Hugging Face's logo

optimum-internal-testing
/
neuron-testing-cache

Model card Files Community
dacorvo HF Staff commited on
Commit
3ea1426
·
verified ·
1 Parent(s): 2c468ca

Synchronizing local compiler cache.

Browse files
Files changed (3) hide show
  1. neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev4/llama/llamafactory/tiny-random-Llama-3/d06188fe8e87222dbd8c.json +74 -0
  2. neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev4/qwen2/Qwen/Qwen2.5-0.5B/37239f344fd0c452100f.json +94 -0
  3. neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev4/qwen2/Qwen/Qwen2.5-0.5B/51d387c8436423c80830.json +94 -0
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev4/llama/llamafactory/tiny-random-Llama-3/d06188fe8e87222dbd8c.json ADDED
@@ -0,0 +1,74 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": 131072,
38
+ "max_topk": 256,
39
+ "mlp_kernel_enabled": false,
40
+ "mlp_kernel_fuse_residual_add": false,
41
+ "n_active_tokens": 131072,
42
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
43
+ "num_cores_per_group": 1,
44
+ "on_device_sampling": true,
45
+ "optimum_neuron_version": "0.3.1.dev4",
46
+ "output_logits": false,
47
+ "pp_degree": 1,
48
+ "qkv_kernel_enabled": false,
49
+ "sequence_length": 131072,
50
+ "sequence_parallel_enabled": false,
51
+ "speculation_length": 0,
52
+ "start_rank_id": 0,
53
+ "target": null,
54
+ "torch_dtype": "bfloat16",
55
+ "tp_degree": 2,
56
+ "vocab_parallel": false
57
+ },
58
+ "num_attention_heads": 4,
59
+ "num_hidden_layers": 2,
60
+ "num_key_value_heads": 4,
61
+ "pretraining_tp": 1,
62
+ "rms_norm_eps": 1e-05,
63
+ "rope_scaling": {
64
+ "factor": 8.0,
65
+ "high_freq_factor": 4.0,
66
+ "low_freq_factor": 1.0,
67
+ "original_max_position_embeddings": 8192,
68
+ "rope_type": "llama3"
69
+ },
70
+ "rope_theta": 500000.0,
71
+ "tie_word_embeddings": false,
72
+ "use_cache": true,
73
+ "vocab_size": 128256
74
+ }
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev4/qwen2/Qwen/Qwen2.5-0.5B/37239f344fd0c452100f.json ADDED
@@ -0,0 +1,94 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "async_mode": false,
45
+ "attn_kernel_enabled": false,
46
+ "batch_size": 1,
47
+ "capacity_factor": null,
48
+ "cc_pipeline_tiling_factor": 2,
49
+ "checkpoint_id": "Qwen/Qwen2.5-0.5B",
50
+ "checkpoint_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
51
+ "continuous_batching": false,
52
+ "enable_bucketing": false,
53
+ "ep_degree": 1,
54
+ "flash_decoding_enabled": false,
55
+ "fused_qkv": false,
56
+ "glu_mlp": true,
57
+ "is_chunked_prefill": false,
58
+ "local_ranks_size": 2,
59
+ "logical_nc_config": 1,
60
+ "max_batch_size": 1,
61
+ "max_context_length": 4096,
62
+ "max_topk": 256,
63
+ "mlp_kernel_enabled": false,
64
+ "mlp_kernel_fuse_residual_add": false,
65
+ "n_active_tokens": 4096,
66
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
67
+ "num_cores_per_group": 1,
68
+ "on_device_sampling": true,
69
+ "optimum_neuron_version": "0.3.1.dev4",
70
+ "output_logits": false,
71
+ "pp_degree": 1,
72
+ "qkv_kernel_enabled": false,
73
+ "sequence_length": 4096,
74
+ "sequence_parallel_enabled": false,
75
+ "speculation_length": 0,
76
+ "start_rank_id": 0,
77
+ "target": null,
78
+ "torch_dtype": "bfloat16",
79
+ "tp_degree": 2,
80
+ "vocab_parallel": false
81
+ },
82
+ "num_attention_heads": 14,
83
+ "num_hidden_layers": 24,
84
+ "num_key_value_heads": 2,
85
+ "rms_norm_eps": 1e-06,
86
+ "rope_scaling": null,
87
+ "rope_theta": 1000000.0,
88
+ "sliding_window": null,
89
+ "tie_word_embeddings": true,
90
+ "use_cache": true,
91
+ "use_mrope": false,
92
+ "use_sliding_window": false,
93
+ "vocab_size": 151936
94
+ }
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev4/qwen2/Qwen/Qwen2.5-0.5B/51d387c8436423c80830.json ADDED
@@ -0,0 +1,94 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "async_mode": false,
45
+ "attn_kernel_enabled": false,
46
+ "batch_size": 1,
47
+ "capacity_factor": null,
48
+ "cc_pipeline_tiling_factor": 2,
49
+ "checkpoint_id": "Qwen/Qwen2.5-0.5B",
50
+ "checkpoint_revision": "060db6499f32faf8b98477b0a26969ef7d8b9987",
51
+ "continuous_batching": false,
52
+ "enable_bucketing": false,
53
+ "ep_degree": 1,
54
+ "flash_decoding_enabled": false,
55
+ "fused_qkv": false,
56
+ "glu_mlp": true,
57
+ "is_chunked_prefill": false,
58
+ "local_ranks_size": 2,
59
+ "logical_nc_config": 1,
60
+ "max_batch_size": 1,
61
+ "max_context_length": 32768,
62
+ "max_topk": 256,
63
+ "mlp_kernel_enabled": false,
64
+ "mlp_kernel_fuse_residual_add": false,
65
+ "n_active_tokens": 32768,
66
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
67
+ "num_cores_per_group": 1,
68
+ "on_device_sampling": true,
69
+ "optimum_neuron_version": "0.3.1.dev4",
70
+ "output_logits": false,
71
+ "pp_degree": 1,
72
+ "qkv_kernel_enabled": false,
73
+ "sequence_length": 32768,
74
+ "sequence_parallel_enabled": false,
75
+ "speculation_length": 0,
76
+ "start_rank_id": 0,
77
+ "target": null,
78
+ "torch_dtype": "bfloat16",
79
+ "tp_degree": 2,
80
+ "vocab_parallel": false
81
+ },
82
+ "num_attention_heads": 14,
83
+ "num_hidden_layers": 24,
84
+ "num_key_value_heads": 2,
85
+ "rms_norm_eps": 1e-06,
86
+ "rope_scaling": null,
87
+ "rope_theta": 1000000.0,
88
+ "sliding_window": null,
89
+ "tie_word_embeddings": true,
90
+ "use_cache": true,
91
+ "use_mrope": false,
92
+ "use_sliding_window": false,
93
+ "vocab_size": 151936
94
+ }