22cfac37c408caa51d61b493ee185fc351c2f9db86171e024b5d0ad5b25c1/484ee555a60558b25118.json +61 -0
  6. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev2/bd8b5aaeb4693c199f97f6eebd1bbca77255bea6e0ed042c0c592ee5adccec97/4409b94110a511ee8fae.json +80 -0
  7. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev2/bd8b5aaeb4693c199f97f6eebd1bbca77255bea6e0ed042c0c592ee5adccec97/ca8ebd0dd1c56ed002d7.json +80 -0
  8. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev2/gemma3_text/google/gemma-3-270m-it/4409b94110a511ee8fae.json +80 -0
  9. neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev2/qwen3/llamafactory/tiny-random-qwen3/484ee555a60558b25118.json +61 -0
  10. neuronxcc-2.21.33363.0+82129205/MODULE_1af22ef97be0fe871938+24129607/compile_flags.json +1 -0
  11. neuronxcc-2.21.33363.0+82129205/MODULE_1af22ef97be0fe871938+24129607/model.done +0 -0
  12. neuronxcc-2.21.33363.0+82129205/MODULE_1af22ef97be0fe871938+24129607/model.hlo_module.pb +3 -0
  13. neuronxcc-2.21.33363.0+82129205/MODULE_1af22ef97be0fe871938+24129607/model.neff +3 -0
  14. neuronxcc-2.21.33363.0+82129205/MODULE_1b5fd23da565a2f62ae1+a02c3a36/compile_flags.json +1 -0
  15. neuronxcc-2.21.33363.0+82129205/MODULE_1b5fd23da565a2f62ae1+a02c3a36/model.done +0 -0
  16. neuronxcc-2.21.33363.0+82129205/MODULE_1b5fd23da565a2f62ae1+a02c3a36/model.hlo_module.pb +3 -0
  17. neuronxcc-2.21.33363.0+82129205/MODULE_1b5fd23da565a2f62ae1+a02c3a36/model.neff +3 -0
  18. neuronxcc-2.21.33363.0+82129205/MODULE_1b5fd23da565a2f62ae1+a02c3a36/wrapped_neff.hlo +3 -0
  19. neuronxcc-2.21.33363.0+82129205/MODULE_2c2347b5dccbe5eb00d1+24129607/compile_flags.json +1 -0
  20. neuronxcc-2.21.33363.0+82129205/MODULE_2c2347b5dccbe5eb00d1+24129607/model.done +0 -0
  21. neuronxcc-2.21.33363.0+82129205/MODULE_2c2347b5dccbe5eb00d1+24129607/model.hlo_module.pb +3 -0
  22. neuronxcc-2.21.33363.0+82129205/MODULE_2c2347b5dccbe5eb00d1+24129607/model.neff +3 -0
  23. neuronxcc-2.21.33363.0+82129205/MODULE_766b26a05d7144c2d188+24129607/compile_flags.json +1 -0
  24. neuronxcc-2.21.33363.0+82129205/MODULE_766b26a05d7144c2d188+24129607/model.done +0 -0
  25. neuronxcc-2.21.33363.0+82129205/MODULE_766b26a05d7144c2d188+24129607/model.hlo_module.pb +3 -0
  26. neuronxcc-2.21.33363.0+82129205/MODULE_766b26a05d7144c2d188+24129607/model.neff +3 -0
  27. neuronxcc-2.21.33363.0+82129205/MODULE_a063aa5293cb29ba4780+24129607/compile_flags.json +1 -0
  28. neuronxcc-2.21.33363.0+82129205/MODULE_a063aa5293cb29ba4780+24129607/model.done +0 -0
  29. neuronxcc-2.21.33363.0+82129205/MODULE_a063aa5293cb29ba4780+24129607/model.hlo_module.pb +3 -0
  30. neuronxcc-2.21.33363.0+82129205/MODULE_a063aa5293cb29ba4780+24129607/model.neff +3 -0
  31. neuronxcc-2.21.33363.0+82129205/MODULE_e1bdba015878a24fad9c+a02c3a36/compile_flags.json +1 -0
  32. neuronxcc-2.21.33363.0+82129205/MODULE_e1bdba015878a24fad9c+a02c3a36/model.done +0 -0
  33. neuronxcc-2.21.33363.0+82129205/MODULE_e1bdba015878a24fad9c+a02c3a36/model.hlo_module.pb +3 -0
  34. neuronxcc-2.21.33363.0+82129205/MODULE_e1bdba015878a24fad9c+a02c3a36/model.neff +3 -0
  35. neuronxcc-2.21.33363.0+82129205/MODULE_e1bdba015878a24fad9c+a02c3a36/wrapped_neff.hlo +3 -0
  36. neuronxcc-2.21.33363.0+82129205/MODULE_f452005788ad9d57f1bf+a02c3a36/compile_flags.json +1 -0
  37. neuronxcc-2.21.33363.0+82129205/MODULE_f452005788ad9d57f1bf+a02c3a36/model.done +0 -0
  38. neuronxcc-2.21.33363.0+82129205/MODULE_f452005788ad9d57f1bf+a02c3a36/model.hlo_module.pb +3 -0
  39. neuronxcc-2.21.33363.0+82129205/MODULE_f452005788ad9d57f1bf+a02c3a36/model.neff +3 -0
  40. neuronxcc-2.21.33363.0+82129205/MODULE_f452005788ad9d57f1bf+a02c3a36/wrapped_neff.hlo +3 -0
  41. neuronxcc-2.21.33363.0+82129205/MODULE_fb7c14df8ae0afce903f+a02c3a36/compile_flags.json +1 -0
  42. neuronxcc-2.21.33363.0+82129205/MODULE_fb7c14df8ae0afce903f+a02c3a36/model.done +0 -0
  43. neuronxcc-2.21.33363.0+82129205/MODULE_fb7c14df8ae0afce903f+a02c3a36/model.hlo_module.pb +3 -0
  44. neuronxcc-2.21.33363.0+82129205/MODULE_fb7c14df8ae0afce903f+a02c3a36/model.neff +3 -0
  45. neuronxcc-2.21.33363.0+82129205/MODULE_fb7c14df8ae0afce903f+a02c3a36/wrapped_neff.hlo +3 -0
.gitattributes CHANGED
@@ -6440,3 +6440,15 @@ neuronxcc-2.21.33363.0+82129205/MODULE_34962ecf0146edb895ab+10d12301/model.neff
6440
  neuronxcc-2.21.33363.0+82129205/MODULE_34962ecf0146edb895ab+10d12301/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
6441
  neuronxcc-2.21.33363.0+82129205/MODULE_6268d8300fb4869a12d0+c832af78/model.neff filter=lfs diff=lfs merge=lfs -text
6442
  neuronxcc-2.21.33363.0+82129205/MODULE_6268d8300fb4869a12d0+c832af78/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
 
 
 
 
 
 
6440
  neuronxcc-2.21.33363.0+82129205/MODULE_34962ecf0146edb895ab+10d12301/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
6441
  neuronxcc-2.21.33363.0+82129205/MODULE_6268d8300fb4869a12d0+c832af78/model.neff filter=lfs diff=lfs merge=lfs -text
6442
  neuronxcc-2.21.33363.0+82129205/MODULE_6268d8300fb4869a12d0+c832af78/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
6443
+ neuronxcc-2.21.33363.0+82129205/MODULE_1af22ef97be0fe871938+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
6444
+ neuronxcc-2.21.33363.0+82129205/MODULE_1b5fd23da565a2f62ae1+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
6445
+ neuronxcc-2.21.33363.0+82129205/MODULE_1b5fd23da565a2f62ae1+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
6446
+ neuronxcc-2.21.33363.0+82129205/MODULE_2c2347b5dccbe5eb00d1+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
6447
+ neuronxcc-2.21.33363.0+82129205/MODULE_766b26a05d7144c2d188+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
6448
+ neuronxcc-2.21.33363.0+82129205/MODULE_a063aa5293cb29ba4780+24129607/model.neff filter=lfs diff=lfs merge=lfs -text
6449
+ neuronxcc-2.21.33363.0+82129205/MODULE_e1bdba015878a24fad9c+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
6450
+ neuronxcc-2.21.33363.0+82129205/MODULE_e1bdba015878a24fad9c+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
6451
+ neuronxcc-2.21.33363.0+82129205/MODULE_f452005788ad9d57f1bf+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
6452
+ neuronxcc-2.21.33363.0+82129205/MODULE_f452005788ad9d57f1bf+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
6453
+ neuronxcc-2.21.33363.0+82129205/MODULE_fb7c14df8ae0afce903f+a02c3a36/model.neff filter=lfs diff=lfs merge=lfs -text
6454
+ neuronxcc-2.21.33363.0+82129205/MODULE_fb7c14df8ae0afce903f+a02c3a36/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4/llama/llamafactory/tiny-random-Llama-3/c18a26167d3d544bfcff.json ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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",
37
+ "output_logits": false,
38
+ "pp_degree": 1,
39
+ "sequence_length": 1024,
40
+ "speculation_length": 0,
41
+ "start_rank_id": 0,
42
+ "target": "trn1",
43
+ "torch_dtype": "float16",
44
+ "tp_degree": 2
45
+ },
46
+ "num_attention_heads": 4,
47
+ "num_hidden_layers": 2,
48
+ "num_key_value_heads": 4,
49
+ "pretraining_tp": 1,
50
+ "rms_norm_eps": 1e-05,
51
+ "rope_scaling": {
52
+ "factor": 8.0,
53
+ "high_freq_factor": 4.0,
54
+ "low_freq_factor": 1.0,
55
+ "original_max_position_embeddings": 8192,
56
+ "rope_type": "llama3"
57
+ },
58
+ "rope_theta": 500000.0,
59
+ "tie_word_embeddings": false,
60
+ "use_cache": true,
61
+ "vocab_size": 128256
62
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.4/llama4_text/tiny-random/llama-4/fc0515836b8f5d1f3ae2.json ADDED
@@ -0,0 +1,81 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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",
51
+ "output_logits": false,
52
+ "pp_degree": 1,
53
+ "sequence_length": 1024,
54
+ "speculation_length": 0,
55
+ "start_rank_id": 0,
56
+ "target": "trn1",
57
+ "torch_dtype": "bfloat16",
58
+ "tp_degree": 2
59
+ },
60
+ "no_rope_layers": [
61
+ 1,
62
+ 1,
63
+ 1,
64
+ 0
65
+ ],
66
+ "num_attention_heads": 1,
67
+ "num_experts_per_tok": 1,
68
+ "num_hidden_layers": 4,
69
+ "num_key_value_heads": 1,
70
+ "num_local_experts": 8,
71
+ "output_router_logits": false,
72
+ "rms_norm_eps": 1e-05,
73
+ "rope_scaling": null,
74
+ "rope_theta": 500000.0,
75
+ "router_aux_loss_coef": 0.001,
76
+ "router_jitter_noise": 0.0,
77
+ "tie_word_embeddings": true,
78
+ "use_cache": true,
79
+ "use_qk_norm": true,
80
+ "vocab_size": 202048
81
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev1/qwen3/Qwen/Qwen3-0.6B/97503f4e2dca15bd6721.json ADDED
@@ -0,0 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen3-0.6B",
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": 40960,
47
+ "max_window_layers": 28,
48
+ "model_type": "qwen3",
49
+ "neuron": {
50
+ "_serialized_key": "NxDNeuronConfig",
51
+ "batch_size": 1,
52
+ "capacity_factor": null,
53
+ "checkpoint_id": "Qwen/Qwen3-0.6B",
54
+ "checkpoint_revision": "c1899de289a04d12100db370d81485cdf75e47ca",
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": 1,
61
+ "max_context_length": 8192,
62
+ "max_topk": 256,
63
+ "n_active_tokens": 8192,
64
+ "neuronxcc_version": "2.21.33363.0+82129205",
65
+ "on_device_sampling": true,
66
+ "optimum_neuron_version": "0.4.5.dev1",
67
+ "output_logits": false,
68
+ "pp_degree": 1,
69
+ "sequence_length": 8192,
70
+ "speculation_length": 0,
71
+ "start_rank_id": 0,
72
+ "target": "trn1",
73
+ "torch_dtype": "bfloat16",
74
+ "tp_degree": 2
75
+ },
76
+ "num_attention_heads": 16,
77
+ "num_hidden_layers": 28,
78
+ "num_key_value_heads": 8,
79
+ "rms_norm_eps": 1e-06,
80
+ "rope_scaling": null,
81
+ "rope_theta": 1000000,
82
+ "sliding_window": null,
83
+ "tie_word_embeddings": true,
84
+ "use_cache": true,
85
+ "use_sliding_window": false,
86
+ "vocab_size": 151936
87
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev2/04e22cfac37c408caa51d61b493ee185fc351c2f9db86171e024b5d0ad5b25c1/484ee555a60558b25118.json ADDED
@@ -0,0 +1,61 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "llamafactory/tiny-random-qwen3",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Qwen3ForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "dtype": "bfloat16",
11
+ "head_dim": 4,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 16,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 64,
16
+ "layer_types": [
17
+ "full_attention",
18
+ "full_attention"
19
+ ],
20
+ "max_position_embeddings": 32768,
21
+ "max_window_layers": 36,
22
+ "model_type": "qwen3",
23
+ "neuron": {
24
+ "_serialized_key": "NxDNeuronConfig",
25
+ "batch_size": 2,
26
+ "capacity_factor": null,
27
+ "checkpoint_id": "llamafactory/tiny-random-qwen3",
28
+ "checkpoint_revision": "81d6f5f5e05ed53ea8a1d19431266a486e46bbd8",
29
+ "continuous_batching": true,
30
+ "ep_degree": 1,
31
+ "fused_qkv": true,
32
+ "glu_mlp": true,
33
+ "local_ranks_size": 2,
34
+ "max_batch_size": 2,
35
+ "max_context_length": 1024,
36
+ "max_topk": 256,
37
+ "n_active_tokens": 1024,
38
+ "neuronxcc_version": "2.21.33363.0+82129205",
39
+ "on_device_sampling": false,
40
+ "optimum_neuron_version": "0.4.5.dev2",
41
+ "output_logits": false,
42
+ "pp_degree": 1,
43
+ "sequence_length": 1024,
44
+ "speculation_length": 0,
45
+ "start_rank_id": 0,
46
+ "target": "trn1",
47
+ "torch_dtype": "bfloat16",
48
+ "tp_degree": 2
49
+ },
50
+ "num_attention_heads": 4,
51
+ "num_hidden_layers": 2,
52
+ "num_key_value_heads": 4,
53
+ "rms_norm_eps": 1e-06,
54
+ "rope_scaling": null,
55
+ "rope_theta": 5000000,
56
+ "sliding_window": null,
57
+ "tie_word_embeddings": true,
58
+ "use_cache": true,
59
+ "use_sliding_window": false,
60
+ "vocab_size": 151936
61
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev2/bd8b5aaeb4693c199f97f6eebd1bbca77255bea6e0ed042c0c592ee5adccec97/4409b94110a511ee8fae.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": 1024,
54
+ "max_topk": 256,
55
+ "n_active_tokens": 1024,
56
+ "neuronxcc_version": "2.21.33363.0+82129205",
57
+ "on_device_sampling": true,
58
+ "optimum_neuron_version": "0.4.5.dev2",
59
+ "output_logits": false,
60
+ "pp_degree": 1,
61
+ "sequence_length": 1024,
62
+ "speculation_length": 0,
63
+ "start_rank_id": 0,
64
+ "target": "trn1",
65
+ "torch_dtype": "bfloat16",
66
+ "tp_degree": 2
67
+ },
68
+ "num_attention_heads": 4,
69
+ "num_hidden_layers": 18,
70
+ "num_key_value_heads": 1,
71
+ "query_pre_attn_scalar": 256,
72
+ "rms_norm_eps": 1e-06,
73
+ "rope_local_base_freq": 10000.0,
74
+ "rope_scaling": null,
75
+ "rope_theta": 1000000.0,
76
+ "sliding_window": 512,
77
+ "use_bidirectional_attention": false,
78
+ "use_cache": true,
79
+ "vocab_size": 262144
80
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev2/bd8b5aaeb4693c199f97f6eebd1bbca77255bea6e0ed042c0c592ee5adccec97/ca8ebd0dd1c56ed002d7.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": 4096,
54
+ "max_topk": 256,
55
+ "n_active_tokens": 4096,
56
+ "neuronxcc_version": "2.21.33363.0+82129205",
57
+ "on_device_sampling": true,
58
+ "optimum_neuron_version": "0.4.5.dev2",
59
+ "output_logits": false,
60
+ "pp_degree": 1,
61
+ "sequence_length": 4096,
62
+ "speculation_length": 0,
63
+ "start_rank_id": 0,
64
+ "target": "trn1",
65
+ "torch_dtype": "bfloat16",
66
+ "tp_degree": 2
67
+ },
68
+ "num_attention_heads": 4,
69
+ "num_hidden_layers": 18,
70
+ "num_key_value_heads": 1,
71
+ "query_pre_attn_scalar": 256,
72
+ "rms_norm_eps": 1e-06,
73
+ "rope_local_base_freq": 10000.0,
74
+ "rope_scaling": null,
75
+ "rope_theta": 1000000.0,
76
+ "sliding_window": 512,
77
+ "use_bidirectional_attention": false,
78
+ "use_cache": true,
79
+ "vocab_size": 262144
80
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev2/gemma3_text/google/gemma-3-270m-it/4409b94110a511ee8fae.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": 1024,
54
+ "max_topk": 256,
55
+ "n_active_tokens": 1024,
56
+ "neuronxcc_version": "2.21.33363.0+82129205",
57
+ "on_device_sampling": true,
58
+ "optimum_neuron_version": "0.4.5.dev2",
59
+ "output_logits": false,
60
+ "pp_degree": 1,
61
+ "sequence_length": 1024,
62
+ "speculation_length": 0,
63
+ "start_rank_id": 0,
64
+ "target": "trn1",
65
+ "torch_dtype": "bfloat16",
66
+ "tp_degree": 2
67
+ },
68
+ "num_attention_heads": 4,
69
+ "num_hidden_layers": 18,
70
+ "num_key_value_heads": 1,
71
+ "query_pre_attn_scalar": 256,
72
+ "rms_norm_eps": 1e-06,
73
+ "rope_local_base_freq": 10000.0,
74
+ "rope_scaling": null,
75
+ "rope_theta": 1000000.0,
76
+ "sliding_window": 512,
77
+ "use_bidirectional_attention": false,
78
+ "use_cache": true,
79
+ "vocab_size": 262144
80
+ }
neuronxcc-2.21.33363.0+82129205/0_REGISTRY/0.4.5.dev2/qwen3/llamafactory/tiny-random-qwen3/484ee555a60558b25118.json ADDED
@@ -0,0 +1,61 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "llamafactory/tiny-random-qwen3",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Qwen3ForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "dtype": "bfloat16",
11
+ "head_dim": 4,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 16,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 64,
16
+ "layer_types": [
17
+ "full_attention",
18
+ "full_attention"
19
+ ],
20
+ "max_position_embeddings": 32768,
21
+ "max_window_layers": 36,
22
+ "model_type": "qwen3",
23
+ "neuron": {
24
+ "_serialized_key": "NxDNeuronConfig",
25
+ "batch_size": 2,
26
+ "capacity_factor": null,
27
+ "checkpoint_id": "llamafactory/tiny-random-qwen3",
28
+ "checkpoint_revision": "81d6f5f5e05ed53ea8a1d19431266a486e46bbd8",
29
+ "continuous_batching": true,
30
+ "ep_degree": 1,
31
+ "fused_qkv": true,
32
+ "glu_mlp": true,
33
+ "local_ranks_size": 2,
34
+ "max_batch_size": 2,
35
+ "max_context_length": 1024,
36
+ "max_topk": 256,
37
+ "n_active_tokens": 1024,
38
+ "neuronxcc_version": "2.21.33363.0+82129205",
39
+ "on_device_sampling": false,
40
+ "optimum_neuron_version": "0.4.5.dev2",
41
+ "output_logits": false,
42
+ "pp_degree": 1,
43
+ "sequence_length": 1024,
44
+ "speculation_length": 0,
45
+ "start_rank_id": 0,
46
+ "target": "trn1",
47
+ "torch_dtype": "bfloat16",
48
+ "tp_degree": 2
49
+ },
50
+ "num_attention_heads": 4,
51
+ "num_hidden_layers": 2,
52
+ "num_key_value_heads": 4,
53
+ "rms_norm_eps": 1e-06,
54
+ "rope_scaling": null,
55
+ "rope_theta": 5000000,
56
+ "sliding_window": null,
57
+ "tie_word_embeddings": true,
58
+ "use_cache": true,
59
+ "use_sliding_window": false,
60
+ "vocab_size": 151936
61
+ }
neuronxcc-2.21.33363.0+82129205/MODULE_1af22ef97be0fe871938+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_1af22ef97be0fe871938+24129607/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_1af22ef97be0fe871938+24129607/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dde78f832bf7b42825123a2072be02643a57f95cbd882555ecd842f6d6419934
3
+ size 708855
neuronxcc-2.21.33363.0+82129205/MODULE_1af22ef97be0fe871938+24129607/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6af5db1b519fd304a769ec6c3571b8730b8f69c2c047ab2289f4d15a66c87fac
3
+ size 7660544
neuronxcc-2.21.33363.0+82129205/MODULE_1b5fd23da565a2f62ae1+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_1b5fd23da565a2f62ae1+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_1b5fd23da565a2f62ae1+a02c3a36/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:26a0120595235859cb5d3b5081228df5cee5f5b7509e5089c2e0b6a2ffdf54cb
3
+ size 692215
neuronxcc-2.21.33363.0+82129205/MODULE_1b5fd23da565a2f62ae1+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e4df5bdd88883f7b8592e7ad15fc56ac4677efbdae065aa8106d8df7484bbc69
3
+ size 1045504
neuronxcc-2.21.33363.0+82129205/MODULE_1b5fd23da565a2f62ae1+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3948819097df7a66219c39985f00a04157938ea56636a068d74f644e3d7b6e06
3
+ size 1167896
neuronxcc-2.21.33363.0+82129205/MODULE_2c2347b5dccbe5eb00d1+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_2c2347b5dccbe5eb00d1+24129607/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_2c2347b5dccbe5eb00d1+24129607/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fa9f88a7adc91f3c99012a220709f8d0099fa64358b63acd3f16030322f1b134
3
+ size 669533
neuronxcc-2.21.33363.0+82129205/MODULE_2c2347b5dccbe5eb00d1+24129607/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f4b6d32fe855702ba97203b0c9eb3a5446060950a269f4450c8b6eb507786e37
3
+ size 1813504
neuronxcc-2.21.33363.0+82129205/MODULE_766b26a05d7144c2d188+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_766b26a05d7144c2d188+24129607/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_766b26a05d7144c2d188+24129607/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8dbc00ee7858cb46fdbb68cacceb3372d7470cbebf8896e39ea2af5d8efc62b9
3
+ size 52101
neuronxcc-2.21.33363.0+82129205/MODULE_766b26a05d7144c2d188+24129607/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6e5ecd710d55ed09335de5557912db1b30262439da7e0e806a26135ecbf1e150
3
+ size 328704
neuronxcc-2.21.33363.0+82129205/MODULE_a063aa5293cb29ba4780+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_a063aa5293cb29ba4780+24129607/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_a063aa5293cb29ba4780+24129607/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7dc28abc15c2cee5a25c29583cb86b7174722924e2345bdb99037bc17e6015a6
3
+ size 843918
neuronxcc-2.21.33363.0+82129205/MODULE_a063aa5293cb29ba4780+24129607/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:26feeebcacf0f72fffda73f862f24323f9cc45c5cfee897da8b4f415eb71ec7b
3
+ size 96881664
neuronxcc-2.21.33363.0+82129205/MODULE_e1bdba015878a24fad9c+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_e1bdba015878a24fad9c+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_e1bdba015878a24fad9c+a02c3a36/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7b5827ec37f20eb1a03f14b76da5b69939e3916eddeea598bfcc6b4e3825399f
3
+ size 758053
neuronxcc-2.21.33363.0+82129205/MODULE_e1bdba015878a24fad9c+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:88bacbb0526fdbc417797b5812937e7f06200b3fbc689b2eddb86241579a58cb
3
+ size 2417664
neuronxcc-2.21.33363.0+82129205/MODULE_e1bdba015878a24fad9c+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5aee6730afe995350e7a9eb65067244b35a8e7c710edde69449a070c803407e4
3
+ size 2554832
neuronxcc-2.21.33363.0+82129205/MODULE_f452005788ad9d57f1bf+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_f452005788ad9d57f1bf+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_f452005788ad9d57f1bf+a02c3a36/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eb3c421871afc527b32f8ff1a4330c7ac88fb42331d3ff7e0acef5406bee1be6
3
+ size 50068
neuronxcc-2.21.33363.0+82129205/MODULE_f452005788ad9d57f1bf+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:77d0ed55734988658a20a6885fd9b8a942c2b49a291fed067ed3adbcaddadb78
3
+ size 287744
neuronxcc-2.21.33363.0+82129205/MODULE_f452005788ad9d57f1bf+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:baf1d33ff412bb350ae4589e582cdf57ebd28be41d4e216ab33c58f7ad744d09
3
+ size 296665
neuronxcc-2.21.33363.0+82129205/MODULE_fb7c14df8ae0afce903f+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_fb7c14df8ae0afce903f+a02c3a36/model.done ADDED
File without changes
neuronxcc-2.21.33363.0+82129205/MODULE_fb7c14df8ae0afce903f+a02c3a36/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5d9da4b130a1408ff7fb0272996084fdfdb12183853e3b8a2849686d0c07e0d6
3
+ size 699082
neuronxcc-2.21.33363.0+82129205/MODULE_fb7c14df8ae0afce903f+a02c3a36/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6690b004795423fee4f05e9adb62198826a53ed08f1ba9698e5ea433d7f238f1
3
+ size 1793024
neuronxcc-2.21.33363.0+82129205/MODULE_fb7c14df8ae0afce903f+a02c3a36/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:865eca2130ef2992aca8b893a2c92910fad9692cc33fd23fbabef42277aca001
3
+ size 1915416