Hugging Face's logo

optimum-internal-testing
/
neuron-testing-cache

Model card Files
xet
Community
dacorvo HF Staff commited on
Commit
1075cec
·
verified ·
1 Parent(s): 9e787f0

Synchronizing local compiler cache.

Browse files
Files changed (45) hide show
  1. .gitattributes +9 -0
  2. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0/llama/meta-llama/Llama-3.1-8B-Instruct/6bbad1c803a240aefcb9.json +77 -0
  3. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.1/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/09344096143dd7670106.json +73 -0
  4. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.1/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/2b0fa8ae0a115fad4c29.json +73 -0
  5. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.1/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/bfc2e217ecf039318115.json +73 -0
  6. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.1/llama/llamafactory/tiny-random-Llama-3/0c86e63e398c5b4e2cfd.json +77 -0
  7. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.1/llama/llamafactory/tiny-random-Llama-3/c58bc107cced18b694d6.json +77 -0
  8. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.1/llama/llamafactory/tiny-random-Llama-3/ff2809df4f513c66db23.json +77 -0
  9. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.1/llama/unsloth/Llama-3.2-1B-Instruct/c7a102927bd4d31a4086.json +78 -0
  10. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.1/mixtral/dacorvo/Mixtral-tiny/2654eeeba6e6693fe3a7.json +73 -0
  11. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.1/mixtral/dacorvo/Mixtral-tiny/3f63a19d4af1b82d4569.json +73 -0
  12. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.1/mixtral/dacorvo/Mixtral-tiny/786147c74e4207def994.json +73 -0
  13. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.1/phi3/yujiepan/phi-4-tiny-random/0f1bfbaf10d2c07abd9a.json +74 -0
  14. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.1/phi3/yujiepan/phi-4-tiny-random/121619e98da8e946a4ce.json +74 -0
  15. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.1/phi3/yujiepan/phi-4-tiny-random/a6d251ae61cee42f8929.json +74 -0
  16. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.1/qwen2/yujiepan/qwen2.5-128k-tiny-random/2d2a3a890f7406761316.json +75 -0
  17. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.1/qwen2/yujiepan/qwen2.5-128k-tiny-random/a638416d74f91230ba38.json +75 -0
  18. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.1/qwen2/yujiepan/qwen2.5-128k-tiny-random/f8e1b768ef174982facc.json +75 -0
  19. neuronxcc-2.17.194.0+d312836f/MODULE_57b107bd0499cc4986ac+63ea7cf2/compile_flags.json +1 -0
  20. neuronxcc-2.17.194.0+d312836f/MODULE_57b107bd0499cc4986ac+63ea7cf2/model.done +0 -0
  21. neuronxcc-2.17.194.0+d312836f/MODULE_57b107bd0499cc4986ac+63ea7cf2/model.hlo_module.pb +3 -0
  22. neuronxcc-2.17.194.0+d312836f/MODULE_57b107bd0499cc4986ac+63ea7cf2/model.neff +3 -0
  23. neuronxcc-2.17.194.0+d312836f/MODULE_57b107bd0499cc4986ac+63ea7cf2/wrapped_neff.hlo +3 -0
  24. neuronxcc-2.17.194.0+d312836f/MODULE_64950c85776a119cdf83+b001bea4/compile_flags.json +1 -0
  25. neuronxcc-2.17.194.0+d312836f/MODULE_64950c85776a119cdf83+b001bea4/model.done +0 -0
  26. neuronxcc-2.17.194.0+d312836f/MODULE_64950c85776a119cdf83+b001bea4/model.hlo_module.pb +3 -0
  27. neuronxcc-2.17.194.0+d312836f/MODULE_64950c85776a119cdf83+b001bea4/model.neff +3 -0
  28. neuronxcc-2.17.194.0+d312836f/MODULE_6bb5680c622866b3b45b+63ea7cf2/compile_flags.json +1 -0
  29. neuronxcc-2.17.194.0+d312836f/MODULE_6bb5680c622866b3b45b+63ea7cf2/model.done +0 -0
  30. neuronxcc-2.17.194.0+d312836f/MODULE_6bb5680c622866b3b45b+63ea7cf2/model.hlo_module.pb +3 -0
  31. neuronxcc-2.17.194.0+d312836f/MODULE_6bb5680c622866b3b45b+63ea7cf2/model.neff +3 -0
  32. neuronxcc-2.17.194.0+d312836f/MODULE_6bb5680c622866b3b45b+63ea7cf2/wrapped_neff.hlo +3 -0
  33. neuronxcc-2.17.194.0+d312836f/MODULE_92d7a6b8bc621dee02b9+b001bea4/compile_flags.json +1 -0
  34. neuronxcc-2.17.194.0+d312836f/MODULE_92d7a6b8bc621dee02b9+b001bea4/model.done +0 -0
  35. neuronxcc-2.17.194.0+d312836f/MODULE_92d7a6b8bc621dee02b9+b001bea4/model.hlo_module.pb +3 -0
  36. neuronxcc-2.17.194.0+d312836f/MODULE_92d7a6b8bc621dee02b9+b001bea4/model.neff +3 -0
  37. neuronxcc-2.17.194.0+d312836f/MODULE_9a53db93d18e769ee7ea+b001bea4/compile_flags.json +1 -0
  38. neuronxcc-2.17.194.0+d312836f/MODULE_9a53db93d18e769ee7ea+b001bea4/model.done +0 -0
  39. neuronxcc-2.17.194.0+d312836f/MODULE_9a53db93d18e769ee7ea+b001bea4/model.hlo_module.pb +3 -0
  40. neuronxcc-2.17.194.0+d312836f/MODULE_9a53db93d18e769ee7ea+b001bea4/model.neff +3 -0
  41. neuronxcc-2.17.194.0+d312836f/MODULE_ae5ca4b91afd03b04b25+63ea7cf2/compile_flags.json +1 -0
  42. neuronxcc-2.17.194.0+d312836f/MODULE_ae5ca4b91afd03b04b25+63ea7cf2/model.done +0 -0
  43. neuronxcc-2.17.194.0+d312836f/MODULE_ae5ca4b91afd03b04b25+63ea7cf2/model.hlo_module.pb +3 -0
  44. neuronxcc-2.17.194.0+d312836f/MODULE_ae5ca4b91afd03b04b25+63ea7cf2/model.neff +3 -0
  45. neuronxcc-2.17.194.0+d312836f/MODULE_ae5ca4b91afd03b04b25+63ea7cf2/wrapped_neff.hlo +3 -0
.gitattributes CHANGED
@@ -3337,3 +3337,12 @@ neuronxcc-2.19.8089.0+8ab9f450/MODULE_d757f0d340c8bb449f21+5be477de/model.neff f
3337
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_d757f0d340c8bb449f21+5be477de/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
3338
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_e716cefc11279f5393c2+84f3e719/model.neff filter=lfs diff=lfs merge=lfs -text
3339
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_f86c96f7ee62c6431f74+431f5505/model.neff filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
 
 
 
3337
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_d757f0d340c8bb449f21+5be477de/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
3338
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_e716cefc11279f5393c2+84f3e719/model.neff filter=lfs diff=lfs merge=lfs -text
3339
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_f86c96f7ee62c6431f74+431f5505/model.neff filter=lfs diff=lfs merge=lfs -text
3340
+ neuronxcc-2.17.194.0+d312836f/MODULE_57b107bd0499cc4986ac+63ea7cf2/model.neff filter=lfs diff=lfs merge=lfs -text
3341
+ neuronxcc-2.17.194.0+d312836f/MODULE_57b107bd0499cc4986ac+63ea7cf2/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
3342
+ neuronxcc-2.17.194.0+d312836f/MODULE_64950c85776a119cdf83+b001bea4/model.neff filter=lfs diff=lfs merge=lfs -text
3343
+ neuronxcc-2.17.194.0+d312836f/MODULE_6bb5680c622866b3b45b+63ea7cf2/model.neff filter=lfs diff=lfs merge=lfs -text
3344
+ neuronxcc-2.17.194.0+d312836f/MODULE_6bb5680c622866b3b45b+63ea7cf2/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
3345
+ neuronxcc-2.17.194.0+d312836f/MODULE_92d7a6b8bc621dee02b9+b001bea4/model.neff filter=lfs diff=lfs merge=lfs -text
3346
+ neuronxcc-2.17.194.0+d312836f/MODULE_9a53db93d18e769ee7ea+b001bea4/model.neff filter=lfs diff=lfs merge=lfs -text
3347
+ neuronxcc-2.17.194.0+d312836f/MODULE_ae5ca4b91afd03b04b25+63ea7cf2/model.neff filter=lfs diff=lfs merge=lfs -text
3348
+ neuronxcc-2.17.194.0+d312836f/MODULE_ae5ca4b91afd03b04b25+63ea7cf2/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0/llama/meta-llama/Llama-3.1-8B-Instruct/6bbad1c803a240aefcb9.json ADDED
@@ -0,0 +1,77 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "meta-llama/Llama-3.1-8B-Instruct",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "LlamaForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "head_dim": 128,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 4096,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 14336,
15
+ "max_position_embeddings": 131072,
16
+ "mlp_bias": false,
17
+ "model_type": "llama",
18
+ "neuron": {
19
+ "_serialized_key": "NxDNeuronConfig",
20
+ "async_mode": false,
21
+ "attn_kernel_enabled": false,
22
+ "batch_size": 32,
23
+ "capacity_factor": null,
24
+ "cc_pipeline_tiling_factor": 2,
25
+ "checkpoint_id": "meta-llama/Llama-3.1-8B-Instruct",
26
+ "checkpoint_revision": "0e9e39f249a16976918f6564b8830bc894c89659",
27
+ "continuous_batching": true,
28
+ "enable_bucketing": false,
29
+ "ep_degree": 1,
30
+ "flash_decoding_enabled": false,
31
+ "fused_qkv": true,
32
+ "glu_mlp": true,
33
+ "is_chunked_prefill": false,
34
+ "local_ranks_size": 8,
35
+ "logical_nc_config": 1,
36
+ "max_batch_size": 32,
37
+ "max_context_length": 4096,
38
+ "max_topk": 256,
39
+ "mlp_kernel_enabled": false,
40
+ "mlp_kernel_fuse_residual_add": false,
41
+ "n_active_tokens": 4096,
42
+ "neuronxcc_version": "2.17.194.0+d312836f",
43
+ "num_cores_per_group": 1,
44
+ "on_device_sampling": true,
45
+ "optimum_neuron_version": "0.2.0",
46
+ "output_logits": false,
47
+ "padding_side": "right",
48
+ "pp_degree": 1,
49
+ "qk_layernorm": false,
50
+ "qkv_kernel_enabled": false,
51
+ "rpl_reduce_dtype": "bfloat16",
52
+ "sequence_length": 4096,
53
+ "sequence_parallel_enabled": false,
54
+ "speculation_length": 0,
55
+ "start_rank_id": 0,
56
+ "target": null,
57
+ "torch_dtype": "bfloat16",
58
+ "tp_degree": 8,
59
+ "vocab_parallel": false
60
+ },
61
+ "num_attention_heads": 32,
62
+ "num_hidden_layers": 32,
63
+ "num_key_value_heads": 8,
64
+ "pretraining_tp": 1,
65
+ "rms_norm_eps": 1e-05,
66
+ "rope_scaling": {
67
+ "factor": 8.0,
68
+ "high_freq_factor": 4.0,
69
+ "low_freq_factor": 1.0,
70
+ "original_max_position_embeddings": 8192,
71
+ "rope_type": "llama3"
72
+ },
73
+ "rope_theta": 500000.0,
74
+ "tie_word_embeddings": false,
75
+ "use_cache": true,
76
+ "vocab_size": 128256
77
+ }
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.1/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/09344096143dd7670106.json ADDED
@@ -0,0 +1,73 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "embedding_multiplier": 1.0,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 32,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 64,
16
+ "logits_scaling": 1.0,
17
+ "max_position_embeddings": 2048,
18
+ "mlp_bias": false,
19
+ "model_type": "granite",
20
+ "neuron": {
21
+ "_serialized_key": "NxDNeuronConfig",
22
+ "async_mode": false,
23
+ "attn_kernel_enabled": false,
24
+ "batch_size": 1,
25
+ "capacity_factor": null,
26
+ "cc_pipeline_tiling_factor": 2,
27
+ "checkpoint_id": "hf-internal-testing/tiny-random-GraniteForCausalLM",
28
+ "checkpoint_revision": "c3074ebc0ac2fe545305f5e5f6cce2cc9b2aa0c5",
29
+ "continuous_batching": false,
30
+ "enable_bucketing": false,
31
+ "ep_degree": 1,
32
+ "flash_decoding_enabled": false,
33
+ "fused_qkv": true,
34
+ "glu_mlp": true,
35
+ "is_chunked_prefill": false,
36
+ "local_ranks_size": 2,
37
+ "logical_nc_config": 1,
38
+ "max_batch_size": 1,
39
+ "max_context_length": 100,
40
+ "max_topk": 256,
41
+ "mlp_kernel_enabled": false,
42
+ "mlp_kernel_fuse_residual_add": false,
43
+ "n_active_tokens": 100,
44
+ "neuronxcc_version": "2.17.194.0+d312836f",
45
+ "num_cores_per_group": 1,
46
+ "on_device_sampling": true,
47
+ "optimum_neuron_version": "0.2.1",
48
+ "output_logits": false,
49
+ "padding_side": "right",
50
+ "pp_degree": 1,
51
+ "qk_layernorm": false,
52
+ "qkv_kernel_enabled": false,
53
+ "rpl_reduce_dtype": "bfloat16",
54
+ "sequence_length": 100,
55
+ "sequence_parallel_enabled": false,
56
+ "speculation_length": 0,
57
+ "start_rank_id": 0,
58
+ "target": null,
59
+ "torch_dtype": "bfloat16",
60
+ "tp_degree": 2,
61
+ "vocab_parallel": false
62
+ },
63
+ "num_attention_heads": 4,
64
+ "num_hidden_layers": 2,
65
+ "num_key_value_heads": 4,
66
+ "residual_multiplier": 1.0,
67
+ "rms_norm_eps": 1e-06,
68
+ "rope_scaling": null,
69
+ "rope_theta": 10000.0,
70
+ "tie_word_embeddings": false,
71
+ "use_cache": true,
72
+ "vocab_size": 49152
73
+ }
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.1/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/2b0fa8ae0a115fad4c29.json ADDED
@@ -0,0 +1,73 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "embedding_multiplier": 1.0,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 32,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 64,
16
+ "logits_scaling": 1.0,
17
+ "max_position_embeddings": 2048,
18
+ "mlp_bias": false,
19
+ "model_type": "granite",
20
+ "neuron": {
21
+ "_serialized_key": "NxDNeuronConfig",
22
+ "async_mode": false,
23
+ "attn_kernel_enabled": false,
24
+ "batch_size": 2,
25
+ "capacity_factor": null,
26
+ "cc_pipeline_tiling_factor": 2,
27
+ "checkpoint_id": "hf-internal-testing/tiny-random-GraniteForCausalLM",
28
+ "checkpoint_revision": "c3074ebc0ac2fe545305f5e5f6cce2cc9b2aa0c5",
29
+ "continuous_batching": true,
30
+ "enable_bucketing": false,
31
+ "ep_degree": 1,
32
+ "flash_decoding_enabled": false,
33
+ "fused_qkv": true,
34
+ "glu_mlp": true,
35
+ "is_chunked_prefill": false,
36
+ "local_ranks_size": 2,
37
+ "logical_nc_config": 1,
38
+ "max_batch_size": 2,
39
+ "max_context_length": 100,
40
+ "max_topk": 256,
41
+ "mlp_kernel_enabled": false,
42
+ "mlp_kernel_fuse_residual_add": false,
43
+ "n_active_tokens": 100,
44
+ "neuronxcc_version": "2.17.194.0+d312836f",
45
+ "num_cores_per_group": 1,
46
+ "on_device_sampling": false,
47
+ "optimum_neuron_version": "0.2.1",
48
+ "output_logits": false,
49
+ "padding_side": "right",
50
+ "pp_degree": 1,
51
+ "qk_layernorm": false,
52
+ "qkv_kernel_enabled": false,
53
+ "rpl_reduce_dtype": "float16",
54
+ "sequence_length": 100,
55
+ "sequence_parallel_enabled": false,
56
+ "speculation_length": 0,
57
+ "start_rank_id": 0,
58
+ "target": null,
59
+ "torch_dtype": "float16",
60
+ "tp_degree": 2,
61
+ "vocab_parallel": false
62
+ },
63
+ "num_attention_heads": 4,
64
+ "num_hidden_layers": 2,
65
+ "num_key_value_heads": 4,
66
+ "residual_multiplier": 1.0,
67
+ "rms_norm_eps": 1e-06,
68
+ "rope_scaling": null,
69
+ "rope_theta": 10000.0,
70
+ "tie_word_embeddings": false,
71
+ "use_cache": true,
72
+ "vocab_size": 49152
73
+ }
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.1/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/bfc2e217ecf039318115.json ADDED
@@ -0,0 +1,73 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "embedding_multiplier": 1.0,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 32,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 64,
16
+ "logits_scaling": 1.0,
17
+ "max_position_embeddings": 2048,
18
+ "mlp_bias": false,
19
+ "model_type": "granite",
20
+ "neuron": {
21
+ "_serialized_key": "NxDNeuronConfig",
22
+ "async_mode": false,
23
+ "attn_kernel_enabled": false,
24
+ "batch_size": 1,
25
+ "capacity_factor": null,
26
+ "cc_pipeline_tiling_factor": 2,
27
+ "checkpoint_id": "hf-internal-testing/tiny-random-GraniteForCausalLM",
28
+ "checkpoint_revision": "c3074ebc0ac2fe545305f5e5f6cce2cc9b2aa0c5",
29
+ "continuous_batching": false,
30
+ "enable_bucketing": false,
31
+ "ep_degree": 1,
32
+ "flash_decoding_enabled": false,
33
+ "fused_qkv": true,
34
+ "glu_mlp": true,
35
+ "is_chunked_prefill": false,
36
+ "local_ranks_size": 2,
37
+ "logical_nc_config": 1,
38
+ "max_batch_size": 1,
39
+ "max_context_length": 100,
40
+ "max_topk": 256,
41
+ "mlp_kernel_enabled": false,
42
+ "mlp_kernel_fuse_residual_add": false,
43
+ "n_active_tokens": 100,
44
+ "neuronxcc_version": "2.17.194.0+d312836f",
45
+ "num_cores_per_group": 1,
46
+ "on_device_sampling": true,
47
+ "optimum_neuron_version": "0.2.1",
48
+ "output_logits": false,
49
+ "padding_side": "right",
50
+ "pp_degree": 1,
51
+ "qk_layernorm": false,
52
+ "qkv_kernel_enabled": false,
53
+ "rpl_reduce_dtype": "float16",
54
+ "sequence_length": 100,
55
+ "sequence_parallel_enabled": false,
56
+ "speculation_length": 0,
57
+ "start_rank_id": 0,
58
+ "target": null,
59
+ "torch_dtype": "float16",
60
+ "tp_degree": 2,
61
+ "vocab_parallel": false
62
+ },
63
+ "num_attention_heads": 4,
64
+ "num_hidden_layers": 2,
65
+ "num_key_value_heads": 4,
66
+ "residual_multiplier": 1.0,
67
+ "rms_norm_eps": 1e-06,
68
+ "rope_scaling": null,
69
+ "rope_theta": 10000.0,
70
+ "tie_word_embeddings": false,
71
+ "use_cache": true,
72
+ "vocab_size": 49152
73
+ }
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.1/llama/llamafactory/tiny-random-Llama-3/0c86e63e398c5b4e2cfd.json ADDED
@@ -0,0 +1,77 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "head_dim": 4,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 16,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 64,
15
+ "max_position_embeddings": 131072,
16
+ "mlp_bias": false,
17
+ "model_type": "llama",
18
+ "neuron": {
19
+ "_serialized_key": "NxDNeuronConfig",
20
+ "async_mode": false,
21
+ "attn_kernel_enabled": false,
22
+ "batch_size": 1,
23
+ "capacity_factor": null,
24
+ "cc_pipeline_tiling_factor": 2,
25
+ "checkpoint_id": "llamafactory/tiny-random-Llama-3",
26
+ "checkpoint_revision": "bf2a2e3bf199ad2ee96f02a3c00246c608db22a8",
27
+ "continuous_batching": false,
28
+ "enable_bucketing": false,
29
+ "ep_degree": 1,
30
+ "flash_decoding_enabled": false,
31
+ "fused_qkv": true,
32
+ "glu_mlp": true,
33
+ "is_chunked_prefill": false,
34
+ "local_ranks_size": 2,
35
+ "logical_nc_config": 1,
36
+ "max_batch_size": 1,
37
+ "max_context_length": 100,
38
+ "max_topk": 256,
39
+ "mlp_kernel_enabled": false,
40
+ "mlp_kernel_fuse_residual_add": false,
41
+ "n_active_tokens": 100,
42
+ "neuronxcc_version": "2.17.194.0+d312836f",
43
+ "num_cores_per_group": 1,
44
+ "on_device_sampling": true,
45
+ "optimum_neuron_version": "0.2.1",
46
+ "output_logits": false,
47
+ "padding_side": "right",
48
+ "pp_degree": 1,
49
+ "qk_layernorm": false,
50
+ "qkv_kernel_enabled": false,
51
+ "rpl_reduce_dtype": "float16",
52
+ "sequence_length": 100,
53
+ "sequence_parallel_enabled": false,
54
+ "speculation_length": 0,
55
+ "start_rank_id": 0,
56
+ "target": null,
57
+ "torch_dtype": "float16",
58
+ "tp_degree": 2,
59
+ "vocab_parallel": false
60
+ },
61
+ "num_attention_heads": 4,
62
+ "num_hidden_layers": 2,
63
+ "num_key_value_heads": 4,
64
+ "pretraining_tp": 1,
65
+ "rms_norm_eps": 1e-05,
66
+ "rope_scaling": {
67
+ "factor": 8.0,
68
+ "high_freq_factor": 4.0,
69
+ "low_freq_factor": 1.0,
70
+ "original_max_position_embeddings": 8192,
71
+ "rope_type": "llama3"
72
+ },
73
+ "rope_theta": 500000.0,
74
+ "tie_word_embeddings": false,
75
+ "use_cache": true,
76
+ "vocab_size": 128256
77
+ }
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.1/llama/llamafactory/tiny-random-Llama-3/c58bc107cced18b694d6.json ADDED
@@ -0,0 +1,77 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "head_dim": 4,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 16,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 64,
15
+ "max_position_embeddings": 131072,
16
+ "mlp_bias": false,
17
+ "model_type": "llama",
18
+ "neuron": {
19
+ "_serialized_key": "NxDNeuronConfig",
20
+ "async_mode": false,
21
+ "attn_kernel_enabled": false,
22
+ "batch_size": 1,
23
+ "capacity_factor": null,
24
+ "cc_pipeline_tiling_factor": 2,
25
+ "checkpoint_id": "llamafactory/tiny-random-Llama-3",
26
+ "checkpoint_revision": "bf2a2e3bf199ad2ee96f02a3c00246c608db22a8",
27
+ "continuous_batching": false,
28
+ "enable_bucketing": false,
29
+ "ep_degree": 1,
30
+ "flash_decoding_enabled": false,
31
+ "fused_qkv": true,
32
+ "glu_mlp": true,
33
+ "is_chunked_prefill": false,
34
+ "local_ranks_size": 2,
35
+ "logical_nc_config": 1,
36
+ "max_batch_size": 1,
37
+ "max_context_length": 100,
38
+ "max_topk": 256,
39
+ "mlp_kernel_enabled": false,
40
+ "mlp_kernel_fuse_residual_add": false,
41
+ "n_active_tokens": 100,
42
+ "neuronxcc_version": "2.17.194.0+d312836f",
43
+ "num_cores_per_group": 1,
44
+ "on_device_sampling": true,
45
+ "optimum_neuron_version": "0.2.1",
46
+ "output_logits": false,
47
+ "padding_side": "right",
48
+ "pp_degree": 1,
49
+ "qk_layernorm": false,
50
+ "qkv_kernel_enabled": false,
51
+ "rpl_reduce_dtype": "bfloat16",
52
+ "sequence_length": 100,
53
+ "sequence_parallel_enabled": false,
54
+ "speculation_length": 0,
55
+ "start_rank_id": 0,
56
+ "target": null,
57
+ "torch_dtype": "bfloat16",
58
+ "tp_degree": 2,
59
+ "vocab_parallel": false
60
+ },
61
+ "num_attention_heads": 4,
62
+ "num_hidden_layers": 2,
63
+ "num_key_value_heads": 4,
64
+ "pretraining_tp": 1,
65
+ "rms_norm_eps": 1e-05,
66
+ "rope_scaling": {
67
+ "factor": 8.0,
68
+ "high_freq_factor": 4.0,
69
+ "low_freq_factor": 1.0,
70
+ "original_max_position_embeddings": 8192,
71
+ "rope_type": "llama3"
72
+ },
73
+ "rope_theta": 500000.0,
74
+ "tie_word_embeddings": false,
75
+ "use_cache": true,
76
+ "vocab_size": 128256
77
+ }
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.1/llama/llamafactory/tiny-random-Llama-3/ff2809df4f513c66db23.json ADDED
@@ -0,0 +1,77 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "head_dim": 4,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 16,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 64,
15
+ "max_position_embeddings": 131072,
16
+ "mlp_bias": false,
17
+ "model_type": "llama",
18
+ "neuron": {
19
+ "_serialized_key": "NxDNeuronConfig",
20
+ "async_mode": false,
21
+ "attn_kernel_enabled": false,
22
+ "batch_size": 2,
23
+ "capacity_factor": null,
24
+ "cc_pipeline_tiling_factor": 2,
25
+ "checkpoint_id": "llamafactory/tiny-random-Llama-3",
26
+ "checkpoint_revision": "bf2a2e3bf199ad2ee96f02a3c00246c608db22a8",
27
+ "continuous_batching": true,
28
+ "enable_bucketing": false,
29
+ "ep_degree": 1,
30
+ "flash_decoding_enabled": false,
31
+ "fused_qkv": true,
32
+ "glu_mlp": true,
33
+ "is_chunked_prefill": false,
34
+ "local_ranks_size": 2,
35
+ "logical_nc_config": 1,
36
+ "max_batch_size": 2,
37
+ "max_context_length": 100,
38
+ "max_topk": 256,
39
+ "mlp_kernel_enabled": false,
40
+ "mlp_kernel_fuse_residual_add": false,
41
+ "n_active_tokens": 100,
42
+ "neuronxcc_version": "2.17.194.0+d312836f",
43
+ "num_cores_per_group": 1,
44
+ "on_device_sampling": false,
45
+ "optimum_neuron_version": "0.2.1",
46
+ "output_logits": false,
47
+ "padding_side": "right",
48
+ "pp_degree": 1,
49
+ "qk_layernorm": false,
50
+ "qkv_kernel_enabled": false,
51
+ "rpl_reduce_dtype": "float16",
52
+ "sequence_length": 100,
53
+ "sequence_parallel_enabled": false,
54
+ "speculation_length": 0,
55
+ "start_rank_id": 0,
56
+ "target": null,
57
+ "torch_dtype": "float16",
58
+ "tp_degree": 2,
59
+ "vocab_parallel": false
60
+ },
61
+ "num_attention_heads": 4,
62
+ "num_hidden_layers": 2,
63
+ "num_key_value_heads": 4,
64
+ "pretraining_tp": 1,
65
+ "rms_norm_eps": 1e-05,
66
+ "rope_scaling": {
67
+ "factor": 8.0,
68
+ "high_freq_factor": 4.0,
69
+ "low_freq_factor": 1.0,
70
+ "original_max_position_embeddings": 8192,
71
+ "rope_type": "llama3"
72
+ },
73
+ "rope_theta": 500000.0,
74
+ "tie_word_embeddings": false,
75
+ "use_cache": true,
76
+ "vocab_size": 128256
77
+ }
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.1/llama/unsloth/Llama-3.2-1B-Instruct/c7a102927bd4d31a4086.json ADDED
@@ -0,0 +1,78 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "unsloth/Llama-3.2-1B-Instruct",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "LlamaForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "head_dim": 64,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 2048,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 8192,
15
+ "max_position_embeddings": 131072,
16
+ "mlp_bias": false,
17
+ "model_type": "llama",
18
+ "neuron": {
19
+ "_serialized_key": "NxDNeuronConfig",
20
+ "async_mode": false,
21
+ "attn_kernel_enabled": false,
22
+ "batch_size": 4,
23
+ "capacity_factor": null,
24
+ "cc_pipeline_tiling_factor": 2,
25
+ "checkpoint_id": "unsloth/Llama-3.2-1B-Instruct",
26
+ "checkpoint_revision": "5a8abab4a5d6f164389b1079fb721cfab8d7126c",
27
+ "continuous_batching": true,
28
+ "enable_bucketing": false,
29
+ "ep_degree": 1,
30
+ "flash_decoding_enabled": false,
31
+ "fused_qkv": true,
32
+ "glu_mlp": true,
33
+ "is_chunked_prefill": false,
34
+ "local_ranks_size": 2,
35
+ "logical_nc_config": 1,
36
+ "max_batch_size": 4,
37
+ "max_context_length": 4096,
38
+ "max_topk": 256,
39
+ "mlp_kernel_enabled": false,
40
+ "mlp_kernel_fuse_residual_add": false,
41
+ "n_active_tokens": 4096,
42
+ "neuronxcc_version": "2.17.194.0+d312836f",
43
+ "num_cores_per_group": 1,
44
+ "on_device_sampling": false,
45
+ "optimum_neuron_version": "0.2.1",
46
+ "output_logits": false,
47
+ "padding_side": "right",
48
+ "pp_degree": 1,
49
+ "qk_layernorm": false,
50
+ "qkv_kernel_enabled": false,
51
+ "rpl_reduce_dtype": "float16",
52
+ "sequence_length": 4096,
53
+ "sequence_parallel_enabled": false,
54
+ "speculation_length": 0,
55
+ "start_rank_id": 0,
56
+ "target": null,
57
+ "torch_dtype": "float16",
58
+ "tp_degree": 2,
59
+ "vocab_parallel": false
60
+ },
61
+ "num_attention_heads": 32,
62
+ "num_hidden_layers": 16,
63
+ "num_key_value_heads": 8,
64
+ "pretraining_tp": 1,
65
+ "rms_norm_eps": 1e-05,
66
+ "rope_scaling": {
67
+ "factor": 32.0,
68
+ "high_freq_factor": 4.0,
69
+ "low_freq_factor": 1.0,
70
+ "original_max_position_embeddings": 8192,
71
+ "rope_type": "llama3"
72
+ },
73
+ "rope_theta": 500000.0,
74
+ "tie_word_embeddings": true,
75
+ "unsloth_fixed": true,
76
+ "use_cache": true,
77
+ "vocab_size": 128256
78
+ }
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.1/mixtral/dacorvo/Mixtral-tiny/2654eeeba6e6693fe3a7.json ADDED
@@ -0,0 +1,73 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "head_dim": 32,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 1024,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 3584,
14
+ "max_position_embeddings": 1024,
15
+ "model_type": "mixtral",
16
+ "neuron": {
17
+ "_serialized_key": "NxDNeuronConfig",
18
+ "async_mode": false,
19
+ "attn_kernel_enabled": false,
20
+ "batch_size": 2,
21
+ "capacity_factor": null,
22
+ "cc_pipeline_tiling_factor": 2,
23
+ "checkpoint_id": "dacorvo/Mixtral-tiny",
24
+ "checkpoint_revision": "c557ba205ddff6ea911f4719e0d543d6c08356b6",
25
+ "continuous_batching": false,
26
+ "enable_bucketing": false,
27
+ "ep_degree": 1,
28
+ "flash_decoding_enabled": false,
29
+ "fused_qkv": false,
30
+ "glu_mlp": true,
31
+ "is_chunked_prefill": false,
32
+ "local_ranks_size": 2,
33
+ "logical_nc_config": 1,
34
+ "max_batch_size": 2,
35
+ "max_context_length": 100,
36
+ "max_topk": 256,
37
+ "mlp_kernel_enabled": false,
38
+ "mlp_kernel_fuse_residual_add": false,
39
+ "n_active_tokens": 100,
40
+ "neuronxcc_version": "2.17.194.0+d312836f",
41
+ "num_cores_per_group": 1,
42
+ "on_device_sampling": false,
43
+ "optimum_neuron_version": "0.2.1",
44
+ "output_logits": false,
45
+ "padding_side": "right",
46
+ "pp_degree": 1,
47
+ "qk_layernorm": false,
48
+ "qkv_kernel_enabled": false,
49
+ "rpl_reduce_dtype": "float16",
50
+ "sequence_length": 100,
51
+ "sequence_parallel_enabled": false,
52
+ "speculation_length": 0,
53
+ "start_rank_id": 0,
54
+ "target": null,
55
+ "torch_dtype": "float16",
56
+ "tp_degree": 2,
57
+ "vocab_parallel": false
58
+ },
59
+ "num_attention_heads": 32,
60
+ "num_experts_per_tok": 2,
61
+ "num_hidden_layers": 2,
62
+ "num_key_value_heads": 8,
63
+ "num_local_experts": 8,
64
+ "output_router_logits": false,
65
+ "rms_norm_eps": 1e-05,
66
+ "rope_theta": 10000.0,
67
+ "router_aux_loss_coef": 0.001,
68
+ "router_jitter_noise": 0.0,
69
+ "sliding_window": 4096,
70
+ "tie_word_embeddings": false,
71
+ "use_cache": true,
72
+ "vocab_size": 32000
73
+ }
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.1/mixtral/dacorvo/Mixtral-tiny/3f63a19d4af1b82d4569.json ADDED
@@ -0,0 +1,73 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "head_dim": 32,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 1024,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 3584,
14
+ "max_position_embeddings": 1024,
15
+ "model_type": "mixtral",
16
+ "neuron": {
17
+ "_serialized_key": "NxDNeuronConfig",
18
+ "async_mode": false,
19
+ "attn_kernel_enabled": false,
20
+ "batch_size": 1,
21
+ "capacity_factor": null,
22
+ "cc_pipeline_tiling_factor": 2,
23
+ "checkpoint_id": "dacorvo/Mixtral-tiny",
24
+ "checkpoint_revision": "c557ba205ddff6ea911f4719e0d543d6c08356b6",
25
+ "continuous_batching": false,
26
+ "enable_bucketing": false,
27
+ "ep_degree": 1,
28
+ "flash_decoding_enabled": false,
29
+ "fused_qkv": false,
30
+ "glu_mlp": true,
31
+ "is_chunked_prefill": false,
32
+ "local_ranks_size": 2,
33
+ "logical_nc_config": 1,
34
+ "max_batch_size": 1,
35
+ "max_context_length": 100,
36
+ "max_topk": 256,
37
+ "mlp_kernel_enabled": false,
38
+ "mlp_kernel_fuse_residual_add": false,
39
+ "n_active_tokens": 100,
40
+ "neuronxcc_version": "2.17.194.0+d312836f",
41
+ "num_cores_per_group": 1,
42
+ "on_device_sampling": false,
43
+ "optimum_neuron_version": "0.2.1",
44
+ "output_logits": false,
45
+ "padding_side": "right",
46
+ "pp_degree": 1,
47
+ "qk_layernorm": false,
48
+ "qkv_kernel_enabled": false,
49
+ "rpl_reduce_dtype": "float16",
50
+ "sequence_length": 100,
51
+ "sequence_parallel_enabled": false,
52
+ "speculation_length": 0,
53
+ "start_rank_id": 0,
54
+ "target": null,
55
+ "torch_dtype": "float16",
56
+ "tp_degree": 2,
57
+ "vocab_parallel": false
58
+ },
59
+ "num_attention_heads": 32,
60
+ "num_experts_per_tok": 2,
61
+ "num_hidden_layers": 2,
62
+ "num_key_value_heads": 8,
63
+ "num_local_experts": 8,
64
+ "output_router_logits": false,
65
+ "rms_norm_eps": 1e-05,
66
+ "rope_theta": 10000.0,
67
+ "router_aux_loss_coef": 0.001,
68
+ "router_jitter_noise": 0.0,
69
+ "sliding_window": 4096,
70
+ "tie_word_embeddings": false,
71
+ "use_cache": true,
72
+ "vocab_size": 32000
73
+ }
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.1/mixtral/dacorvo/Mixtral-tiny/786147c74e4207def994.json ADDED
@@ -0,0 +1,73 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "head_dim": 32,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 1024,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 3584,
14
+ "max_position_embeddings": 1024,
15
+ "model_type": "mixtral",
16
+ "neuron": {
17
+ "_serialized_key": "NxDNeuronConfig",
18
+ "async_mode": false,
19
+ "attn_kernel_enabled": false,
20
+ "batch_size": 1,
21
+ "capacity_factor": null,
22
+ "cc_pipeline_tiling_factor": 2,
23
+ "checkpoint_id": "dacorvo/Mixtral-tiny",
24
+ "checkpoint_revision": "c557ba205ddff6ea911f4719e0d543d6c08356b6",
25
+ "continuous_batching": false,
26
+ "enable_bucketing": false,
27
+ "ep_degree": 1,
28
+ "flash_decoding_enabled": false,
29
+ "fused_qkv": false,
30
+ "glu_mlp": true,
31
+ "is_chunked_prefill": false,
32
+ "local_ranks_size": 2,
33
+ "logical_nc_config": 1,
34
+ "max_batch_size": 1,
35
+ "max_context_length": 100,
36
+ "max_topk": 256,
37
+ "mlp_kernel_enabled": false,
38
+ "mlp_kernel_fuse_residual_add": false,
39
+ "n_active_tokens": 100,
40
+ "neuronxcc_version": "2.17.194.0+d312836f",
41
+ "num_cores_per_group": 1,
42
+ "on_device_sampling": false,
43
+ "optimum_neuron_version": "0.2.1",
44
+ "output_logits": false,
45
+ "padding_side": "right",
46
+ "pp_degree": 1,
47
+ "qk_layernorm": false,
48
+ "qkv_kernel_enabled": false,
49
+ "rpl_reduce_dtype": "bfloat16",
50
+ "sequence_length": 100,
51
+ "sequence_parallel_enabled": false,
52
+ "speculation_length": 0,
53
+ "start_rank_id": 0,
54
+ "target": null,
55
+ "torch_dtype": "bfloat16",
56
+ "tp_degree": 2,
57
+ "vocab_parallel": false
58
+ },
59
+ "num_attention_heads": 32,
60
+ "num_experts_per_tok": 2,
61
+ "num_hidden_layers": 2,
62
+ "num_key_value_heads": 8,
63
+ "num_local_experts": 8,
64
+ "output_router_logits": false,
65
+ "rms_norm_eps": 1e-05,
66
+ "rope_theta": 10000.0,
67
+ "router_aux_loss_coef": 0.001,
68
+ "router_jitter_noise": 0.0,
69
+ "sliding_window": 4096,
70
+ "tie_word_embeddings": false,
71
+ "use_cache": true,
72
+ "vocab_size": 32000
73
+ }
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.1/phi3/yujiepan/phi-4-tiny-random/0f1bfbaf10d2c07abd9a.json ADDED
@@ -0,0 +1,74 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "embd_pdrop": 0.0,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 16,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 32,
16
+ "max_position_embeddings": 16384,
17
+ "model_type": "phi3",
18
+ "neuron": {
19
+ "_serialized_key": "NxDNeuronConfig",
20
+ "async_mode": false,
21
+ "attn_kernel_enabled": false,
22
+ "batch_size": 1,
23
+ "capacity_factor": null,
24
+ "cc_pipeline_tiling_factor": 2,
25
+ "checkpoint_id": "yujiepan/phi-4-tiny-random",
26
+ "checkpoint_revision": "18a9a1168dc97ac6d128f811925670c275610f5a",
27
+ "continuous_batching": false,
28
+ "enable_bucketing": false,
29
+ "ep_degree": 1,
30
+ "flash_decoding_enabled": false,
31
+ "fused_qkv": true,
32
+ "glu_mlp": true,
33
+ "is_chunked_prefill": false,
34
+ "local_ranks_size": 2,
35
+ "logical_nc_config": 1,
36
+ "max_batch_size": 1,
37
+ "max_context_length": 100,
38
+ "max_topk": 256,
39
+ "mlp_kernel_enabled": false,
40
+ "mlp_kernel_fuse_residual_add": false,
41
+ "n_active_tokens": 100,
42
+ "neuronxcc_version": "2.17.194.0+d312836f",
43
+ "num_cores_per_group": 1,
44
+ "on_device_sampling": true,
45
+ "optimum_neuron_version": "0.2.1",
46
+ "output_logits": false,
47
+ "padding_side": "right",
48
+ "pp_degree": 1,
49
+ "qk_layernorm": false,
50
+ "qkv_kernel_enabled": false,
51
+ "rpl_reduce_dtype": "float16",
52
+ "sequence_length": 100,
53
+ "sequence_parallel_enabled": false,
54
+ "speculation_length": 0,
55
+ "start_rank_id": 0,
56
+ "target": null,
57
+ "torch_dtype": "float16",
58
+ "tp_degree": 2,
59
+ "vocab_parallel": false
60
+ },
61
+ "num_attention_heads": 2,
62
+ "num_hidden_layers": 2,
63
+ "num_key_value_heads": 1,
64
+ "original_max_position_embeddings": 16384,
65
+ "partial_rotary_factor": 1.0,
66
+ "resid_pdrop": 0.0,
67
+ "rms_norm_eps": 1e-05,
68
+ "rope_scaling": null,
69
+ "rope_theta": 250000,
70
+ "sliding_window": null,
71
+ "tie_word_embeddings": false,
72
+ "use_cache": true,
73
+ "vocab_size": 100352
74
+ }
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.1/phi3/yujiepan/phi-4-tiny-random/121619e98da8e946a4ce.json ADDED
@@ -0,0 +1,74 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "embd_pdrop": 0.0,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 16,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 32,
16
+ "max_position_embeddings": 16384,
17
+ "model_type": "phi3",
18
+ "neuron": {
19
+ "_serialized_key": "NxDNeuronConfig",
20
+ "async_mode": false,
21
+ "attn_kernel_enabled": false,
22
+ "batch_size": 1,
23
+ "capacity_factor": null,
24
+ "cc_pipeline_tiling_factor": 2,
25
+ "checkpoint_id": "yujiepan/phi-4-tiny-random",
26
+ "checkpoint_revision": "18a9a1168dc97ac6d128f811925670c275610f5a",
27
+ "continuous_batching": false,
28
+ "enable_bucketing": false,
29
+ "ep_degree": 1,
30
+ "flash_decoding_enabled": false,
31
+ "fused_qkv": true,
32
+ "glu_mlp": true,
33
+ "is_chunked_prefill": false,
34
+ "local_ranks_size": 2,
35
+ "logical_nc_config": 1,
36
+ "max_batch_size": 1,
37
+ "max_context_length": 100,
38
+ "max_topk": 256,
39
+ "mlp_kernel_enabled": false,
40
+ "mlp_kernel_fuse_residual_add": false,
41
+ "n_active_tokens": 100,
42
+ "neuronxcc_version": "2.17.194.0+d312836f",
43
+ "num_cores_per_group": 1,
44
+ "on_device_sampling": true,
45
+ "optimum_neuron_version": "0.2.1",
46
+ "output_logits": false,
47
+ "padding_side": "right",
48
+ "pp_degree": 1,
49
+ "qk_layernorm": false,
50
+ "qkv_kernel_enabled": false,
51
+ "rpl_reduce_dtype": "bfloat16",
52
+ "sequence_length": 100,
53
+ "sequence_parallel_enabled": false,
54
+ "speculation_length": 0,
55
+ "start_rank_id": 0,
56
+ "target": null,
57
+ "torch_dtype": "bfloat16",
58
+ "tp_degree": 2,
59
+ "vocab_parallel": false
60
+ },
61
+ "num_attention_heads": 2,
62
+ "num_hidden_layers": 2,
63
+ "num_key_value_heads": 1,
64
+ "original_max_position_embeddings": 16384,
65
+ "partial_rotary_factor": 1.0,
66
+ "resid_pdrop": 0.0,
67
+ "rms_norm_eps": 1e-05,
68
+ "rope_scaling": null,
69
+ "rope_theta": 250000,
70
+ "sliding_window": null,
71
+ "tie_word_embeddings": false,
72
+ "use_cache": true,
73
+ "vocab_size": 100352
74
+ }
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.1/phi3/yujiepan/phi-4-tiny-random/a6d251ae61cee42f8929.json ADDED
@@ -0,0 +1,74 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "embd_pdrop": 0.0,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 16,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 32,
16
+ "max_position_embeddings": 16384,
17
+ "model_type": "phi3",
18
+ "neuron": {
19
+ "_serialized_key": "NxDNeuronConfig",
20
+ "async_mode": false,
21
+ "attn_kernel_enabled": false,
22
+ "batch_size": 2,
23
+ "capacity_factor": null,
24
+ "cc_pipeline_tiling_factor": 2,
25
+ "checkpoint_id": "yujiepan/phi-4-tiny-random",
26
+ "checkpoint_revision": "18a9a1168dc97ac6d128f811925670c275610f5a",
27
+ "continuous_batching": true,
28
+ "enable_bucketing": false,
29
+ "ep_degree": 1,
30
+ "flash_decoding_enabled": false,
31
+ "fused_qkv": true,
32
+ "glu_mlp": true,
33
+ "is_chunked_prefill": false,
34
+ "local_ranks_size": 2,
35
+ "logical_nc_config": 1,
36
+ "max_batch_size": 2,
37
+ "max_context_length": 100,
38
+ "max_topk": 256,
39
+ "mlp_kernel_enabled": false,
40
+ "mlp_kernel_fuse_residual_add": false,
41
+ "n_active_tokens": 100,
42
+ "neuronxcc_version": "2.17.194.0+d312836f",
43
+ "num_cores_per_group": 1,
44
+ "on_device_sampling": false,
45
+ "optimum_neuron_version": "0.2.1",
46
+ "output_logits": false,
47
+ "padding_side": "right",
48
+ "pp_degree": 1,
49
+ "qk_layernorm": false,
50
+ "qkv_kernel_enabled": false,
51
+ "rpl_reduce_dtype": "float16",
52
+ "sequence_length": 100,
53
+ "sequence_parallel_enabled": false,
54
+ "speculation_length": 0,
55
+ "start_rank_id": 0,
56
+ "target": null,
57
+ "torch_dtype": "float16",
58
+ "tp_degree": 2,
59
+ "vocab_parallel": false
60
+ },
61
+ "num_attention_heads": 2,
62
+ "num_hidden_layers": 2,
63
+ "num_key_value_heads": 1,
64
+ "original_max_position_embeddings": 16384,
65
+ "partial_rotary_factor": 1.0,
66
+ "resid_pdrop": 0.0,
67
+ "rms_norm_eps": 1e-05,
68
+ "rope_scaling": null,
69
+ "rope_theta": 250000,
70
+ "sliding_window": null,
71
+ "tie_word_embeddings": false,
72
+ "use_cache": true,
73
+ "vocab_size": 100352
74
+ }
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.1/qwen2/yujiepan/qwen2.5-128k-tiny-random/2d2a3a890f7406761316.json ADDED
@@ -0,0 +1,75 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "hidden_act": "silu",
10
+ "hidden_size": 8,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 16,
13
+ "max_position_embeddings": 32768,
14
+ "max_window_layers": 1,
15
+ "model_type": "qwen2",
16
+ "neuron": {
17
+ "_serialized_key": "NxDNeuronConfig",
18
+ "async_mode": false,
19
+ "attn_kernel_enabled": false,
20
+ "batch_size": 2,
21
+ "capacity_factor": null,
22
+ "cc_pipeline_tiling_factor": 2,
23
+ "checkpoint_id": "yujiepan/qwen2.5-128k-tiny-random",
24
+ "checkpoint_revision": "c8296d4ca3f87782876d2382fbb6481d1beb8ef0",
25
+ "continuous_batching": true,
26
+ "enable_bucketing": false,
27
+ "ep_degree": 1,
28
+ "flash_decoding_enabled": false,
29
+ "fused_qkv": false,
30
+ "glu_mlp": true,
31
+ "is_chunked_prefill": false,
32
+ "local_ranks_size": 2,
33
+ "logical_nc_config": 1,
34
+ "max_batch_size": 2,
35
+ "max_context_length": 100,
36
+ "max_topk": 256,
37
+ "mlp_kernel_enabled": false,
38
+ "mlp_kernel_fuse_residual_add": false,
39
+ "n_active_tokens": 100,
40
+ "neuronxcc_version": "2.17.194.0+d312836f",
41
+ "num_cores_per_group": 1,
42
+ "on_device_sampling": false,
43
+ "optimum_neuron_version": "0.2.1",
44
+ "output_logits": false,
45
+ "padding_side": "right",
46
+ "pp_degree": 1,
47
+ "qk_layernorm": false,
48
+ "qkv_kernel_enabled": false,
49
+ "rpl_reduce_dtype": "float16",
50
+ "sequence_length": 100,
51
+ "sequence_parallel_enabled": false,
52
+ "speculation_length": 0,
53
+ "start_rank_id": 0,
54
+ "target": null,
55
+ "torch_dtype": "float16",
56
+ "tp_degree": 2,
57
+ "vocab_parallel": false
58
+ },
59
+ "num_attention_heads": 4,
60
+ "num_hidden_layers": 2,
61
+ "num_key_value_heads": 2,
62
+ "rms_norm_eps": 1e-06,
63
+ "rope_scaling": {
64
+ "factor": 4.0,
65
+ "original_max_position_embeddings": 32768,
66
+ "rope_type": "yarn",
67
+ "type": "yarn"
68
+ },
69
+ "rope_theta": 1000000.0,
70
+ "sliding_window": 131072,
71
+ "tie_word_embeddings": false,
72
+ "use_cache": true,
73
+ "use_sliding_window": false,
74
+ "vocab_size": 152064
75
+ }
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.1/qwen2/yujiepan/qwen2.5-128k-tiny-random/a638416d74f91230ba38.json ADDED
@@ -0,0 +1,75 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "hidden_act": "silu",
10
+ "hidden_size": 8,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 16,
13
+ "max_position_embeddings": 32768,
14
+ "max_window_layers": 1,
15
+ "model_type": "qwen2",
16
+ "neuron": {
17
+ "_serialized_key": "NxDNeuronConfig",
18
+ "async_mode": false,
19
+ "attn_kernel_enabled": false,
20
+ "batch_size": 1,
21
+ "capacity_factor": null,
22
+ "cc_pipeline_tiling_factor": 2,
23
+ "checkpoint_id": "yujiepan/qwen2.5-128k-tiny-random",
24
+ "checkpoint_revision": "c8296d4ca3f87782876d2382fbb6481d1beb8ef0",
25
+ "continuous_batching": false,
26
+ "enable_bucketing": false,
27
+ "ep_degree": 1,
28
+ "flash_decoding_enabled": false,
29
+ "fused_qkv": false,
30
+ "glu_mlp": true,
31
+ "is_chunked_prefill": false,
32
+ "local_ranks_size": 2,
33
+ "logical_nc_config": 1,
34
+ "max_batch_size": 1,
35
+ "max_context_length": 100,
36
+ "max_topk": 256,
37
+ "mlp_kernel_enabled": false,
38
+ "mlp_kernel_fuse_residual_add": false,
39
+ "n_active_tokens": 100,
40
+ "neuronxcc_version": "2.17.194.0+d312836f",
41
+ "num_cores_per_group": 1,
42
+ "on_device_sampling": true,
43
+ "optimum_neuron_version": "0.2.1",
44
+ "output_logits": false,
45
+ "padding_side": "right",
46
+ "pp_degree": 1,
47
+ "qk_layernorm": false,
48
+ "qkv_kernel_enabled": false,
49
+ "rpl_reduce_dtype": "float16",
50
+ "sequence_length": 100,
51
+ "sequence_parallel_enabled": false,
52
+ "speculation_length": 0,
53
+ "start_rank_id": 0,
54
+ "target": null,
55
+ "torch_dtype": "float16",
56
+ "tp_degree": 2,
57
+ "vocab_parallel": false
58
+ },
59
+ "num_attention_heads": 4,
60
+ "num_hidden_layers": 2,
61
+ "num_key_value_heads": 2,
62
+ "rms_norm_eps": 1e-06,
63
+ "rope_scaling": {
64
+ "factor": 4.0,
65
+ "original_max_position_embeddings": 32768,
66
+ "rope_type": "yarn",
67
+ "type": "yarn"
68
+ },
69
+ "rope_theta": 1000000.0,
70
+ "sliding_window": 131072,
71
+ "tie_word_embeddings": false,
72
+ "use_cache": true,
73
+ "use_sliding_window": false,
74
+ "vocab_size": 152064
75
+ }
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.1/qwen2/yujiepan/qwen2.5-128k-tiny-random/f8e1b768ef174982facc.json ADDED
@@ -0,0 +1,75 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "hidden_act": "silu",
10
+ "hidden_size": 8,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 16,
13
+ "max_position_embeddings": 32768,
14
+ "max_window_layers": 1,
15
+ "model_type": "qwen2",
16
+ "neuron": {
17
+ "_serialized_key": "NxDNeuronConfig",
18
+ "async_mode": false,
19
+ "attn_kernel_enabled": false,
20
+ "batch_size": 1,
21
+ "capacity_factor": null,
22
+ "cc_pipeline_tiling_factor": 2,
23
+ "checkpoint_id": "yujiepan/qwen2.5-128k-tiny-random",
24
+ "checkpoint_revision": "c8296d4ca3f87782876d2382fbb6481d1beb8ef0",
25
+ "continuous_batching": false,
26
+ "enable_bucketing": false,
27
+ "ep_degree": 1,
28
+ "flash_decoding_enabled": false,
29
+ "fused_qkv": false,
30
+ "glu_mlp": true,
31
+ "is_chunked_prefill": false,
32
+ "local_ranks_size": 2,
33
+ "logical_nc_config": 1,
34
+ "max_batch_size": 1,
35
+ "max_context_length": 100,
36
+ "max_topk": 256,
37
+ "mlp_kernel_enabled": false,
38
+ "mlp_kernel_fuse_residual_add": false,
39
+ "n_active_tokens": 100,
40
+ "neuronxcc_version": "2.17.194.0+d312836f",
41
+ "num_cores_per_group": 1,
42
+ "on_device_sampling": true,
43
+ "optimum_neuron_version": "0.2.1",
44
+ "output_logits": false,
45
+ "padding_side": "right",
46
+ "pp_degree": 1,
47
+ "qk_layernorm": false,
48
+ "qkv_kernel_enabled": false,
49
+ "rpl_reduce_dtype": "bfloat16",
50
+ "sequence_length": 100,
51
+ "sequence_parallel_enabled": false,
52
+ "speculation_length": 0,
53
+ "start_rank_id": 0,
54
+ "target": null,
55
+ "torch_dtype": "bfloat16",
56
+ "tp_degree": 2,
57
+ "vocab_parallel": false
58
+ },
59
+ "num_attention_heads": 4,
60
+ "num_hidden_layers": 2,
61
+ "num_key_value_heads": 2,
62
+ "rms_norm_eps": 1e-06,
63
+ "rope_scaling": {
64
+ "factor": 4.0,
65
+ "original_max_position_embeddings": 32768,
66
+ "rope_type": "yarn",
67
+ "type": "yarn"
68
+ },
69
+ "rope_theta": 1000000.0,
70
+ "sliding_window": 131072,
71
+ "tie_word_embeddings": false,
72
+ "use_cache": true,
73
+ "use_sliding_window": false,
74
+ "vocab_size": 152064
75
+ }
neuronxcc-2.17.194.0+d312836f/MODULE_57b107bd0499cc4986ac+63ea7cf2/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ "--enable-saturate-infinity --enable-mixed-precision-accumulation --model-type transformer -O1 --tensorizer-options='--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2' --auto-cast=none --internal-enable-dge-levels vector_dynamic_offsets --logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt --enable-internal-neff-wrapper"
neuronxcc-2.17.194.0+d312836f/MODULE_57b107bd0499cc4986ac+63ea7cf2/model.done ADDED
File without changes
neuronxcc-2.17.194.0+d312836f/MODULE_57b107bd0499cc4986ac+63ea7cf2/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a914031f26bb159d35140c96a1bdb74370696834acf754811e40b142622a02a4
3
+ size 68277
neuronxcc-2.17.194.0+d312836f/MODULE_57b107bd0499cc4986ac+63ea7cf2/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c868ec53737adc2fe50d3708f1a3cedfc2e769e9e5d707555376ee2a9a153028
3
+ size 257024
neuronxcc-2.17.194.0+d312836f/MODULE_57b107bd0499cc4986ac+63ea7cf2/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9eee4b03603e48bfdd262029efc30ecd40c1dbfcb10b94f627cf4e6696aa82f6
3
+ size 268322
neuronxcc-2.17.194.0+d312836f/MODULE_64950c85776a119cdf83+b001bea4/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ "--enable-saturate-infinity --enable-mixed-precision-accumulation --model-type transformer -O1 --tensorizer-options='--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2' --auto-cast=none --internal-enable-dge-levels vector_dynamic_offsets --logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"
neuronxcc-2.17.194.0+d312836f/MODULE_64950c85776a119cdf83+b001bea4/model.done ADDED
File without changes
neuronxcc-2.17.194.0+d312836f/MODULE_64950c85776a119cdf83+b001bea4/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d161deceb48f8f73a0aa6227a14f23c73b193b8c2f54ec986673930820610f16
3
+ size 82456
neuronxcc-2.17.194.0+d312836f/MODULE_64950c85776a119cdf83+b001bea4/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:06018e131b181177574a276f6294a56b20726b844088fb5bc1b9799f8bc4eef5
3
+ size 420864
neuronxcc-2.17.194.0+d312836f/MODULE_6bb5680c622866b3b45b+63ea7cf2/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ "--enable-saturate-infinity --enable-mixed-precision-accumulation --model-type transformer -O1 --tensorizer-options='--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2' --auto-cast=none --internal-enable-dge-levels vector_dynamic_offsets --logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt --enable-internal-neff-wrapper"
neuronxcc-2.17.194.0+d312836f/MODULE_6bb5680c622866b3b45b+63ea7cf2/model.done ADDED
File without changes
neuronxcc-2.17.194.0+d312836f/MODULE_6bb5680c622866b3b45b+63ea7cf2/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:12aceb27cdf723d91c999dfa8673d18f8cdabe145b276c11445d19e86a5f5145
3
+ size 79431
neuronxcc-2.17.194.0+d312836f/MODULE_6bb5680c622866b3b45b+63ea7cf2/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6cc031ed02ba0c6ba5aa974a455517d4c142fb3079c57115af3c0a4117b4e9f4
3
+ size 318464
neuronxcc-2.17.194.0+d312836f/MODULE_6bb5680c622866b3b45b+63ea7cf2/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:85e3ceea8c3aeb3881b637508b4f0bd79debc800f3a70ebf651b7c68f3ee8665
3
+ size 329762
neuronxcc-2.17.194.0+d312836f/MODULE_92d7a6b8bc621dee02b9+b001bea4/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ "--enable-saturate-infinity --enable-mixed-precision-accumulation --model-type transformer -O1 --tensorizer-options='--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2' --auto-cast=none --internal-enable-dge-levels vector_dynamic_offsets --logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"
neuronxcc-2.17.194.0+d312836f/MODULE_92d7a6b8bc621dee02b9+b001bea4/model.done ADDED
File without changes
neuronxcc-2.17.194.0+d312836f/MODULE_92d7a6b8bc621dee02b9+b001bea4/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6d96dab97289c7e8f0f86606e3e28fe6ba499bac93101c635261281f994837bf
3
+ size 81609
neuronxcc-2.17.194.0+d312836f/MODULE_92d7a6b8bc621dee02b9+b001bea4/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6283c2b8221af2ea89f2bedbae48a1b2f8d7fb42b4e630fc3ad3e16519d01b14
3
+ size 359424
neuronxcc-2.17.194.0+d312836f/MODULE_9a53db93d18e769ee7ea+b001bea4/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ "--enable-saturate-infinity --enable-mixed-precision-accumulation --model-type transformer -O1 --tensorizer-options='--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2' --auto-cast=none --internal-enable-dge-levels vector_dynamic_offsets --logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"
neuronxcc-2.17.194.0+d312836f/MODULE_9a53db93d18e769ee7ea+b001bea4/model.done ADDED
File without changes
neuronxcc-2.17.194.0+d312836f/MODULE_9a53db93d18e769ee7ea+b001bea4/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:316aa6d670ec51cbf810083a84be88f91b1bb9cbc07460058f55b9f20cc9dfc2
3
+ size 81607
neuronxcc-2.17.194.0+d312836f/MODULE_9a53db93d18e769ee7ea+b001bea4/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6613c5481ae06b85bb8abb953bc30cc50c4addff26b3478d9e71c68116482f9d
3
+ size 359424
neuronxcc-2.17.194.0+d312836f/MODULE_ae5ca4b91afd03b04b25+63ea7cf2/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ "--enable-saturate-infinity --enable-mixed-precision-accumulation --model-type transformer -O1 --tensorizer-options='--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2' --auto-cast=none --internal-enable-dge-levels vector_dynamic_offsets --logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt --enable-internal-neff-wrapper"
neuronxcc-2.17.194.0+d312836f/MODULE_ae5ca4b91afd03b04b25+63ea7cf2/model.done ADDED
File without changes
neuronxcc-2.17.194.0+d312836f/MODULE_ae5ca4b91afd03b04b25+63ea7cf2/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7bf1de332dd0f1615aa9bb39e8b50034c3b02b64240d386640900156271b83e8
3
+ size 68279
neuronxcc-2.17.194.0+d312836f/MODULE_ae5ca4b91afd03b04b25+63ea7cf2/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c809a8eebc804bd1bd9bddf1d8d93146a057e5c1d57af08032b900a9ab6d0830
3
+ size 257024
neuronxcc-2.17.194.0+d312836f/MODULE_ae5ca4b91afd03b04b25+63ea7cf2/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e3a45648500fccd8d6bada22f3ccf68bafd048f7a678b92125321c7667da7b02
3
+ size 268322