Hugging Face's logo

optimum-internal-testing
/
neuron-testing-cache

Model card Files Community
optimum-internal-testing-user commited on
Commit
8e8946a
·
verified ·
1 Parent(s): 5d6f7cf

Synchronizing local compiler cache.

Browse files
Files changed (47) hide show
  1. .gitattributes +9 -0
  2. neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/0f0ac5938a0e8b8b473f.json +59 -0
  3. neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/4cdc2bc5149e677e4f2b.json +59 -0
  4. neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/feb78fb2c8a20f0a119c.json +59 -0
  5. neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/llama/llamafactory/tiny-random-Llama-3/6e02957b9795bfe1abda.json +63 -0
  6. neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/llama/llamafactory/tiny-random-Llama-3/848f201c15fc7bba1d8b.json +63 -0
  7. neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/llama/llamafactory/tiny-random-Llama-3/9c711dd85a2d4618b475.json +63 -0
  8. neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/mixtral/dacorvo/Mixtral-tiny/14adc923cf600e750ce2.json +59 -0
  9. neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/mixtral/dacorvo/Mixtral-tiny/6e078a302b1325238d89.json +59 -0
  10. neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/mixtral/dacorvo/Mixtral-tiny/8bd0bdaa409da65b0e34.json +59 -0
  11. neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/phi3/yujiepan/phi-4-tiny-random/41ce44fb47656404cc82.json +60 -0
  12. neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/phi3/yujiepan/phi-4-tiny-random/86fcc21d5c777fc845a7.json +60 -0
  13. neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/phi3/yujiepan/phi-4-tiny-random/e9b3555aae4ea1a390db.json +60 -0
  14. neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/qwen2/Qwen/Qwen2.5-0.5B/1001cb40169f602f416d.json +83 -0
  15. neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/qwen2/yujiepan/qwen2.5-128k-tiny-random/4efceebded25852079ca.json +65 -0
  16. neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/qwen2/yujiepan/qwen2.5-128k-tiny-random/6cc739f6dbdb1b0e1676.json +65 -0
  17. neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/qwen2/yujiepan/qwen2.5-128k-tiny-random/8a9598c30e93d592df6b.json +65 -0
  18. neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/qwen3_moe/optimum-internal-testing/tiny-random-qwen3_moe/1550081bb53b23759b89.json +66 -0
  19. neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/qwen3_moe/optimum-internal-testing/tiny-random-qwen3_moe/3a9ad8456bbc98d3a4b5.json +66 -0
  20. neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/qwen3_moe/optimum-internal-testing/tiny-random-qwen3_moe/9a119236f3406ad16638.json +66 -0
  21. neuronxcc-2.19.8089.0+8ab9f450/MODULE_147ae45766ca99a75121+253d6470/compile_flags.json +1 -0
  22. neuronxcc-2.19.8089.0+8ab9f450/MODULE_147ae45766ca99a75121+253d6470/model.done +0 -0
  23. neuronxcc-2.19.8089.0+8ab9f450/MODULE_147ae45766ca99a75121+253d6470/model.hlo_module.pb +3 -0
  24. neuronxcc-2.19.8089.0+8ab9f450/MODULE_147ae45766ca99a75121+253d6470/model.neff +3 -0
  25. neuronxcc-2.19.8089.0+8ab9f450/MODULE_7f8f6cb9ef9069b9a192+cd3419b6/compile_flags.json +1 -0
  26. neuronxcc-2.19.8089.0+8ab9f450/MODULE_7f8f6cb9ef9069b9a192+cd3419b6/model.done +0 -0
  27. neuronxcc-2.19.8089.0+8ab9f450/MODULE_7f8f6cb9ef9069b9a192+cd3419b6/model.hlo_module.pb +3 -0
  28. neuronxcc-2.19.8089.0+8ab9f450/MODULE_7f8f6cb9ef9069b9a192+cd3419b6/model.neff +3 -0
  29. neuronxcc-2.19.8089.0+8ab9f450/MODULE_7f8f6cb9ef9069b9a192+cd3419b6/wrapped_neff.hlo +3 -0
  30. neuronxcc-2.19.8089.0+8ab9f450/MODULE_afd876d1d56e17288e52+253d6470/compile_flags.json +1 -0
  31. neuronxcc-2.19.8089.0+8ab9f450/MODULE_afd876d1d56e17288e52+253d6470/model.done +0 -0
  32. neuronxcc-2.19.8089.0+8ab9f450/MODULE_afd876d1d56e17288e52+253d6470/model.hlo_module.pb +3 -0
  33. neuronxcc-2.19.8089.0+8ab9f450/MODULE_afd876d1d56e17288e52+253d6470/model.neff +3 -0
  34. neuronxcc-2.19.8089.0+8ab9f450/MODULE_bb03c2c3bd153e48eeea+cd3419b6/compile_flags.json +1 -0
  35. neuronxcc-2.19.8089.0+8ab9f450/MODULE_bb03c2c3bd153e48eeea+cd3419b6/model.done +0 -0
  36. neuronxcc-2.19.8089.0+8ab9f450/MODULE_bb03c2c3bd153e48eeea+cd3419b6/model.hlo_module.pb +3 -0
  37. neuronxcc-2.19.8089.0+8ab9f450/MODULE_bb03c2c3bd153e48eeea+cd3419b6/model.neff +3 -0
  38. neuronxcc-2.19.8089.0+8ab9f450/MODULE_bb03c2c3bd153e48eeea+cd3419b6/wrapped_neff.hlo +3 -0
  39. neuronxcc-2.19.8089.0+8ab9f450/MODULE_bb945fbdfaf46f441eed+cd3419b6/compile_flags.json +1 -0
  40. neuronxcc-2.19.8089.0+8ab9f450/MODULE_bb945fbdfaf46f441eed+cd3419b6/model.done +0 -0
  41. neuronxcc-2.19.8089.0+8ab9f450/MODULE_bb945fbdfaf46f441eed+cd3419b6/model.hlo_module.pb +3 -0
  42. neuronxcc-2.19.8089.0+8ab9f450/MODULE_bb945fbdfaf46f441eed+cd3419b6/model.neff +3 -0
  43. neuronxcc-2.19.8089.0+8ab9f450/MODULE_bb945fbdfaf46f441eed+cd3419b6/wrapped_neff.hlo +3 -0
  44. neuronxcc-2.19.8089.0+8ab9f450/MODULE_ff4faa8edb2d2eec282d+253d6470/compile_flags.json +1 -0
  45. neuronxcc-2.19.8089.0+8ab9f450/MODULE_ff4faa8edb2d2eec282d+253d6470/model.done +0 -0
  46. neuronxcc-2.19.8089.0+8ab9f450/MODULE_ff4faa8edb2d2eec282d+253d6470/model.hlo_module.pb +3 -0
  47. neuronxcc-2.19.8089.0+8ab9f450/MODULE_ff4faa8edb2d2eec282d+253d6470/model.neff +3 -0
.gitattributes CHANGED
@@ -4570,3 +4570,12 @@ neuronxcc-2.19.8089.0+8ab9f450/MODULE_d379451a3eeeb9a21ff6+a9d440f5/model.neff f
4570
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_d379451a3eeeb9a21ff6+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
4571
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_d7030396f798ba07e1a9+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
4572
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_d7030396f798ba07e1a9+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
 
 
 
4570
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_d379451a3eeeb9a21ff6+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
4571
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_d7030396f798ba07e1a9+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
4572
  neuronxcc-2.19.8089.0+8ab9f450/MODULE_d7030396f798ba07e1a9+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
4573
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_147ae45766ca99a75121+253d6470/model.neff filter=lfs diff=lfs merge=lfs -text
4574
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_7f8f6cb9ef9069b9a192+cd3419b6/model.neff filter=lfs diff=lfs merge=lfs -text
4575
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_7f8f6cb9ef9069b9a192+cd3419b6/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
4576
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_afd876d1d56e17288e52+253d6470/model.neff filter=lfs diff=lfs merge=lfs -text
4577
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_bb03c2c3bd153e48eeea+cd3419b6/model.neff filter=lfs diff=lfs merge=lfs -text
4578
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_bb03c2c3bd153e48eeea+cd3419b6/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
4579
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_bb945fbdfaf46f441eed+cd3419b6/model.neff filter=lfs diff=lfs merge=lfs -text
4580
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_bb945fbdfaf46f441eed+cd3419b6/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
4581
+ neuronxcc-2.19.8089.0+8ab9f450/MODULE_ff4faa8edb2d2eec282d+253d6470/model.neff filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/0f0ac5938a0e8b8b473f.json ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "hf-internal-testing/tiny-random-GraniteForCausalLM",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "GraniteForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "attention_multiplier": 1.0,
11
+ "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
+ "batch_size": 1,
23
+ "capacity_factor": null,
24
+ "checkpoint_id": "hf-internal-testing/tiny-random-GraniteForCausalLM",
25
+ "checkpoint_revision": "c3074ebc0ac2fe545305f5e5f6cce2cc9b2aa0c5",
26
+ "continuous_batching": false,
27
+ "enable_bucketing": false,
28
+ "ep_degree": 1,
29
+ "fused_qkv": true,
30
+ "glu_mlp": true,
31
+ "local_ranks_size": 2,
32
+ "logical_nc_config": 1,
33
+ "max_batch_size": 1,
34
+ "max_context_length": 100,
35
+ "max_topk": 256,
36
+ "n_active_tokens": 100,
37
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
38
+ "on_device_sampling": true,
39
+ "optimum_neuron_version": "0.3.1.dev5",
40
+ "output_logits": false,
41
+ "pp_degree": 1,
42
+ "sequence_length": 100,
43
+ "speculation_length": 0,
44
+ "start_rank_id": 0,
45
+ "target": null,
46
+ "torch_dtype": "float16",
47
+ "tp_degree": 2
48
+ },
49
+ "num_attention_heads": 4,
50
+ "num_hidden_layers": 2,
51
+ "num_key_value_heads": 4,
52
+ "residual_multiplier": 1.0,
53
+ "rms_norm_eps": 1e-06,
54
+ "rope_scaling": null,
55
+ "rope_theta": 10000.0,
56
+ "tie_word_embeddings": false,
57
+ "use_cache": true,
58
+ "vocab_size": 49152
59
+ }
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/4cdc2bc5149e677e4f2b.json ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "hf-internal-testing/tiny-random-GraniteForCausalLM",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "GraniteForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "attention_multiplier": 1.0,
11
+ "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
+ "batch_size": 1,
23
+ "capacity_factor": null,
24
+ "checkpoint_id": "hf-internal-testing/tiny-random-GraniteForCausalLM",
25
+ "checkpoint_revision": "c3074ebc0ac2fe545305f5e5f6cce2cc9b2aa0c5",
26
+ "continuous_batching": false,
27
+ "enable_bucketing": false,
28
+ "ep_degree": 1,
29
+ "fused_qkv": true,
30
+ "glu_mlp": true,
31
+ "local_ranks_size": 2,
32
+ "logical_nc_config": 1,
33
+ "max_batch_size": 1,
34
+ "max_context_length": 100,
35
+ "max_topk": 256,
36
+ "n_active_tokens": 100,
37
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
38
+ "on_device_sampling": true,
39
+ "optimum_neuron_version": "0.3.1.dev5",
40
+ "output_logits": false,
41
+ "pp_degree": 1,
42
+ "sequence_length": 100,
43
+ "speculation_length": 0,
44
+ "start_rank_id": 0,
45
+ "target": null,
46
+ "torch_dtype": "bfloat16",
47
+ "tp_degree": 2
48
+ },
49
+ "num_attention_heads": 4,
50
+ "num_hidden_layers": 2,
51
+ "num_key_value_heads": 4,
52
+ "residual_multiplier": 1.0,
53
+ "rms_norm_eps": 1e-06,
54
+ "rope_scaling": null,
55
+ "rope_theta": 10000.0,
56
+ "tie_word_embeddings": false,
57
+ "use_cache": true,
58
+ "vocab_size": 49152
59
+ }
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/granite/hf-internal-testing/tiny-random-GraniteForCausalLM/feb78fb2c8a20f0a119c.json ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "hf-internal-testing/tiny-random-GraniteForCausalLM",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "GraniteForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "attention_multiplier": 1.0,
11
+ "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
+ "batch_size": 2,
23
+ "capacity_factor": null,
24
+ "checkpoint_id": "hf-internal-testing/tiny-random-GraniteForCausalLM",
25
+ "checkpoint_revision": "c3074ebc0ac2fe545305f5e5f6cce2cc9b2aa0c5",
26
+ "continuous_batching": true,
27
+ "enable_bucketing": false,
28
+ "ep_degree": 1,
29
+ "fused_qkv": true,
30
+ "glu_mlp": true,
31
+ "local_ranks_size": 2,
32
+ "logical_nc_config": 1,
33
+ "max_batch_size": 2,
34
+ "max_context_length": 100,
35
+ "max_topk": 256,
36
+ "n_active_tokens": 100,
37
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
38
+ "on_device_sampling": true,
39
+ "optimum_neuron_version": "0.3.1.dev5",
40
+ "output_logits": false,
41
+ "pp_degree": 1,
42
+ "sequence_length": 100,
43
+ "speculation_length": 0,
44
+ "start_rank_id": 0,
45
+ "target": null,
46
+ "torch_dtype": "float16",
47
+ "tp_degree": 2
48
+ },
49
+ "num_attention_heads": 4,
50
+ "num_hidden_layers": 2,
51
+ "num_key_value_heads": 4,
52
+ "residual_multiplier": 1.0,
53
+ "rms_norm_eps": 1e-06,
54
+ "rope_scaling": null,
55
+ "rope_theta": 10000.0,
56
+ "tie_word_embeddings": false,
57
+ "use_cache": true,
58
+ "vocab_size": 49152
59
+ }
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/llama/llamafactory/tiny-random-Llama-3/6e02957b9795bfe1abda.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "llamafactory/tiny-random-Llama-3",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "LlamaForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "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
+ "batch_size": 1,
21
+ "capacity_factor": null,
22
+ "checkpoint_id": "llamafactory/tiny-random-Llama-3",
23
+ "checkpoint_revision": "bf2a2e3bf199ad2ee96f02a3c00246c608db22a8",
24
+ "continuous_batching": false,
25
+ "enable_bucketing": false,
26
+ "ep_degree": 1,
27
+ "fused_qkv": true,
28
+ "glu_mlp": true,
29
+ "local_ranks_size": 2,
30
+ "logical_nc_config": 1,
31
+ "max_batch_size": 1,
32
+ "max_context_length": 100,
33
+ "max_topk": 256,
34
+ "n_active_tokens": 100,
35
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
36
+ "on_device_sampling": true,
37
+ "optimum_neuron_version": "0.3.1.dev5",
38
+ "output_logits": false,
39
+ "pp_degree": 1,
40
+ "sequence_length": 100,
41
+ "speculation_length": 0,
42
+ "start_rank_id": 0,
43
+ "target": null,
44
+ "torch_dtype": "float16",
45
+ "tp_degree": 2
46
+ },
47
+ "num_attention_heads": 4,
48
+ "num_hidden_layers": 2,
49
+ "num_key_value_heads": 4,
50
+ "pretraining_tp": 1,
51
+ "rms_norm_eps": 1e-05,
52
+ "rope_scaling": {
53
+ "factor": 8.0,
54
+ "high_freq_factor": 4.0,
55
+ "low_freq_factor": 1.0,
56
+ "original_max_position_embeddings": 8192,
57
+ "rope_type": "llama3"
58
+ },
59
+ "rope_theta": 500000.0,
60
+ "tie_word_embeddings": false,
61
+ "use_cache": true,
62
+ "vocab_size": 128256
63
+ }
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/llama/llamafactory/tiny-random-Llama-3/848f201c15fc7bba1d8b.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "llamafactory/tiny-random-Llama-3",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "LlamaForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "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
+ "batch_size": 1,
21
+ "capacity_factor": null,
22
+ "checkpoint_id": "llamafactory/tiny-random-Llama-3",
23
+ "checkpoint_revision": "bf2a2e3bf199ad2ee96f02a3c00246c608db22a8",
24
+ "continuous_batching": false,
25
+ "enable_bucketing": false,
26
+ "ep_degree": 1,
27
+ "fused_qkv": true,
28
+ "glu_mlp": true,
29
+ "local_ranks_size": 2,
30
+ "logical_nc_config": 1,
31
+ "max_batch_size": 1,
32
+ "max_context_length": 100,
33
+ "max_topk": 256,
34
+ "n_active_tokens": 100,
35
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
36
+ "on_device_sampling": true,
37
+ "optimum_neuron_version": "0.3.1.dev5",
38
+ "output_logits": false,
39
+ "pp_degree": 1,
40
+ "sequence_length": 100,
41
+ "speculation_length": 0,
42
+ "start_rank_id": 0,
43
+ "target": null,
44
+ "torch_dtype": "bfloat16",
45
+ "tp_degree": 2
46
+ },
47
+ "num_attention_heads": 4,
48
+ "num_hidden_layers": 2,
49
+ "num_key_value_heads": 4,
50
+ "pretraining_tp": 1,
51
+ "rms_norm_eps": 1e-05,
52
+ "rope_scaling": {
53
+ "factor": 8.0,
54
+ "high_freq_factor": 4.0,
55
+ "low_freq_factor": 1.0,
56
+ "original_max_position_embeddings": 8192,
57
+ "rope_type": "llama3"
58
+ },
59
+ "rope_theta": 500000.0,
60
+ "tie_word_embeddings": false,
61
+ "use_cache": true,
62
+ "vocab_size": 128256
63
+ }
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/llama/llamafactory/tiny-random-Llama-3/9c711dd85a2d4618b475.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "llamafactory/tiny-random-Llama-3",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "LlamaForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "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
+ "batch_size": 2,
21
+ "capacity_factor": null,
22
+ "checkpoint_id": "llamafactory/tiny-random-Llama-3",
23
+ "checkpoint_revision": "bf2a2e3bf199ad2ee96f02a3c00246c608db22a8",
24
+ "continuous_batching": true,
25
+ "enable_bucketing": false,
26
+ "ep_degree": 1,
27
+ "fused_qkv": true,
28
+ "glu_mlp": true,
29
+ "local_ranks_size": 2,
30
+ "logical_nc_config": 1,
31
+ "max_batch_size": 2,
32
+ "max_context_length": 100,
33
+ "max_topk": 256,
34
+ "n_active_tokens": 100,
35
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
36
+ "on_device_sampling": true,
37
+ "optimum_neuron_version": "0.3.1.dev5",
38
+ "output_logits": false,
39
+ "pp_degree": 1,
40
+ "sequence_length": 100,
41
+ "speculation_length": 0,
42
+ "start_rank_id": 0,
43
+ "target": null,
44
+ "torch_dtype": "float16",
45
+ "tp_degree": 2
46
+ },
47
+ "num_attention_heads": 4,
48
+ "num_hidden_layers": 2,
49
+ "num_key_value_heads": 4,
50
+ "pretraining_tp": 1,
51
+ "rms_norm_eps": 1e-05,
52
+ "rope_scaling": {
53
+ "factor": 8.0,
54
+ "high_freq_factor": 4.0,
55
+ "low_freq_factor": 1.0,
56
+ "original_max_position_embeddings": 8192,
57
+ "rope_type": "llama3"
58
+ },
59
+ "rope_theta": 500000.0,
60
+ "tie_word_embeddings": false,
61
+ "use_cache": true,
62
+ "vocab_size": 128256
63
+ }
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/mixtral/dacorvo/Mixtral-tiny/14adc923cf600e750ce2.json ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "dacorvo/Mixtral-tiny",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "MixtralForCausalLM"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "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
+ "batch_size": 1,
19
+ "capacity_factor": null,
20
+ "checkpoint_id": "dacorvo/Mixtral-tiny",
21
+ "checkpoint_revision": "c557ba205ddff6ea911f4719e0d543d6c08356b6",
22
+ "continuous_batching": false,
23
+ "enable_bucketing": false,
24
+ "ep_degree": 1,
25
+ "fused_qkv": false,
26
+ "glu_mlp": true,
27
+ "local_ranks_size": 2,
28
+ "logical_nc_config": 1,
29
+ "max_batch_size": 1,
30
+ "max_context_length": 100,
31
+ "max_topk": 256,
32
+ "n_active_tokens": 100,
33
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
34
+ "on_device_sampling": false,
35
+ "optimum_neuron_version": "0.3.1.dev5",
36
+ "output_logits": false,
37
+ "pp_degree": 1,
38
+ "sequence_length": 100,
39
+ "speculation_length": 0,
40
+ "start_rank_id": 0,
41
+ "target": null,
42
+ "torch_dtype": "bfloat16",
43
+ "tp_degree": 2
44
+ },
45
+ "num_attention_heads": 32,
46
+ "num_experts_per_tok": 2,
47
+ "num_hidden_layers": 2,
48
+ "num_key_value_heads": 8,
49
+ "num_local_experts": 8,
50
+ "output_router_logits": false,
51
+ "rms_norm_eps": 1e-05,
52
+ "rope_theta": 10000.0,
53
+ "router_aux_loss_coef": 0.001,
54
+ "router_jitter_noise": 0.0,
55
+ "sliding_window": 4096,
56
+ "tie_word_embeddings": false,
57
+ "use_cache": true,
58
+ "vocab_size": 32000
59
+ }
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/mixtral/dacorvo/Mixtral-tiny/6e078a302b1325238d89.json ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "dacorvo/Mixtral-tiny",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "MixtralForCausalLM"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "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
+ "batch_size": 2,
19
+ "capacity_factor": null,
20
+ "checkpoint_id": "dacorvo/Mixtral-tiny",
21
+ "checkpoint_revision": "c557ba205ddff6ea911f4719e0d543d6c08356b6",
22
+ "continuous_batching": false,
23
+ "enable_bucketing": false,
24
+ "ep_degree": 1,
25
+ "fused_qkv": false,
26
+ "glu_mlp": true,
27
+ "local_ranks_size": 2,
28
+ "logical_nc_config": 1,
29
+ "max_batch_size": 2,
30
+ "max_context_length": 100,
31
+ "max_topk": 256,
32
+ "n_active_tokens": 100,
33
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
34
+ "on_device_sampling": false,
35
+ "optimum_neuron_version": "0.3.1.dev5",
36
+ "output_logits": false,
37
+ "pp_degree": 1,
38
+ "sequence_length": 100,
39
+ "speculation_length": 0,
40
+ "start_rank_id": 0,
41
+ "target": null,
42
+ "torch_dtype": "float16",
43
+ "tp_degree": 2
44
+ },
45
+ "num_attention_heads": 32,
46
+ "num_experts_per_tok": 2,
47
+ "num_hidden_layers": 2,
48
+ "num_key_value_heads": 8,
49
+ "num_local_experts": 8,
50
+ "output_router_logits": false,
51
+ "rms_norm_eps": 1e-05,
52
+ "rope_theta": 10000.0,
53
+ "router_aux_loss_coef": 0.001,
54
+ "router_jitter_noise": 0.0,
55
+ "sliding_window": 4096,
56
+ "tie_word_embeddings": false,
57
+ "use_cache": true,
58
+ "vocab_size": 32000
59
+ }
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/mixtral/dacorvo/Mixtral-tiny/8bd0bdaa409da65b0e34.json ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "dacorvo/Mixtral-tiny",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "MixtralForCausalLM"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "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
+ "batch_size": 1,
19
+ "capacity_factor": null,
20
+ "checkpoint_id": "dacorvo/Mixtral-tiny",
21
+ "checkpoint_revision": "c557ba205ddff6ea911f4719e0d543d6c08356b6",
22
+ "continuous_batching": false,
23
+ "enable_bucketing": false,
24
+ "ep_degree": 1,
25
+ "fused_qkv": false,
26
+ "glu_mlp": true,
27
+ "local_ranks_size": 2,
28
+ "logical_nc_config": 1,
29
+ "max_batch_size": 1,
30
+ "max_context_length": 100,
31
+ "max_topk": 256,
32
+ "n_active_tokens": 100,
33
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
34
+ "on_device_sampling": false,
35
+ "optimum_neuron_version": "0.3.1.dev5",
36
+ "output_logits": false,
37
+ "pp_degree": 1,
38
+ "sequence_length": 100,
39
+ "speculation_length": 0,
40
+ "start_rank_id": 0,
41
+ "target": null,
42
+ "torch_dtype": "float16",
43
+ "tp_degree": 2
44
+ },
45
+ "num_attention_heads": 32,
46
+ "num_experts_per_tok": 2,
47
+ "num_hidden_layers": 2,
48
+ "num_key_value_heads": 8,
49
+ "num_local_experts": 8,
50
+ "output_router_logits": false,
51
+ "rms_norm_eps": 1e-05,
52
+ "rope_theta": 10000.0,
53
+ "router_aux_loss_coef": 0.001,
54
+ "router_jitter_noise": 0.0,
55
+ "sliding_window": 4096,
56
+ "tie_word_embeddings": false,
57
+ "use_cache": true,
58
+ "vocab_size": 32000
59
+ }
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/phi3/yujiepan/phi-4-tiny-random/41ce44fb47656404cc82.json ADDED
@@ -0,0 +1,60 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "yujiepan/phi-4-tiny-random",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Phi3ForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "auto_map": {},
11
+ "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
+ "batch_size": 2,
21
+ "capacity_factor": null,
22
+ "checkpoint_id": "yujiepan/phi-4-tiny-random",
23
+ "checkpoint_revision": "18a9a1168dc97ac6d128f811925670c275610f5a",
24
+ "continuous_batching": true,
25
+ "enable_bucketing": false,
26
+ "ep_degree": 1,
27
+ "fused_qkv": true,
28
+ "glu_mlp": true,
29
+ "local_ranks_size": 2,
30
+ "logical_nc_config": 1,
31
+ "max_batch_size": 2,
32
+ "max_context_length": 100,
33
+ "max_topk": 256,
34
+ "n_active_tokens": 100,
35
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
36
+ "on_device_sampling": true,
37
+ "optimum_neuron_version": "0.3.1.dev5",
38
+ "output_logits": false,
39
+ "pp_degree": 1,
40
+ "sequence_length": 100,
41
+ "speculation_length": 0,
42
+ "start_rank_id": 0,
43
+ "target": null,
44
+ "torch_dtype": "float16",
45
+ "tp_degree": 2
46
+ },
47
+ "num_attention_heads": 2,
48
+ "num_hidden_layers": 2,
49
+ "num_key_value_heads": 1,
50
+ "original_max_position_embeddings": 16384,
51
+ "partial_rotary_factor": 1.0,
52
+ "resid_pdrop": 0.0,
53
+ "rms_norm_eps": 1e-05,
54
+ "rope_scaling": null,
55
+ "rope_theta": 250000,
56
+ "sliding_window": null,
57
+ "tie_word_embeddings": false,
58
+ "use_cache": true,
59
+ "vocab_size": 100352
60
+ }
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/phi3/yujiepan/phi-4-tiny-random/86fcc21d5c777fc845a7.json ADDED
@@ -0,0 +1,60 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "yujiepan/phi-4-tiny-random",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Phi3ForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "auto_map": {},
11
+ "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
+ "batch_size": 1,
21
+ "capacity_factor": null,
22
+ "checkpoint_id": "yujiepan/phi-4-tiny-random",
23
+ "checkpoint_revision": "18a9a1168dc97ac6d128f811925670c275610f5a",
24
+ "continuous_batching": false,
25
+ "enable_bucketing": false,
26
+ "ep_degree": 1,
27
+ "fused_qkv": true,
28
+ "glu_mlp": true,
29
+ "local_ranks_size": 2,
30
+ "logical_nc_config": 1,
31
+ "max_batch_size": 1,
32
+ "max_context_length": 100,
33
+ "max_topk": 256,
34
+ "n_active_tokens": 100,
35
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
36
+ "on_device_sampling": true,
37
+ "optimum_neuron_version": "0.3.1.dev5",
38
+ "output_logits": false,
39
+ "pp_degree": 1,
40
+ "sequence_length": 100,
41
+ "speculation_length": 0,
42
+ "start_rank_id": 0,
43
+ "target": null,
44
+ "torch_dtype": "float16",
45
+ "tp_degree": 2
46
+ },
47
+ "num_attention_heads": 2,
48
+ "num_hidden_layers": 2,
49
+ "num_key_value_heads": 1,
50
+ "original_max_position_embeddings": 16384,
51
+ "partial_rotary_factor": 1.0,
52
+ "resid_pdrop": 0.0,
53
+ "rms_norm_eps": 1e-05,
54
+ "rope_scaling": null,
55
+ "rope_theta": 250000,
56
+ "sliding_window": null,
57
+ "tie_word_embeddings": false,
58
+ "use_cache": true,
59
+ "vocab_size": 100352
60
+ }
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/phi3/yujiepan/phi-4-tiny-random/e9b3555aae4ea1a390db.json ADDED
@@ -0,0 +1,60 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "yujiepan/phi-4-tiny-random",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Phi3ForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "auto_map": {},
11
+ "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
+ "batch_size": 1,
21
+ "capacity_factor": null,
22
+ "checkpoint_id": "yujiepan/phi-4-tiny-random",
23
+ "checkpoint_revision": "18a9a1168dc97ac6d128f811925670c275610f5a",
24
+ "continuous_batching": false,
25
+ "enable_bucketing": false,
26
+ "ep_degree": 1,
27
+ "fused_qkv": true,
28
+ "glu_mlp": true,
29
+ "local_ranks_size": 2,
30
+ "logical_nc_config": 1,
31
+ "max_batch_size": 1,
32
+ "max_context_length": 100,
33
+ "max_topk": 256,
34
+ "n_active_tokens": 100,
35
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
36
+ "on_device_sampling": true,
37
+ "optimum_neuron_version": "0.3.1.dev5",
38
+ "output_logits": false,
39
+ "pp_degree": 1,
40
+ "sequence_length": 100,
41
+ "speculation_length": 0,
42
+ "start_rank_id": 0,
43
+ "target": null,
44
+ "torch_dtype": "bfloat16",
45
+ "tp_degree": 2
46
+ },
47
+ "num_attention_heads": 2,
48
+ "num_hidden_layers": 2,
49
+ "num_key_value_heads": 1,
50
+ "original_max_position_embeddings": 16384,
51
+ "partial_rotary_factor": 1.0,
52
+ "resid_pdrop": 0.0,
53
+ "rms_norm_eps": 1e-05,
54
+ "rope_scaling": null,
55
+ "rope_theta": 250000,
56
+ "sliding_window": null,
57
+ "tie_word_embeddings": false,
58
+ "use_cache": true,
59
+ "vocab_size": 100352
60
+ }
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/qwen2/Qwen/Qwen2.5-0.5B/1001cb40169f602f416d.json ADDED
@@ -0,0 +1,83 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "Qwen/Qwen2.5-0.5B",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Qwen2ForCausalLM"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 896,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 4864,
13
+ "layer_types": [
14
+ "full_attention",
15
+ "full_attention",
16
+ "full_attention",
17
+ "full_attention",
18
+ "full_attention",
19
+ "full_attention",
20
+ "full_attention",
21
+ "full_attention",
22
+ "full_attention",
23
+ "full_attention",
24
+ "full_attention",
25
+ "full_attention",
26
+ "full_attention",
27
+ "full_attention",
28
+ "full_attention",
29
+ "full_attention",
30
+ "full_attention",
31
+ "full_attention",
32
+ "full_attention",
33
+ "full_attention",
34
+ "full_attention",
35
+ "full_attention",
36
+ "full_attention",
37
+ "full_attention"
38
+ ],
39
+ "max_position_embeddings": 32768,
40
+ "max_window_layers": 24,
41
+ "model_type": "qwen2",
42
+ "neuron": {
43
+ "_serialized_key": "NxDNeuronConfig",
44
+ "batch_size": 4,
45
+ "capacity_factor": null,
46
+ "checkpoint_id": "Qwen/Qwen2.5-0.5B",
47
+ "checkpoint_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
48
+ "continuous_batching": true,
49
+ "enable_bucketing": false,
50
+ "ep_degree": 1,
51
+ "fused_qkv": false,
52
+ "glu_mlp": true,
53
+ "local_ranks_size": 2,
54
+ "logical_nc_config": 1,
55
+ "max_batch_size": 4,
56
+ "max_context_length": 4096,
57
+ "max_topk": 256,
58
+ "n_active_tokens": 4096,
59
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
60
+ "on_device_sampling": false,
61
+ "optimum_neuron_version": "0.3.1.dev5",
62
+ "output_logits": false,
63
+ "pp_degree": 1,
64
+ "sequence_length": 4096,
65
+ "speculation_length": 0,
66
+ "start_rank_id": 0,
67
+ "target": null,
68
+ "torch_dtype": "float16",
69
+ "tp_degree": 2
70
+ },
71
+ "num_attention_heads": 14,
72
+ "num_hidden_layers": 24,
73
+ "num_key_value_heads": 2,
74
+ "rms_norm_eps": 1e-06,
75
+ "rope_scaling": null,
76
+ "rope_theta": 1000000.0,
77
+ "sliding_window": null,
78
+ "tie_word_embeddings": true,
79
+ "use_cache": true,
80
+ "use_mrope": false,
81
+ "use_sliding_window": false,
82
+ "vocab_size": 151936
83
+ }
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/qwen2/yujiepan/qwen2.5-128k-tiny-random/4efceebded25852079ca.json ADDED
@@ -0,0 +1,65 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "yujiepan/qwen2.5-128k-tiny-random",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Qwen2ForCausalLM"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 8,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 16,
13
+ "layer_types": [
14
+ "full_attention",
15
+ "full_attention"
16
+ ],
17
+ "max_position_embeddings": 32768,
18
+ "max_window_layers": 1,
19
+ "model_type": "qwen2",
20
+ "neuron": {
21
+ "_serialized_key": "NxDNeuronConfig",
22
+ "batch_size": 2,
23
+ "capacity_factor": null,
24
+ "checkpoint_id": "yujiepan/qwen2.5-128k-tiny-random",
25
+ "checkpoint_revision": "c8296d4ca3f87782876d2382fbb6481d1beb8ef0",
26
+ "continuous_batching": true,
27
+ "enable_bucketing": false,
28
+ "ep_degree": 1,
29
+ "fused_qkv": false,
30
+ "glu_mlp": true,
31
+ "local_ranks_size": 2,
32
+ "logical_nc_config": 1,
33
+ "max_batch_size": 2,
34
+ "max_context_length": 100,
35
+ "max_topk": 256,
36
+ "n_active_tokens": 100,
37
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
38
+ "on_device_sampling": false,
39
+ "optimum_neuron_version": "0.3.1.dev5",
40
+ "output_logits": false,
41
+ "pp_degree": 1,
42
+ "sequence_length": 100,
43
+ "speculation_length": 0,
44
+ "start_rank_id": 0,
45
+ "target": null,
46
+ "torch_dtype": "float16",
47
+ "tp_degree": 2
48
+ },
49
+ "num_attention_heads": 4,
50
+ "num_hidden_layers": 2,
51
+ "num_key_value_heads": 2,
52
+ "rms_norm_eps": 1e-06,
53
+ "rope_scaling": {
54
+ "factor": 4.0,
55
+ "original_max_position_embeddings": 32768,
56
+ "rope_type": "yarn",
57
+ "type": "yarn"
58
+ },
59
+ "rope_theta": 1000000.0,
60
+ "sliding_window": null,
61
+ "tie_word_embeddings": false,
62
+ "use_cache": true,
63
+ "use_sliding_window": false,
64
+ "vocab_size": 152064
65
+ }
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/qwen2/yujiepan/qwen2.5-128k-tiny-random/6cc739f6dbdb1b0e1676.json ADDED
@@ -0,0 +1,65 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "yujiepan/qwen2.5-128k-tiny-random",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Qwen2ForCausalLM"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 8,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 16,
13
+ "layer_types": [
14
+ "full_attention",
15
+ "full_attention"
16
+ ],
17
+ "max_position_embeddings": 32768,
18
+ "max_window_layers": 1,
19
+ "model_type": "qwen2",
20
+ "neuron": {
21
+ "_serialized_key": "NxDNeuronConfig",
22
+ "batch_size": 1,
23
+ "capacity_factor": null,
24
+ "checkpoint_id": "yujiepan/qwen2.5-128k-tiny-random",
25
+ "checkpoint_revision": "c8296d4ca3f87782876d2382fbb6481d1beb8ef0",
26
+ "continuous_batching": false,
27
+ "enable_bucketing": false,
28
+ "ep_degree": 1,
29
+ "fused_qkv": false,
30
+ "glu_mlp": true,
31
+ "local_ranks_size": 2,
32
+ "logical_nc_config": 1,
33
+ "max_batch_size": 1,
34
+ "max_context_length": 100,
35
+ "max_topk": 256,
36
+ "n_active_tokens": 100,
37
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
38
+ "on_device_sampling": true,
39
+ "optimum_neuron_version": "0.3.1.dev5",
40
+ "output_logits": false,
41
+ "pp_degree": 1,
42
+ "sequence_length": 100,
43
+ "speculation_length": 0,
44
+ "start_rank_id": 0,
45
+ "target": null,
46
+ "torch_dtype": "bfloat16",
47
+ "tp_degree": 2
48
+ },
49
+ "num_attention_heads": 4,
50
+ "num_hidden_layers": 2,
51
+ "num_key_value_heads": 2,
52
+ "rms_norm_eps": 1e-06,
53
+ "rope_scaling": {
54
+ "factor": 4.0,
55
+ "original_max_position_embeddings": 32768,
56
+ "rope_type": "yarn",
57
+ "type": "yarn"
58
+ },
59
+ "rope_theta": 1000000.0,
60
+ "sliding_window": null,
61
+ "tie_word_embeddings": false,
62
+ "use_cache": true,
63
+ "use_sliding_window": false,
64
+ "vocab_size": 152064
65
+ }
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/qwen2/yujiepan/qwen2.5-128k-tiny-random/8a9598c30e93d592df6b.json ADDED
@@ -0,0 +1,65 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "yujiepan/qwen2.5-128k-tiny-random",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Qwen2ForCausalLM"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 8,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 16,
13
+ "layer_types": [
14
+ "full_attention",
15
+ "full_attention"
16
+ ],
17
+ "max_position_embeddings": 32768,
18
+ "max_window_layers": 1,
19
+ "model_type": "qwen2",
20
+ "neuron": {
21
+ "_serialized_key": "NxDNeuronConfig",
22
+ "batch_size": 1,
23
+ "capacity_factor": null,
24
+ "checkpoint_id": "yujiepan/qwen2.5-128k-tiny-random",
25
+ "checkpoint_revision": "c8296d4ca3f87782876d2382fbb6481d1beb8ef0",
26
+ "continuous_batching": false,
27
+ "enable_bucketing": false,
28
+ "ep_degree": 1,
29
+ "fused_qkv": false,
30
+ "glu_mlp": true,
31
+ "local_ranks_size": 2,
32
+ "logical_nc_config": 1,
33
+ "max_batch_size": 1,
34
+ "max_context_length": 100,
35
+ "max_topk": 256,
36
+ "n_active_tokens": 100,
37
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
38
+ "on_device_sampling": true,
39
+ "optimum_neuron_version": "0.3.1.dev5",
40
+ "output_logits": false,
41
+ "pp_degree": 1,
42
+ "sequence_length": 100,
43
+ "speculation_length": 0,
44
+ "start_rank_id": 0,
45
+ "target": null,
46
+ "torch_dtype": "float16",
47
+ "tp_degree": 2
48
+ },
49
+ "num_attention_heads": 4,
50
+ "num_hidden_layers": 2,
51
+ "num_key_value_heads": 2,
52
+ "rms_norm_eps": 1e-06,
53
+ "rope_scaling": {
54
+ "factor": 4.0,
55
+ "original_max_position_embeddings": 32768,
56
+ "rope_type": "yarn",
57
+ "type": "yarn"
58
+ },
59
+ "rope_theta": 1000000.0,
60
+ "sliding_window": null,
61
+ "tie_word_embeddings": false,
62
+ "use_cache": true,
63
+ "use_sliding_window": false,
64
+ "vocab_size": 152064
65
+ }
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/qwen3_moe/optimum-internal-testing/tiny-random-qwen3_moe/1550081bb53b23759b89.json ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "optimum-internal-testing/tiny-random-qwen3_moe",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Qwen3MoeForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "decoder_sparse_step": 2,
11
+ "head_dim": 32,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 64,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 128,
16
+ "max_position_embeddings": 40960,
17
+ "max_window_layers": 1,
18
+ "mlp_only_layers": [],
19
+ "model_type": "qwen3_moe",
20
+ "moe_intermediate_size": 128,
21
+ "neuron": {
22
+ "_serialized_key": "NxDNeuronConfig",
23
+ "batch_size": 2,
24
+ "capacity_factor": null,
25
+ "checkpoint_id": "optimum-internal-testing/tiny-random-qwen3_moe",
26
+ "checkpoint_revision": "e0230be2839556b44b7400a233c73c74b4abb7af",
27
+ "continuous_batching": false,
28
+ "enable_bucketing": false,
29
+ "ep_degree": 1,
30
+ "fused_qkv": false,
31
+ "glu_mlp": true,
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
+ "n_active_tokens": 100,
38
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
39
+ "on_device_sampling": false,
40
+ "optimum_neuron_version": "0.3.1.dev5",
41
+ "output_logits": false,
42
+ "pp_degree": 1,
43
+ "sequence_length": 100,
44
+ "speculation_length": 0,
45
+ "start_rank_id": 0,
46
+ "target": null,
47
+ "torch_dtype": "float16",
48
+ "tp_degree": 2
49
+ },
50
+ "norm_topk_prob": true,
51
+ "num_attention_heads": 2,
52
+ "num_experts": 8,
53
+ "num_experts_per_tok": 2,
54
+ "num_hidden_layers": 2,
55
+ "num_key_value_heads": 1,
56
+ "output_router_logits": false,
57
+ "rms_norm_eps": 1e-06,
58
+ "rope_scaling": null,
59
+ "rope_theta": 1000000.0,
60
+ "router_aux_loss_coef": 0.001,
61
+ "sliding_window": null,
62
+ "tie_word_embeddings": true,
63
+ "use_cache": true,
64
+ "use_sliding_window": false,
65
+ "vocab_size": 151936
66
+ }
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/qwen3_moe/optimum-internal-testing/tiny-random-qwen3_moe/3a9ad8456bbc98d3a4b5.json ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "optimum-internal-testing/tiny-random-qwen3_moe",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Qwen3MoeForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "decoder_sparse_step": 2,
11
+ "head_dim": 32,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 64,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 128,
16
+ "max_position_embeddings": 40960,
17
+ "max_window_layers": 1,
18
+ "mlp_only_layers": [],
19
+ "model_type": "qwen3_moe",
20
+ "moe_intermediate_size": 128,
21
+ "neuron": {
22
+ "_serialized_key": "NxDNeuronConfig",
23
+ "batch_size": 1,
24
+ "capacity_factor": null,
25
+ "checkpoint_id": "optimum-internal-testing/tiny-random-qwen3_moe",
26
+ "checkpoint_revision": "e0230be2839556b44b7400a233c73c74b4abb7af",
27
+ "continuous_batching": false,
28
+ "enable_bucketing": false,
29
+ "ep_degree": 1,
30
+ "fused_qkv": false,
31
+ "glu_mlp": true,
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
+ "n_active_tokens": 100,
38
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
39
+ "on_device_sampling": false,
40
+ "optimum_neuron_version": "0.3.1.dev5",
41
+ "output_logits": false,
42
+ "pp_degree": 1,
43
+ "sequence_length": 100,
44
+ "speculation_length": 0,
45
+ "start_rank_id": 0,
46
+ "target": null,
47
+ "torch_dtype": "bfloat16",
48
+ "tp_degree": 2
49
+ },
50
+ "norm_topk_prob": true,
51
+ "num_attention_heads": 2,
52
+ "num_experts": 8,
53
+ "num_experts_per_tok": 2,
54
+ "num_hidden_layers": 2,
55
+ "num_key_value_heads": 1,
56
+ "output_router_logits": false,
57
+ "rms_norm_eps": 1e-06,
58
+ "rope_scaling": null,
59
+ "rope_theta": 1000000.0,
60
+ "router_aux_loss_coef": 0.001,
61
+ "sliding_window": null,
62
+ "tie_word_embeddings": true,
63
+ "use_cache": true,
64
+ "use_sliding_window": false,
65
+ "vocab_size": 151936
66
+ }
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev5/qwen3_moe/optimum-internal-testing/tiny-random-qwen3_moe/9a119236f3406ad16638.json ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "optimum-internal-testing/tiny-random-qwen3_moe",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Qwen3MoeForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "decoder_sparse_step": 2,
11
+ "head_dim": 32,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 64,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 128,
16
+ "max_position_embeddings": 40960,
17
+ "max_window_layers": 1,
18
+ "mlp_only_layers": [],
19
+ "model_type": "qwen3_moe",
20
+ "moe_intermediate_size": 128,
21
+ "neuron": {
22
+ "_serialized_key": "NxDNeuronConfig",
23
+ "batch_size": 1,
24
+ "capacity_factor": null,
25
+ "checkpoint_id": "optimum-internal-testing/tiny-random-qwen3_moe",
26
+ "checkpoint_revision": "e0230be2839556b44b7400a233c73c74b4abb7af",
27
+ "continuous_batching": false,
28
+ "enable_bucketing": false,
29
+ "ep_degree": 1,
30
+ "fused_qkv": false,
31
+ "glu_mlp": true,
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
+ "n_active_tokens": 100,
38
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
39
+ "on_device_sampling": false,
40
+ "optimum_neuron_version": "0.3.1.dev5",
41
+ "output_logits": false,
42
+ "pp_degree": 1,
43
+ "sequence_length": 100,
44
+ "speculation_length": 0,
45
+ "start_rank_id": 0,
46
+ "target": null,
47
+ "torch_dtype": "float16",
48
+ "tp_degree": 2
49
+ },
50
+ "norm_topk_prob": true,
51
+ "num_attention_heads": 2,
52
+ "num_experts": 8,
53
+ "num_experts_per_tok": 2,
54
+ "num_hidden_layers": 2,
55
+ "num_key_value_heads": 1,
56
+ "output_router_logits": false,
57
+ "rms_norm_eps": 1e-06,
58
+ "rope_scaling": null,
59
+ "rope_theta": 1000000.0,
60
+ "router_aux_loss_coef": 0.001,
61
+ "sliding_window": null,
62
+ "tie_word_embeddings": true,
63
+ "use_cache": true,
64
+ "use_sliding_window": false,
65
+ "vocab_size": 151936
66
+ }
neuronxcc-2.19.8089.0+8ab9f450/MODULE_147ae45766ca99a75121+253d6470/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn1", "--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", "--internal-hlo2tensorizer-options=--verify-hlo=true", "--logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.19.8089.0+8ab9f450/MODULE_147ae45766ca99a75121+253d6470/model.done ADDED
File without changes
neuronxcc-2.19.8089.0+8ab9f450/MODULE_147ae45766ca99a75121+253d6470/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e337899cceac9fa29b9f9e18aed5193cb8c8ad9a0179029e141e4fef7bb88cd0
3
+ size 63460
neuronxcc-2.19.8089.0+8ab9f450/MODULE_147ae45766ca99a75121+253d6470/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:47a9663af1ba0f41f105c6c305206c147eba1c3df777f77f4c9df596a6bc8605
3
+ size 287744
neuronxcc-2.19.8089.0+8ab9f450/MODULE_7f8f6cb9ef9069b9a192+cd3419b6/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn1", "--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", "--internal-hlo2tensorizer-options=--verify-hlo=true", "--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.19.8089.0+8ab9f450/MODULE_7f8f6cb9ef9069b9a192+cd3419b6/model.done ADDED
File without changes
neuronxcc-2.19.8089.0+8ab9f450/MODULE_7f8f6cb9ef9069b9a192+cd3419b6/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4852a8338a1d9d344df084329d7286e82de06fd0420b50b169409061b9a1f34f
3
+ size 54991
neuronxcc-2.19.8089.0+8ab9f450/MODULE_7f8f6cb9ef9069b9a192+cd3419b6/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5ddbb1b4253a04ab6347f1ee2b1bcad3c4c9d7f273412b76a35c9477f9885186
3
+ size 287744
neuronxcc-2.19.8089.0+8ab9f450/MODULE_7f8f6cb9ef9069b9a192+cd3419b6/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a53fd364c9df45f6d70aa4c9b9bf9721d35574ae3ecea3d85d9fb6d6dfd93cf6
3
+ size 297990
neuronxcc-2.19.8089.0+8ab9f450/MODULE_afd876d1d56e17288e52+253d6470/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn1", "--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", "--internal-hlo2tensorizer-options=--verify-hlo=true", "--logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.19.8089.0+8ab9f450/MODULE_afd876d1d56e17288e52+253d6470/model.done ADDED
File without changes
neuronxcc-2.19.8089.0+8ab9f450/MODULE_afd876d1d56e17288e52+253d6470/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c4ac7fc1389ea5ee07b96defa7ca191a12b4d9f9d141c86f2debc5a4e0535ed1
3
+ size 64132
neuronxcc-2.19.8089.0+8ab9f450/MODULE_afd876d1d56e17288e52+253d6470/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:66c3049bfa3918307e61a6bd10f4f0717b2549e2b28d24b32c27106d6642f297
3
+ size 267264
neuronxcc-2.19.8089.0+8ab9f450/MODULE_bb03c2c3bd153e48eeea+cd3419b6/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn1", "--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", "--internal-hlo2tensorizer-options=--verify-hlo=true", "--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.19.8089.0+8ab9f450/MODULE_bb03c2c3bd153e48eeea+cd3419b6/model.done ADDED
File without changes
neuronxcc-2.19.8089.0+8ab9f450/MODULE_bb03c2c3bd153e48eeea+cd3419b6/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a52fb5639f14fd0b6c5005656dc5368e9f6e26cfa77bdabeeba39bc08779c455
3
+ size 60703
neuronxcc-2.19.8089.0+8ab9f450/MODULE_bb03c2c3bd153e48eeea+cd3419b6/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:80eff9b6cdd6e8290e1eaf2a6735906363619eb37832c80db2e5fe220f50fd54
3
+ size 277504
neuronxcc-2.19.8089.0+8ab9f450/MODULE_bb03c2c3bd153e48eeea+cd3419b6/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:db19b80a6325210718994aae2e80e10ac8996b77608bf29b77addfe6aa2cde12
3
+ size 287750
neuronxcc-2.19.8089.0+8ab9f450/MODULE_bb945fbdfaf46f441eed+cd3419b6/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn1", "--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", "--internal-hlo2tensorizer-options=--verify-hlo=true", "--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.19.8089.0+8ab9f450/MODULE_bb945fbdfaf46f441eed+cd3419b6/model.done ADDED
File without changes
neuronxcc-2.19.8089.0+8ab9f450/MODULE_bb945fbdfaf46f441eed+cd3419b6/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:85fa1dc2d01483952b893050dde84142316ccf28250b5ba845b34c6991b808b6
3
+ size 54990
neuronxcc-2.19.8089.0+8ab9f450/MODULE_bb945fbdfaf46f441eed+cd3419b6/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f4e0dc0590cfc7f8996b44e26b6550e454c4c57a808c7f52a9bf58accd7e57f4
3
+ size 287744
neuronxcc-2.19.8089.0+8ab9f450/MODULE_bb945fbdfaf46f441eed+cd3419b6/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2832140ca235a73ecaf128ead1dc28902e8f2a902307e1a7d076bd1bd17e2454
3
+ size 297990
neuronxcc-2.19.8089.0+8ab9f450/MODULE_ff4faa8edb2d2eec282d+253d6470/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn1", "--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", "--internal-hlo2tensorizer-options=--verify-hlo=true", "--logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.19.8089.0+8ab9f450/MODULE_ff4faa8edb2d2eec282d+253d6470/model.done ADDED
File without changes
neuronxcc-2.19.8089.0+8ab9f450/MODULE_ff4faa8edb2d2eec282d+253d6470/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a8b4b8c77f61111bed9ebd6fc802d768af68c809b2ae58ae1f1661d3546a0254
3
+ size 63459
neuronxcc-2.19.8089.0+8ab9f450/MODULE_ff4faa8edb2d2eec282d+253d6470/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:088106e4447109c542d444ffffc3877e0ed00a87f29adbe5b75e49d410ff0607
3
+ size 287744