Hugging Face's logo

optimum-internal-testing
/
neuron-testing-cache

Model card Files Community
dacorvo HF Staff commited on
Commit
2f6b9d4
·
verified ·
1 Parent(s): f578bfb

Synchronizing local compiler cache.

Browse files
Files changed (39) hide show
  1. .gitattributes +3 -0
  2. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/05c9cb13b974d4bfbb70.json +51 -0
  3. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/182f570e395a76ac2d2c.json +51 -0
  4. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/7def1dfbe4c4fd9f87fd.json +51 -0
  5. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0/llama/llamafactory/tiny-random-Llama-3/081c145e831213ab718b.json +77 -0
  6. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0/llama/llamafactory/tiny-random-Llama-3/087436195a7333857a9f.json +77 -0
  7. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0/llama/llamafactory/tiny-random-Llama-3/2308b99b68d27ecfdc41.json +77 -0
  8. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0/llama/llamafactory/tiny-random-Llama-3/2c27b201142fed60ee5b.json +55 -0
  9. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0/llama/llamafactory/tiny-random-Llama-3/304cd41ac7d9c2303bd3.json +55 -0
  10. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0/llama/llamafactory/tiny-random-Llama-3/7f5cf6b207ab7e9d6384.json +55 -0
  11. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0/llama/unsloth/Llama-3.2-1B-Instruct/5e144164d13014b24277.json +78 -0
  12. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0/mixtral/dacorvo/Mixtral-tiny/4e14b2bd4d311cf1902f.json +73 -0
  13. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0/mixtral/dacorvo/Mixtral-tiny/7d7a8f9e25ad161672e4.json +73 -0
  14. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0/mixtral/dacorvo/Mixtral-tiny/fa8340534c295e77e3b6.json +73 -0
  15. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0/phi3/yujiepan/phi-4-tiny-random/732f0a646f3fead04cc6.json +52 -0
  16. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0/phi3/yujiepan/phi-4-tiny-random/dd2dd68e4186f608717a.json +52 -0
  17. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0/phi3/yujiepan/phi-4-tiny-random/ebcb161fd1f8a2eaaf0a.json +52 -0
  18. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0/qwen2/yujiepan/qwen2.5-128k-tiny-random/7823ff8dffd276c75ce4.json +53 -0
  19. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0/qwen2/yujiepan/qwen2.5-128k-tiny-random/d8e67fd7c8f8e393dc42.json +53 -0
  20. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0/qwen2/yujiepan/qwen2.5-128k-tiny-random/f245d59e6049a5b7842a.json +53 -0
  21. neuronxcc-2.17.194.0+d312836f/MODULE_306e43064416feb5de44+165e9558/compile_flags.json +1 -0
  22. neuronxcc-2.17.194.0+d312836f/MODULE_306e43064416feb5de44+165e9558/model.done +0 -0
  23. neuronxcc-2.17.194.0+d312836f/MODULE_306e43064416feb5de44+165e9558/model.hlo_module.pb +3 -0
  24. neuronxcc-2.17.194.0+d312836f/MODULE_306e43064416feb5de44+165e9558/model.neff +3 -0
  25. neuronxcc-2.17.194.0+d312836f/MODULE_306e43064416feb5de44+165e9558/wrapped_neff.hlo +3 -0
  26. neuronxcc-2.17.194.0+d312836f/MODULE_3678b7fe58e0d712139a+431f5505/compile_flags.json +1 -0
  27. neuronxcc-2.17.194.0+d312836f/MODULE_3678b7fe58e0d712139a+431f5505/model.done +0 -0
  28. neuronxcc-2.17.194.0+d312836f/MODULE_3678b7fe58e0d712139a+431f5505/model.hlo_module.pb +3 -0
  29. neuronxcc-2.17.194.0+d312836f/MODULE_3678b7fe58e0d712139a+431f5505/model.neff +0 -0
  30. neuronxcc-2.17.194.0+d312836f/MODULE_3d29f316a0cedf338ff9+bfe5714b/compile_flags.json +1 -0
  31. neuronxcc-2.17.194.0+d312836f/MODULE_3d29f316a0cedf338ff9+bfe5714b/model.done +0 -0
  32. neuronxcc-2.17.194.0+d312836f/MODULE_3d29f316a0cedf338ff9+bfe5714b/model.hlo_module.pb +3 -0
  33. neuronxcc-2.17.194.0+d312836f/MODULE_3d29f316a0cedf338ff9+bfe5714b/model.neff +3 -0
  34. neuronxcc-2.17.194.0+d312836f/MODULE_6b226c5a27dcc0947566+bfc62e4c/model.hlo_module.pb +1 -1
  35. neuronxcc-2.17.194.0+d312836f/MODULE_6b226c5a27dcc0947566+bfc62e4c/model.neff +1 -1
  36. neuronxcc-2.17.194.0+d312836f/MODULE_6b226c5a27dcc0947566+bfc62e4c/wrapped_neff.hlo +1 -1
  37. neuronxcc-2.17.194.0+d312836f/MODULE_73a8fcccb40e156a3330+26ac6be0/model.hlo_module.pb +1 -1
  38. neuronxcc-2.17.194.0+d312836f/MODULE_73a8fcccb40e156a3330+26ac6be0/model.neff +1 -1
  39. neuronxcc-2.17.194.0+d312836f/MODULE_bfedca8ab047e39c175b+431f5505/model.neff +1 -1
.gitattributes CHANGED
@@ -2806,3 +2806,6 @@ neuronxcc-2.18.121.0+9e31e41a/MODULE_a77b880fafe6f97b3d8e+5be477de/wrapped_neff.
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
 
 
 
 
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
2809
+ neuronxcc-2.17.194.0+d312836f/MODULE_306e43064416feb5de44+165e9558/model.neff filter=lfs diff=lfs merge=lfs -text
2810
+ neuronxcc-2.17.194.0+d312836f/MODULE_306e43064416feb5de44+165e9558/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
2811
+ neuronxcc-2.17.194.0+d312836f/MODULE_3d29f316a0cedf338ff9+bfe5714b/model.neff filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/05c9cb13b974d4bfbb70.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "HloNeuronConfig",
22
+ "all_reduce_dtype": null,
23
+ "allow_flash_attention": true,
24
+ "attention_layout": "HSB",
25
+ "attn_output_transposed": false,
26
+ "auto_cast_type": "fp16",
27
+ "batch_size": 2,
28
+ "checkpoint_id": "hf-internal-testing/tiny-random-GraniteForCausalLM",
29
+ "checkpoint_revision": "c3074ebc0ac2fe545305f5e5f6cce2cc9b2aa0c5",
30
+ "collectives_layout": "HSB",
31
+ "continuous_batching": true,
32
+ "fuse_qkv": true,
33
+ "group_query_attention": null,
34
+ "log_softmax_scores": false,
35
+ "neuronxcc_version": "2.17.194.0+d312836f",
36
+ "optimum_neuron_version": "0.2.0",
37
+ "output_all_logits": false,
38
+ "sequence_length": 100,
39
+ "tp_degree": 2
40
+ },
41
+ "num_attention_heads": 4,
42
+ "num_hidden_layers": 2,
43
+ "num_key_value_heads": 4,
44
+ "residual_multiplier": 1.0,
45
+ "rms_norm_eps": 1e-06,
46
+ "rope_scaling": null,
47
+ "rope_theta": 10000.0,
48
+ "tie_word_embeddings": false,
49
+ "use_cache": true,
50
+ "vocab_size": 49152
51
+ }
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/182f570e395a76ac2d2c.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "HloNeuronConfig",
22
+ "all_reduce_dtype": null,
23
+ "allow_flash_attention": true,
24
+ "attention_layout": "HSB",
25
+ "attn_output_transposed": false,
26
+ "auto_cast_type": "bf16",
27
+ "batch_size": 1,
28
+ "checkpoint_id": "hf-internal-testing/tiny-random-GraniteForCausalLM",
29
+ "checkpoint_revision": "c3074ebc0ac2fe545305f5e5f6cce2cc9b2aa0c5",
30
+ "collectives_layout": "HSB",
31
+ "continuous_batching": false,
32
+ "fuse_qkv": true,
33
+ "group_query_attention": null,
34
+ "log_softmax_scores": false,
35
+ "neuronxcc_version": "2.17.194.0+d312836f",
36
+ "optimum_neuron_version": "0.2.0",
37
+ "output_all_logits": false,
38
+ "sequence_length": 100,
39
+ "tp_degree": 2
40
+ },
41
+ "num_attention_heads": 4,
42
+ "num_hidden_layers": 2,
43
+ "num_key_value_heads": 4,
44
+ "residual_multiplier": 1.0,
45
+ "rms_norm_eps": 1e-06,
46
+ "rope_scaling": null,
47
+ "rope_theta": 10000.0,
48
+ "tie_word_embeddings": false,
49
+ "use_cache": true,
50
+ "vocab_size": 49152
51
+ }
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/7def1dfbe4c4fd9f87fd.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "HloNeuronConfig",
22
+ "all_reduce_dtype": null,
23
+ "allow_flash_attention": true,
24
+ "attention_layout": "HSB",
25
+ "attn_output_transposed": false,
26
+ "auto_cast_type": "fp16",
27
+ "batch_size": 1,
28
+ "checkpoint_id": "hf-internal-testing/tiny-random-GraniteForCausalLM",
29
+ "checkpoint_revision": "c3074ebc0ac2fe545305f5e5f6cce2cc9b2aa0c5",
30
+ "collectives_layout": "HSB",
31
+ "continuous_batching": false,
32
+ "fuse_qkv": true,
33
+ "group_query_attention": null,
34
+ "log_softmax_scores": false,
35
+ "neuronxcc_version": "2.17.194.0+d312836f",
36
+ "optimum_neuron_version": "0.2.0",
37
+ "output_all_logits": false,
38
+ "sequence_length": 100,
39
+ "tp_degree": 2
40
+ },
41
+ "num_attention_heads": 4,
42
+ "num_hidden_layers": 2,
43
+ "num_key_value_heads": 4,
44
+ "residual_multiplier": 1.0,
45
+ "rms_norm_eps": 1e-06,
46
+ "rope_scaling": null,
47
+ "rope_theta": 10000.0,
48
+ "tie_word_embeddings": false,
49
+ "use_cache": true,
50
+ "vocab_size": 49152
51
+ }
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0/llama/llamafactory/tiny-random-Llama-3/081c145e831213ab718b.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": "0.2.0",
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": "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/0.2.0/llama/llamafactory/tiny-random-Llama-3/087436195a7333857a9f.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": "0.2.0",
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": "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/0.2.0/llama/llamafactory/tiny-random-Llama-3/2308b99b68d27ecfdc41.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": "0.2.0",
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": 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/0.2.0/llama/llamafactory/tiny-random-Llama-3/2c27b201142fed60ee5b.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": "2.17.194.0+d312836f",
34
+ "optimum_neuron_version": "0.2.0",
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-2.17.194.0+d312836f/0_REGISTRY/0.2.0/llama/llamafactory/tiny-random-Llama-3/304cd41ac7d9c2303bd3.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": "2.17.194.0+d312836f",
34
+ "optimum_neuron_version": "0.2.0",
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-2.17.194.0+d312836f/0_REGISTRY/0.2.0/llama/llamafactory/tiny-random-Llama-3/7f5cf6b207ab7e9d6384.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": "2.17.194.0+d312836f",
34
+ "optimum_neuron_version": "0.2.0",
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-2.17.194.0+d312836f/0_REGISTRY/0.2.0/llama/unsloth/Llama-3.2-1B-Instruct/5e144164d13014b24277.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": "0.2.0",
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": "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/0.2.0/mixtral/dacorvo/Mixtral-tiny/4e14b2bd4d311cf1902f.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": "0.2.0",
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/0.2.0/mixtral/dacorvo/Mixtral-tiny/7d7a8f9e25ad161672e4.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": "0.2.0",
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/0.2.0/mixtral/dacorvo/Mixtral-tiny/fa8340534c295e77e3b6.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": "0.2.0",
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/0.2.0/phi3/yujiepan/phi-4-tiny-random/732f0a646f3fead04cc6.json ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "HloNeuronConfig",
20
+ "all_reduce_dtype": null,
21
+ "allow_flash_attention": false,
22
+ "attention_layout": "HSB",
23
+ "attn_output_transposed": false,
24
+ "auto_cast_type": "bf16",
25
+ "batch_size": 1,
26
+ "checkpoint_id": "yujiepan/phi-4-tiny-random",
27
+ "checkpoint_revision": "18a9a1168dc97ac6d128f811925670c275610f5a",
28
+ "collectives_layout": "HSB",
29
+ "continuous_batching": false,
30
+ "fuse_qkv": true,
31
+ "group_query_attention": "replicated-heads",
32
+ "log_softmax_scores": false,
33
+ "neuronxcc_version": "2.17.194.0+d312836f",
34
+ "optimum_neuron_version": "0.2.0",
35
+ "output_all_logits": false,
36
+ "sequence_length": 100,
37
+ "tp_degree": 2
38
+ },
39
+ "num_attention_heads": 2,
40
+ "num_hidden_layers": 2,
41
+ "num_key_value_heads": 1,
42
+ "original_max_position_embeddings": 16384,
43
+ "partial_rotary_factor": 1.0,
44
+ "resid_pdrop": 0.0,
45
+ "rms_norm_eps": 1e-05,
46
+ "rope_scaling": null,
47
+ "rope_theta": 250000,
48
+ "sliding_window": null,
49
+ "tie_word_embeddings": false,
50
+ "use_cache": true,
51
+ "vocab_size": 100352
52
+ }
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0/phi3/yujiepan/phi-4-tiny-random/dd2dd68e4186f608717a.json ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "HloNeuronConfig",
20
+ "all_reduce_dtype": null,
21
+ "allow_flash_attention": false,
22
+ "attention_layout": "HSB",
23
+ "attn_output_transposed": false,
24
+ "auto_cast_type": "fp16",
25
+ "batch_size": 1,
26
+ "checkpoint_id": "yujiepan/phi-4-tiny-random",
27
+ "checkpoint_revision": "18a9a1168dc97ac6d128f811925670c275610f5a",
28
+ "collectives_layout": "HSB",
29
+ "continuous_batching": false,
30
+ "fuse_qkv": true,
31
+ "group_query_attention": "replicated-heads",
32
+ "log_softmax_scores": false,
33
+ "neuronxcc_version": "2.17.194.0+d312836f",
34
+ "optimum_neuron_version": "0.2.0",
35
+ "output_all_logits": false,
36
+ "sequence_length": 100,
37
+ "tp_degree": 2
38
+ },
39
+ "num_attention_heads": 2,
40
+ "num_hidden_layers": 2,
41
+ "num_key_value_heads": 1,
42
+ "original_max_position_embeddings": 16384,
43
+ "partial_rotary_factor": 1.0,
44
+ "resid_pdrop": 0.0,
45
+ "rms_norm_eps": 1e-05,
46
+ "rope_scaling": null,
47
+ "rope_theta": 250000,
48
+ "sliding_window": null,
49
+ "tie_word_embeddings": false,
50
+ "use_cache": true,
51
+ "vocab_size": 100352
52
+ }
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0/phi3/yujiepan/phi-4-tiny-random/ebcb161fd1f8a2eaaf0a.json ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "HloNeuronConfig",
20
+ "all_reduce_dtype": null,
21
+ "allow_flash_attention": false,
22
+ "attention_layout": "HSB",
23
+ "attn_output_transposed": false,
24
+ "auto_cast_type": "fp16",
25
+ "batch_size": 2,
26
+ "checkpoint_id": "yujiepan/phi-4-tiny-random",
27
+ "checkpoint_revision": "18a9a1168dc97ac6d128f811925670c275610f5a",
28
+ "collectives_layout": "HSB",
29
+ "continuous_batching": true,
30
+ "fuse_qkv": true,
31
+ "group_query_attention": "replicated-heads",
32
+ "log_softmax_scores": false,
33
+ "neuronxcc_version": "2.17.194.0+d312836f",
34
+ "optimum_neuron_version": "0.2.0",
35
+ "output_all_logits": false,
36
+ "sequence_length": 100,
37
+ "tp_degree": 2
38
+ },
39
+ "num_attention_heads": 2,
40
+ "num_hidden_layers": 2,
41
+ "num_key_value_heads": 1,
42
+ "original_max_position_embeddings": 16384,
43
+ "partial_rotary_factor": 1.0,
44
+ "resid_pdrop": 0.0,
45
+ "rms_norm_eps": 1e-05,
46
+ "rope_scaling": null,
47
+ "rope_theta": 250000,
48
+ "sliding_window": null,
49
+ "tie_word_embeddings": false,
50
+ "use_cache": true,
51
+ "vocab_size": 100352
52
+ }
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0/qwen2/yujiepan/qwen2.5-128k-tiny-random/7823ff8dffd276c75ce4.json ADDED
@@ -0,0 +1,53 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "HloNeuronConfig",
18
+ "all_reduce_dtype": null,
19
+ "allow_flash_attention": true,
20
+ "attention_layout": "HSB",
21
+ "attn_output_transposed": false,
22
+ "auto_cast_type": "fp16",
23
+ "batch_size": 1,
24
+ "checkpoint_id": "yujiepan/qwen2.5-128k-tiny-random",
25
+ "checkpoint_revision": "c8296d4ca3f87782876d2382fbb6481d1beb8ef0",
26
+ "collectives_layout": "HSB",
27
+ "continuous_batching": false,
28
+ "fuse_qkv": false,
29
+ "group_query_attention": "shard-over-heads",
30
+ "log_softmax_scores": false,
31
+ "neuronxcc_version": "2.17.194.0+d312836f",
32
+ "optimum_neuron_version": "0.2.0",
33
+ "output_all_logits": false,
34
+ "sequence_length": 100,
35
+ "tp_degree": 2
36
+ },
37
+ "num_attention_heads": 4,
38
+ "num_hidden_layers": 2,
39
+ "num_key_value_heads": 2,
40
+ "rms_norm_eps": 1e-06,
41
+ "rope_scaling": {
42
+ "factor": 4.0,
43
+ "original_max_position_embeddings": 32768,
44
+ "rope_type": "yarn",
45
+ "type": "yarn"
46
+ },
47
+ "rope_theta": 1000000.0,
48
+ "sliding_window": 131072,
49
+ "tie_word_embeddings": false,
50
+ "use_cache": true,
51
+ "use_sliding_window": false,
52
+ "vocab_size": 152064
53
+ }
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0/qwen2/yujiepan/qwen2.5-128k-tiny-random/d8e67fd7c8f8e393dc42.json ADDED
@@ -0,0 +1,53 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "HloNeuronConfig",
18
+ "all_reduce_dtype": null,
19
+ "allow_flash_attention": true,
20
+ "attention_layout": "HSB",
21
+ "attn_output_transposed": false,
22
+ "auto_cast_type": "fp16",
23
+ "batch_size": 2,
24
+ "checkpoint_id": "yujiepan/qwen2.5-128k-tiny-random",
25
+ "checkpoint_revision": "c8296d4ca3f87782876d2382fbb6481d1beb8ef0",
26
+ "collectives_layout": "HSB",
27
+ "continuous_batching": true,
28
+ "fuse_qkv": false,
29
+ "group_query_attention": "shard-over-heads",
30
+ "log_softmax_scores": false,
31
+ "neuronxcc_version": "2.17.194.0+d312836f",
32
+ "optimum_neuron_version": "0.2.0",
33
+ "output_all_logits": false,
34
+ "sequence_length": 100,
35
+ "tp_degree": 2
36
+ },
37
+ "num_attention_heads": 4,
38
+ "num_hidden_layers": 2,
39
+ "num_key_value_heads": 2,
40
+ "rms_norm_eps": 1e-06,
41
+ "rope_scaling": {
42
+ "factor": 4.0,
43
+ "original_max_position_embeddings": 32768,
44
+ "rope_type": "yarn",
45
+ "type": "yarn"
46
+ },
47
+ "rope_theta": 1000000.0,
48
+ "sliding_window": 131072,
49
+ "tie_word_embeddings": false,
50
+ "use_cache": true,
51
+ "use_sliding_window": false,
52
+ "vocab_size": 152064
53
+ }
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0/qwen2/yujiepan/qwen2.5-128k-tiny-random/f245d59e6049a5b7842a.json ADDED
@@ -0,0 +1,53 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "HloNeuronConfig",
18
+ "all_reduce_dtype": null,
19
+ "allow_flash_attention": true,
20
+ "attention_layout": "HSB",
21
+ "attn_output_transposed": false,
22
+ "auto_cast_type": "bf16",
23
+ "batch_size": 1,
24
+ "checkpoint_id": "yujiepan/qwen2.5-128k-tiny-random",
25
+ "checkpoint_revision": "c8296d4ca3f87782876d2382fbb6481d1beb8ef0",
26
+ "collectives_layout": "HSB",
27
+ "continuous_batching": false,
28
+ "fuse_qkv": false,
29
+ "group_query_attention": "shard-over-heads",
30
+ "log_softmax_scores": false,
31
+ "neuronxcc_version": "2.17.194.0+d312836f",
32
+ "optimum_neuron_version": "0.2.0",
33
+ "output_all_logits": false,
34
+ "sequence_length": 100,
35
+ "tp_degree": 2
36
+ },
37
+ "num_attention_heads": 4,
38
+ "num_hidden_layers": 2,
39
+ "num_key_value_heads": 2,
40
+ "rms_norm_eps": 1e-06,
41
+ "rope_scaling": {
42
+ "factor": 4.0,
43
+ "original_max_position_embeddings": 32768,
44
+ "rope_type": "yarn",
45
+ "type": "yarn"
46
+ },
47
+ "rope_theta": 1000000.0,
48
+ "sliding_window": 131072,
49
+ "tie_word_embeddings": false,
50
+ "use_cache": true,
51
+ "use_sliding_window": false,
52
+ "vocab_size": 152064
53
+ }
neuronxcc-2.17.194.0+d312836f/MODULE_306e43064416feb5de44+165e9558/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ "--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-2.17.194.0+d312836f/MODULE_306e43064416feb5de44+165e9558/model.done ADDED
File without changes
neuronxcc-2.17.194.0+d312836f/MODULE_306e43064416feb5de44+165e9558/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9cb48e74dbeaafb58771e45b6d0b36e41454443bf74bf0695899909b9dc170f1
3
+ size 86903
neuronxcc-2.17.194.0+d312836f/MODULE_306e43064416feb5de44+165e9558/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e1e68e361d8fb5e36eac16d440957c4f5e4bef7149005ab3900ae8fb702be76e
3
+ size 267264
neuronxcc-2.17.194.0+d312836f/MODULE_306e43064416feb5de44+165e9558/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:357a75efd755e85350572569c6c48811082802b6ee3759554db1c834f2c345c9
3
+ size 276419
neuronxcc-2.17.194.0+d312836f/MODULE_3678b7fe58e0d712139a+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-2.17.194.0+d312836f/MODULE_3678b7fe58e0d712139a+431f5505/model.done ADDED
File without changes
neuronxcc-2.17.194.0+d312836f/MODULE_3678b7fe58e0d712139a+431f5505/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ac926c6dd51d86cf6a76e668ff035fbbcedb1769803a4e4b4c802af52ec03de2
3
+ size 8648
neuronxcc-2.17.194.0+d312836f/MODULE_3678b7fe58e0d712139a+431f5505/model.neff ADDED
Binary file (42 kB). View file
 
neuronxcc-2.17.194.0+d312836f/MODULE_3d29f316a0cedf338ff9+bfe5714b/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ "--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-2.17.194.0+d312836f/MODULE_3d29f316a0cedf338ff9+bfe5714b/model.done ADDED
File without changes
neuronxcc-2.17.194.0+d312836f/MODULE_3d29f316a0cedf338ff9+bfe5714b/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8ffbf0bb377b4db400d4ac37a60449bf98f88f3abe14a23cc93b4dfbfb30c8a3
3
+ size 88689
neuronxcc-2.17.194.0+d312836f/MODULE_3d29f316a0cedf338ff9+bfe5714b/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fc1229629f4fe60e90fe49941317a7311410a6bf20a39522e45c41d59a5ba2a5
3
+ size 257024
neuronxcc-2.17.194.0+d312836f/MODULE_6b226c5a27dcc0947566+bfc62e4c/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9a1d4737b03880a59edac5cd564c0422b31b3c8ccbfad4955f43f95f58c6d124
3
  size 81486
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4000a55e4f5dbd1eca9566ff60f1bf9acb1bc00e35f71e7a22220e1b5e977684
3
  size 81486
neuronxcc-2.17.194.0+d312836f/MODULE_6b226c5a27dcc0947566+bfc62e4c/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f18468903a9ceeef62d4958073ae541496eb85c8ee3538f98edf5891e8cacff4
3
  size 318464
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fda983457f78de472af959d6a4efea929f84f621f853fb5e08ad31af5af251b3
3
  size 318464
neuronxcc-2.17.194.0+d312836f/MODULE_6b226c5a27dcc0947566+bfc62e4c/wrapped_neff.hlo CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b494ee982233e61acffe827b108989690eeb03b5f8005efe631e69e73b2efacc
3
  size 329762
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:331f4ffb1d6e3629bb5003aa1cd32632d430b5e262306d5b1cb1a87610bcd14e
3
  size 329762
neuronxcc-2.17.194.0+d312836f/MODULE_73a8fcccb40e156a3330+26ac6be0/model.hlo_module.pb CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cab792fcef1befda3768bc4e611e90d447b0aefbfaa0694e517974ec7e4ac274
3
  size 82168
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:99959d517b441937c4fe7e78cd58d5e7ab0cdb4c53b3e466bbfa1712ad58ff0a
3
  size 82168
neuronxcc-2.17.194.0+d312836f/MODULE_73a8fcccb40e156a3330+26ac6be0/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:973458cca0e15fa1a9194d279f46149198e89590ccfaf3c8a4d1f2446b649692
3
  size 420864
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ddffdeed53d210d00069356f58e0f87bbf716ad1d2b87cd9f149b0506071b80b
3
  size 420864
neuronxcc-2.17.194.0+d312836f/MODULE_bfedca8ab047e39c175b+431f5505/model.neff CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:01496855c4e7b176e667784ffdd3eabd3dca35b2da7c2cab39bb2f544093aa4a
3
  size 103424
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d2f8209490daedb50f0809528a0f4f4b512ee9b2a4d690d986a8cf30b1c22aad
3
  size 103424