Hugging Face's logo

optimum-internal-testing
/
neuron-testing-cache

Model card Files Community
dacorvo HF Staff commited on
Commit
f578bfb
·
verified ·
1 Parent(s): 3ba696d

Synchronizing local compiler cache.

Browse files
Files changed (50) hide show
  1. .gitattributes +6 -0
  2. neuronxcc-2.18.121.0+9e31e41a/0_REGISTRY/0.2.0.dev8/llama/llamafactory/tiny-random-Llama-3/34f8b63bd243bc941ae1.json +55 -0
  3. neuronxcc-2.18.121.0+9e31e41a/0_REGISTRY/0.2.0.dev8/llama/llamafactory/tiny-random-Llama-3/5ac530b0eedece55f109.json +77 -0
  4. neuronxcc-2.18.121.0+9e31e41a/0_REGISTRY/0.2.0.dev8/llama/llamafactory/tiny-random-Llama-3/ac78980b4bce50b6fbc4.json +55 -0
  5. neuronxcc-2.18.121.0+9e31e41a/0_REGISTRY/0.2.0.dev8/llama/llamafactory/tiny-random-Llama-3/acfb4a1a634ed1fd984f.json +77 -0
  6. neuronxcc-2.18.121.0+9e31e41a/0_REGISTRY/0.2.0.dev8/llama/llamafactory/tiny-random-Llama-3/df95ebb72707172e4141.json +77 -0
  7. neuronxcc-2.18.121.0+9e31e41a/0_REGISTRY/0.2.0.dev8/llama/llamafactory/tiny-random-Llama-3/eb93770853f294e67a7b.json +55 -0
  8. neuronxcc-2.18.121.0+9e31e41a/0_REGISTRY/0.2.0.dev8/llama/unsloth/Llama-3.2-1B-Instruct/972a896364f30c9bc91a.json +78 -0
  9. neuronxcc-2.18.121.0+9e31e41a/MODULE_152c4fbd51eb1f6f90ec+84f3e719/compile_flags.json +1 -0
  10. neuronxcc-2.18.121.0+9e31e41a/MODULE_152c4fbd51eb1f6f90ec+84f3e719/model.done +0 -0
  11. neuronxcc-2.18.121.0+9e31e41a/MODULE_152c4fbd51eb1f6f90ec+84f3e719/model.hlo_module.pb +3 -0
  12. neuronxcc-2.18.121.0+9e31e41a/MODULE_152c4fbd51eb1f6f90ec+84f3e719/model.neff +3 -0
  13. neuronxcc-2.18.121.0+9e31e41a/MODULE_21d49e164d26352245e5+84f3e719/model.hlo_module.pb +1 -1
  14. neuronxcc-2.18.121.0+9e31e41a/MODULE_21d49e164d26352245e5+84f3e719/model.neff +1 -1
  15. neuronxcc-2.18.121.0+9e31e41a/MODULE_24ff9ac2787ce9a1d276+613edded/model.neff +1 -1
  16. neuronxcc-2.18.121.0+9e31e41a/MODULE_320f2622d4d0c9fdd0f1+613edded/model.neff +1 -1
  17. neuronxcc-2.18.121.0+9e31e41a/MODULE_331276a07386ee77d52e+431f5505/compile_flags.json +1 -0
  18. neuronxcc-2.18.121.0+9e31e41a/MODULE_331276a07386ee77d52e+431f5505/model.done +0 -0
  19. neuronxcc-2.18.121.0+9e31e41a/MODULE_331276a07386ee77d52e+431f5505/model.hlo_module.pb +3 -0
  20. neuronxcc-2.18.121.0+9e31e41a/MODULE_331276a07386ee77d52e+431f5505/model.neff +0 -0
  21. neuronxcc-2.18.121.0+9e31e41a/MODULE_34a6b42796c8b4e2f58b+431f5505/model.neff +1 -1
  22. neuronxcc-2.18.121.0+9e31e41a/MODULE_40a0e75a65ac51fdd01a+613edded/model.neff +1 -1
  23. neuronxcc-2.18.121.0+9e31e41a/MODULE_5368928916623911b1f2+84f3e719/model.hlo_module.pb +1 -1
  24. neuronxcc-2.18.121.0+9e31e41a/MODULE_5368928916623911b1f2+84f3e719/model.neff +1 -1
  25. neuronxcc-2.18.121.0+9e31e41a/MODULE_788eb0c6a9b0ca759eca+5be477de/model.neff +1 -1
  26. neuronxcc-2.18.121.0+9e31e41a/MODULE_788eb0c6a9b0ca759eca+5be477de/wrapped_neff.hlo +1 -1
  27. neuronxcc-2.18.121.0+9e31e41a/MODULE_96404ade088398b2f3d5+431f5505/model.neff +0 -0
  28. neuronxcc-2.18.121.0+9e31e41a/MODULE_a669c4bb07e1a9435b7c+431f5505/compile_flags.json +1 -0
  29. neuronxcc-2.18.121.0+9e31e41a/MODULE_a669c4bb07e1a9435b7c+431f5505/model.done +0 -0
  30. neuronxcc-2.18.121.0+9e31e41a/MODULE_a669c4bb07e1a9435b7c+431f5505/model.hlo_module.pb +3 -0
  31. neuronxcc-2.18.121.0+9e31e41a/MODULE_a669c4bb07e1a9435b7c+431f5505/model.neff +0 -0
  32. neuronxcc-2.18.121.0+9e31e41a/MODULE_a77b880fafe6f97b3d8e+5be477de/compile_flags.json +1 -0
  33. neuronxcc-2.18.121.0+9e31e41a/MODULE_a77b880fafe6f97b3d8e+5be477de/model.done +0 -0
  34. neuronxcc-2.18.121.0+9e31e41a/MODULE_a77b880fafe6f97b3d8e+5be477de/model.hlo_module.pb +3 -0
  35. neuronxcc-2.18.121.0+9e31e41a/MODULE_a77b880fafe6f97b3d8e+5be477de/model.neff +3 -0
  36. neuronxcc-2.18.121.0+9e31e41a/MODULE_a77b880fafe6f97b3d8e+5be477de/wrapped_neff.hlo +3 -0
  37. neuronxcc-2.18.121.0+9e31e41a/MODULE_b94c33a1002b2de0e6f0+5be477de/model.neff +1 -1
  38. neuronxcc-2.18.121.0+9e31e41a/MODULE_b94c33a1002b2de0e6f0+5be477de/wrapped_neff.hlo +1 -1
  39. neuronxcc-2.18.121.0+9e31e41a/MODULE_ba6a6bbc03ee6dbd2e65+5be477de/compile_flags.json +1 -0
  40. neuronxcc-2.18.121.0+9e31e41a/MODULE_ba6a6bbc03ee6dbd2e65+5be477de/model.done +0 -0
  41. neuronxcc-2.18.121.0+9e31e41a/MODULE_ba6a6bbc03ee6dbd2e65+5be477de/model.hlo_module.pb +3 -0
  42. neuronxcc-2.18.121.0+9e31e41a/MODULE_ba6a6bbc03ee6dbd2e65+5be477de/model.neff +3 -0
  43. neuronxcc-2.18.121.0+9e31e41a/MODULE_ba6a6bbc03ee6dbd2e65+5be477de/wrapped_neff.hlo +3 -0
  44. neuronxcc-2.18.121.0+9e31e41a/MODULE_c49c0c3715f68c22b32f+613edded/model.neff +1 -1
  45. neuronxcc-2.18.121.0+9e31e41a/MODULE_df48af4bf01af7f3857e+613edded/model.neff +1 -1
  46. neuronxcc-2.18.121.0+9e31e41a/MODULE_e3e99dc2a67feb5066f9+84f3e719/compile_flags.json +1 -0
  47. neuronxcc-2.18.121.0+9e31e41a/MODULE_e3e99dc2a67feb5066f9+84f3e719/model.done +0 -0
  48. neuronxcc-2.18.121.0+9e31e41a/MODULE_e3e99dc2a67feb5066f9+84f3e719/model.hlo_module.pb +3 -0
  49. neuronxcc-2.18.121.0+9e31e41a/MODULE_e3e99dc2a67feb5066f9+84f3e719/model.neff +3 -0
  50. neuronxcc-2.18.121.0+9e31e41a/MODULE_fda7fb53c475ba393ed7+613edded/model.neff +1 -1
.gitattributes CHANGED
@@ -2800,3 +2800,9 @@ neuronxcc-2.18.121.0+9e31e41a/MODULE_5e70482efa3553464212+613edded/model.neff fi
2800
  neuronxcc-2.18.121.0+9e31e41a/MODULE_b94c33a1002b2de0e6f0+5be477de/model.neff filter=lfs diff=lfs merge=lfs -text
2801
  neuronxcc-2.18.121.0+9e31e41a/MODULE_b94c33a1002b2de0e6f0+5be477de/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
2802
  neuronxcc-2.18.121.0+9e31e41a/MODULE_c63be8d0579327b9cd06+613edded/model.neff filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
2800
  neuronxcc-2.18.121.0+9e31e41a/MODULE_b94c33a1002b2de0e6f0+5be477de/model.neff filter=lfs diff=lfs merge=lfs -text
2801
  neuronxcc-2.18.121.0+9e31e41a/MODULE_b94c33a1002b2de0e6f0+5be477de/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
2802
  neuronxcc-2.18.121.0+9e31e41a/MODULE_c63be8d0579327b9cd06+613edded/model.neff filter=lfs diff=lfs merge=lfs -text
2803
+ neuronxcc-2.18.121.0+9e31e41a/MODULE_152c4fbd51eb1f6f90ec+84f3e719/model.neff filter=lfs diff=lfs merge=lfs -text
2804
+ neuronxcc-2.18.121.0+9e31e41a/MODULE_a77b880fafe6f97b3d8e+5be477de/model.neff filter=lfs diff=lfs merge=lfs -text
2805
+ neuronxcc-2.18.121.0+9e31e41a/MODULE_a77b880fafe6f97b3d8e+5be477de/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
2806
+ neuronxcc-2.18.121.0+9e31e41a/MODULE_ba6a6bbc03ee6dbd2e65+5be477de/model.neff filter=lfs diff=lfs merge=lfs -text
2807
+ neuronxcc-2.18.121.0+9e31e41a/MODULE_ba6a6bbc03ee6dbd2e65+5be477de/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
2808
+ neuronxcc-2.18.121.0+9e31e41a/MODULE_e3e99dc2a67feb5066f9+84f3e719/model.neff filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.18.121.0+9e31e41a/0_REGISTRY/0.2.0.dev8/llama/llamafactory/tiny-random-Llama-3/34f8b63bd243bc941ae1.json ADDED
@@ -0,0 +1,55 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "HloNeuronConfig",
20
+ "all_reduce_dtype": null,
21
+ "allow_flash_attention": true,
22
+ "attention_layout": "BSH",
23
+ "attn_output_transposed": false,
24
+ "auto_cast_type": "fp16",
25
+ "batch_size": 2,
26
+ "checkpoint_id": "llamafactory/tiny-random-Llama-3",
27
+ "checkpoint_revision": "bf2a2e3bf199ad2ee96f02a3c00246c608db22a8",
28
+ "collectives_layout": "HSB",
29
+ "continuous_batching": true,
30
+ "fuse_qkv": true,
31
+ "group_query_attention": null,
32
+ "log_softmax_scores": false,
33
+ "neuronxcc_version": "2.18.121.0+9e31e41a",
34
+ "optimum_neuron_version": "0.2.0.dev8",
35
+ "output_all_logits": false,
36
+ "sequence_length": 100,
37
+ "tp_degree": 2
38
+ },
39
+ "num_attention_heads": 4,
40
+ "num_hidden_layers": 2,
41
+ "num_key_value_heads": 4,
42
+ "pretraining_tp": 1,
43
+ "rms_norm_eps": 1e-05,
44
+ "rope_scaling": {
45
+ "factor": 8.0,
46
+ "high_freq_factor": 4.0,
47
+ "low_freq_factor": 1.0,
48
+ "original_max_position_embeddings": 8192,
49
+ "rope_type": "llama3"
50
+ },
51
+ "rope_theta": 500000.0,
52
+ "tie_word_embeddings": false,
53
+ "use_cache": true,
54
+ "vocab_size": 128256
55
+ }
neuronxcc-2.18.121.0+9e31e41a/0_REGISTRY/0.2.0.dev8/llama/llamafactory/tiny-random-Llama-3/5ac530b0eedece55f109.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": "2.18.121.0+9e31e41a",
43
+ "num_cores_per_group": 1,
44
+ "on_device_sampling": true,
45
+ "optimum_neuron_version": "0.2.0.dev8",
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-2.18.121.0+9e31e41a/0_REGISTRY/0.2.0.dev8/llama/llamafactory/tiny-random-Llama-3/ac78980b4bce50b6fbc4.json ADDED
@@ -0,0 +1,55 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "HloNeuronConfig",
20
+ "all_reduce_dtype": null,
21
+ "allow_flash_attention": true,
22
+ "attention_layout": "BSH",
23
+ "attn_output_transposed": false,
24
+ "auto_cast_type": "fp16",
25
+ "batch_size": 1,
26
+ "checkpoint_id": "llamafactory/tiny-random-Llama-3",
27
+ "checkpoint_revision": "bf2a2e3bf199ad2ee96f02a3c00246c608db22a8",
28
+ "collectives_layout": "HSB",
29
+ "continuous_batching": false,
30
+ "fuse_qkv": true,
31
+ "group_query_attention": null,
32
+ "log_softmax_scores": false,
33
+ "neuronxcc_version": "2.18.121.0+9e31e41a",
34
+ "optimum_neuron_version": "0.2.0.dev8",
35
+ "output_all_logits": false,
36
+ "sequence_length": 100,
37
+ "tp_degree": 2
38
+ },
39
+ "num_attention_heads": 4,
40
+ "num_hidden_layers": 2,
41
+ "num_key_value_heads": 4,
42
+ "pretraining_tp": 1,
43
+ "rms_norm_eps": 1e-05,
44
+ "rope_scaling": {
45
+ "factor": 8.0,
46
+ "high_freq_factor": 4.0,
47
+ "low_freq_factor": 1.0,
48
+ "original_max_position_embeddings": 8192,
49
+ "rope_type": "llama3"
50
+ },
51
+ "rope_theta": 500000.0,
52
+ "tie_word_embeddings": false,
53
+ "use_cache": true,
54
+ "vocab_size": 128256
55
+ }
neuronxcc-2.18.121.0+9e31e41a/0_REGISTRY/0.2.0.dev8/llama/llamafactory/tiny-random-Llama-3/acfb4a1a634ed1fd984f.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": "2.18.121.0+9e31e41a",
43
+ "num_cores_per_group": 1,
44
+ "on_device_sampling": false,
45
+ "optimum_neuron_version": "0.2.0.dev8",
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-2.18.121.0+9e31e41a/0_REGISTRY/0.2.0.dev8/llama/llamafactory/tiny-random-Llama-3/df95ebb72707172e4141.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": "2.18.121.0+9e31e41a",
43
+ "num_cores_per_group": 1,
44
+ "on_device_sampling": true,
45
+ "optimum_neuron_version": "0.2.0.dev8",
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-2.18.121.0+9e31e41a/0_REGISTRY/0.2.0.dev8/llama/llamafactory/tiny-random-Llama-3/eb93770853f294e67a7b.json ADDED
@@ -0,0 +1,55 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "HloNeuronConfig",
20
+ "all_reduce_dtype": null,
21
+ "allow_flash_attention": true,
22
+ "attention_layout": "BSH",
23
+ "attn_output_transposed": false,
24
+ "auto_cast_type": "bf16",
25
+ "batch_size": 1,
26
+ "checkpoint_id": "llamafactory/tiny-random-Llama-3",
27
+ "checkpoint_revision": "bf2a2e3bf199ad2ee96f02a3c00246c608db22a8",
28
+ "collectives_layout": "HSB",
29
+ "continuous_batching": false,
30
+ "fuse_qkv": true,
31
+ "group_query_attention": null,
32
+ "log_softmax_scores": false,
33
+ "neuronxcc_version": "2.18.121.0+9e31e41a",
34
+ "optimum_neuron_version": "0.2.0.dev8",
35
+ "output_all_logits": false,
36
+ "sequence_length": 100,
37
+ "tp_degree": 2
38
+ },
39
+ "num_attention_heads": 4,
40
+ "num_hidden_layers": 2,
41
+ "num_key_value_heads": 4,
42
+ "pretraining_tp": 1,
43
+ "rms_norm_eps": 1e-05,
44
+ "rope_scaling": {
45
+ "factor": 8.0,
46
+ "high_freq_factor": 4.0,
47
+ "low_freq_factor": 1.0,
48
+ "original_max_position_embeddings": 8192,
49
+ "rope_type": "llama3"
50
+ },
51
+ "rope_theta": 500000.0,
52
+ "tie_word_embeddings": false,
53
+ "use_cache": true,
54
+ "vocab_size": 128256
55
+ }
neuronxcc-2.18.121.0+9e31e41a/0_REGISTRY/0.2.0.dev8/llama/unsloth/Llama-3.2-1B-Instruct/972a896364f30c9bc91a.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": "2.18.121.0+9e31e41a",
43
+ "num_cores_per_group": 1,
44
+ "on_device_sampling": false,
45
+ "optimum_neuron_version": "0.2.0.dev8",
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-2.18.121.0+9e31e41a/MODULE_152c4fbd51eb1f6f90ec+84f3e719/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-dge-dma --vectorize-strided-dma ", "-O2", "--internal-num-neuroncores-per-sengine=1", "--logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.18.121.0+9e31e41a/MODULE_152c4fbd51eb1f6f90ec+84f3e719/model.done ADDED
File without changes
neuronxcc-2.18.121.0+9e31e41a/MODULE_152c4fbd51eb1f6f90ec+84f3e719/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:940d63493c5e24e852e93fd211ef389b1227d8cdbda0e5f60d4deb32479db79b
3
+ size 80975
neuronxcc-2.18.121.0+9e31e41a/MODULE_152c4fbd51eb1f6f90ec+84f3e719/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:40a5b41908151f72aed02aeff1551d6172c274387a8a3a1769203344c9947102
3
+ size 226304
neuronxcc-2.18.121.0+9e31e41a/MODULE_21d49e164d26352245e5+84f3e719/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0090e72266856b033ebceba4fb07996b8adf223cbb1ccd95b9a110b4d718439d
3
  size 448722
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:862a0c3cbaa7b7892a69f499f79a983d7850007fb3f3a804332b3619497b002b
3
  size 448722
neuronxcc-2.18.121.0+9e31e41a/MODULE_21d49e164d26352245e5+84f3e719/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2a483dab1a41dc12ef27c79af0d8c5e5aa6509a3edd23b3d7ee09e76443dd49e
3
  size 32257024
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:92e447adc7b251e6a456c38bee619cded3e494eabe5ad1f47e44a3b2b103a877
3
  size 32257024
neuronxcc-2.18.121.0+9e31e41a/MODULE_24ff9ac2787ce9a1d276+613edded/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:86c30dbf951f5d15bdd34dcfb86020380523ef739f105813aaac0285d876c62f
3
  size 134144
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:45ab2bbb5224185ad9fbc969770f2d85563af272d116c01ade65fa7bbef2bab2
3
  size 134144
neuronxcc-2.18.121.0+9e31e41a/MODULE_320f2622d4d0c9fdd0f1+613edded/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:068b4f6f4c1613a75105e409abf759c46f57c0f0d64a92b74152c59fd1910ebb
3
  size 144384
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ff43ee8a129d30924ff9b3562bbd18130313581226e9677b542793d39037b07f
3
  size 144384
neuronxcc-2.18.121.0+9e31e41a/MODULE_331276a07386ee77d52e+431f5505/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ "--model-type=transformer -O1 --lnc=1 --internal-hlo2tensorizer-options=--experimental-unsafe-fp8e4m3fn-as-fp8e4m3 --logfile=/tmp/nxd_model/layout_opt/log-neuron-cc.txt"
neuronxcc-2.18.121.0+9e31e41a/MODULE_331276a07386ee77d52e+431f5505/model.done ADDED
File without changes
neuronxcc-2.18.121.0+9e31e41a/MODULE_331276a07386ee77d52e+431f5505/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f32d5db6cdf19c397941133ad17626fbb60335e60d3115b19046ba4d38e111f0
3
+ size 7106
neuronxcc-2.18.121.0+9e31e41a/MODULE_331276a07386ee77d52e+431f5505/model.neff ADDED
Binary file (42 kB). View file
 
neuronxcc-2.18.121.0+9e31e41a/MODULE_34a6b42796c8b4e2f58b+431f5505/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f6a8a8051d1ee442c0d6104a3d4a28cb465b0c7177888657e7c78e356356f4ef
3
  size 1158144
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e0d579312347183e39ad4ecc3e6ef0c730cda4cb4c6a01803cfe4d6e66774e16
3
  size 1158144
neuronxcc-2.18.121.0+9e31e41a/MODULE_40a0e75a65ac51fdd01a+613edded/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6fc8fe930d1a7fdc8ec51cb1e2b0ba151a8debdf8bbb1c84f2f4d178e0775bc2
3
  size 144384
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:257dd149284f429d33e419d6fea0298c8961b7ec25e04772dc082cacbc243760
3
  size 144384
neuronxcc-2.18.121.0+9e31e41a/MODULE_5368928916623911b1f2+84f3e719/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7b51a446849f966d6ee4b830fc99920ecc9859a68782b65ad45e9c6fc509ddbc
3
  size 47194
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5fa7bb6da1ff6c9a4d1024e95303dbffef7b4210925cc91757a7428c02bbff16
3
  size 47194
neuronxcc-2.18.121.0+9e31e41a/MODULE_5368928916623911b1f2+84f3e719/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1c480ea5d6af516f1f97d236e725d9b3147e57ce598c8060ff2a07e672b4f5c1
3
  size 246784
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:54f874a5471d91d2e953da4bfbe2a66a8574358f5707ad3d72a39e168730d866
3
  size 246784
neuronxcc-2.18.121.0+9e31e41a/MODULE_788eb0c6a9b0ca759eca+5be477de/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f90d1f38858638b69e5de8f1291e2f4ede577bac86dcd5ea181017993834edb3
3
  size 2386944
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:147853014224cbc7cf3e8b734afc60b9b7959c2f15d08ec0a2b2472bbfa75eaf
3
  size 2386944
neuronxcc-2.18.121.0+9e31e41a/MODULE_788eb0c6a9b0ca759eca+5be477de/wrapped_neff.hlo CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:189e509999514654134b86d2bb8db0c944b9f48c60670c1ea97184c15b42e419
3
  size 2394734
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dfbeae3bc8e69c9f4248160b6aa576552d407098cf4df09a8df262355c50cc0b
3
  size 2394734
neuronxcc-2.18.121.0+9e31e41a/MODULE_96404ade088398b2f3d5+431f5505/model.neff CHANGED
Binary files a/neuronxcc-2.18.121.0+9e31e41a/MODULE_96404ade088398b2f3d5+431f5505/model.neff and b/neuronxcc-2.18.121.0+9e31e41a/MODULE_96404ade088398b2f3d5+431f5505/model.neff differ
 
neuronxcc-2.18.121.0+9e31e41a/MODULE_a669c4bb07e1a9435b7c+431f5505/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ "--model-type=transformer -O1 --lnc=1 --internal-hlo2tensorizer-options=--experimental-unsafe-fp8e4m3fn-as-fp8e4m3 --logfile=/tmp/nxd_model/layout_opt/log-neuron-cc.txt"
neuronxcc-2.18.121.0+9e31e41a/MODULE_a669c4bb07e1a9435b7c+431f5505/model.done ADDED
File without changes
neuronxcc-2.18.121.0+9e31e41a/MODULE_a669c4bb07e1a9435b7c+431f5505/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:01d4291445f0fad2b3267c9fc11713f66b969c4b7def78292de1e0ef96a0c206
3
+ size 7106
neuronxcc-2.18.121.0+9e31e41a/MODULE_a669c4bb07e1a9435b7c+431f5505/model.neff ADDED
Binary file (42 kB). View file
 
neuronxcc-2.18.121.0+9e31e41a/MODULE_a77b880fafe6f97b3d8e+5be477de/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-dge-dma --vectorize-strided-dma ", "-O2", "--internal-num-neuroncores-per-sengine=1", "--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.18.121.0+9e31e41a/MODULE_a77b880fafe6f97b3d8e+5be477de/model.done ADDED
File without changes
neuronxcc-2.18.121.0+9e31e41a/MODULE_a77b880fafe6f97b3d8e+5be477de/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6bfc502457c8140618e43fa0f9fb04353838e80b331f8ade3e4532f88f525db5
3
+ size 80942
neuronxcc-2.18.121.0+9e31e41a/MODULE_a77b880fafe6f97b3d8e+5be477de/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1289cc53f0e3dac59961968ae68e43e2dae98441db1897ba313cb027ca05e1a7
3
+ size 236544
neuronxcc-2.18.121.0+9e31e41a/MODULE_a77b880fafe6f97b3d8e+5be477de/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:37fbef15e62dab940aa9a22718cdf5d74f79105c792b47ee410cccb4258571d6
3
+ size 244319
neuronxcc-2.18.121.0+9e31e41a/MODULE_b94c33a1002b2de0e6f0+5be477de/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:289cd7236f8fbddca6e02f3c8d9951ec5d65aea27b6ecc132c5762ec422659e0
3
  size 4445184
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c1903569ae78d37715e5f8d11ebc5b24addba75be3a234f0214183d8a07bc0dd
3
  size 4445184
neuronxcc-2.18.121.0+9e31e41a/MODULE_b94c33a1002b2de0e6f0+5be477de/wrapped_neff.hlo CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b5f4259b3ebb6cedd70a109e2fc61525ae8c75050c4e6f77eca71f03eec5e9be
3
  size 4514940
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9abb8316b56717215eda6d723ee785d74073d435832ffc9003ecfb6698cc08d4
3
  size 4514940
neuronxcc-2.18.121.0+9e31e41a/MODULE_ba6a6bbc03ee6dbd2e65+5be477de/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-dge-dma --vectorize-strided-dma ", "-O2", "--internal-num-neuroncores-per-sengine=1", "--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.18.121.0+9e31e41a/MODULE_ba6a6bbc03ee6dbd2e65+5be477de/model.done ADDED
File without changes
neuronxcc-2.18.121.0+9e31e41a/MODULE_ba6a6bbc03ee6dbd2e65+5be477de/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9c607d1e4a4cb0983a535f92bf7b97ff3fb7c2707ed725f9ad27d5f8c836131f
3
+ size 80942
neuronxcc-2.18.121.0+9e31e41a/MODULE_ba6a6bbc03ee6dbd2e65+5be477de/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:73d0a18efe648561975b69d38af7585eb0c0e4fa9b7e27e99a15202a0d7eff15
3
+ size 236544
neuronxcc-2.18.121.0+9e31e41a/MODULE_ba6a6bbc03ee6dbd2e65+5be477de/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c4f1c0d429d458ac177c4bd9b7200f93fbe8cb8b87670bfe361df7c91daf0b71
3
+ size 244319
neuronxcc-2.18.121.0+9e31e41a/MODULE_c49c0c3715f68c22b32f+613edded/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:bbf7c4a0ddf297c453d95bab5559254af19ee4d821e65eebee5d23f8309ec444
3
  size 144384
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cbf9f8d57a03f1bac0ae367b13f554dc9c7478fe47e5a972c8d6a67732635d0e
3
  size 144384
neuronxcc-2.18.121.0+9e31e41a/MODULE_df48af4bf01af7f3857e+613edded/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5cd94ff48573d55bdf90669a80680899501150cd062d2f02825ba50c894af47c
3
  size 144384
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:939153dcc0b4f1f803bd5737b5903aadbec23f1fba79df1ffe4c3207701a4747
3
  size 144384
neuronxcc-2.18.121.0+9e31e41a/MODULE_e3e99dc2a67feb5066f9+84f3e719/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-dge-dma --vectorize-strided-dma ", "-O2", "--internal-num-neuroncores-per-sengine=1", "--logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.18.121.0+9e31e41a/MODULE_e3e99dc2a67feb5066f9+84f3e719/model.done ADDED
File without changes
neuronxcc-2.18.121.0+9e31e41a/MODULE_e3e99dc2a67feb5066f9+84f3e719/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6908efe96eae9688ed6cf368a3011b2cf14d6d014cf6a9df424253966396ef5e
3
+ size 80975
neuronxcc-2.18.121.0+9e31e41a/MODULE_e3e99dc2a67feb5066f9+84f3e719/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:53b77dc392558fc18d230ef2b894c3a8b57cf844a4f2dcdb6379728b3d77ac8e
3
+ size 226304
neuronxcc-2.18.121.0+9e31e41a/MODULE_fda7fb53c475ba393ed7+613edded/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a83e7ca5aecc3279fb272e4b92d13bc2093090ed7f92d55d49a6ddc0513c6a48
3
  size 134144
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a5936747a55881bb7dc5c04f2802097806055b84f58411b8c413d0ad6ed19b28
3
  size 134144