dacorvo HF Staff commited on
Commit
ee9e952
·
verified ·
1 Parent(s): a4609aa

Synchronizing local compiler cache.

Browse files
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev4/qwen2/deepseek-ai/DeepSeek-R1-Distill-Qwen-14B/72950cfdcce39e7372a3.json ADDED
@@ -0,0 +1,117 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Qwen2ForCausalLM"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 5120,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 13824,
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
+ "full_attention",
39
+ "full_attention",
40
+ "full_attention",
41
+ "full_attention",
42
+ "full_attention",
43
+ "full_attention",
44
+ "full_attention",
45
+ "full_attention",
46
+ "full_attention",
47
+ "full_attention",
48
+ "full_attention",
49
+ "full_attention",
50
+ "full_attention",
51
+ "full_attention",
52
+ "full_attention",
53
+ "full_attention",
54
+ "full_attention",
55
+ "full_attention",
56
+ "full_attention",
57
+ "full_attention",
58
+ "full_attention",
59
+ "full_attention",
60
+ "full_attention",
61
+ "full_attention"
62
+ ],
63
+ "max_position_embeddings": 131072,
64
+ "max_window_layers": 48,
65
+ "model_type": "qwen2",
66
+ "neuron": {
67
+ "_serialized_key": "NxDNeuronConfig",
68
+ "async_mode": false,
69
+ "attn_kernel_enabled": false,
70
+ "batch_size": 16,
71
+ "capacity_factor": null,
72
+ "cc_pipeline_tiling_factor": 2,
73
+ "checkpoint_id": "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B",
74
+ "checkpoint_revision": "1df8507178afcc1bef68cd8c393f61a886323761",
75
+ "continuous_batching": true,
76
+ "enable_bucketing": false,
77
+ "ep_degree": 1,
78
+ "flash_decoding_enabled": false,
79
+ "fused_qkv": false,
80
+ "glu_mlp": true,
81
+ "is_chunked_prefill": false,
82
+ "local_ranks_size": 8,
83
+ "logical_nc_config": 1,
84
+ "max_batch_size": 16,
85
+ "max_context_length": 4096,
86
+ "max_topk": 256,
87
+ "mlp_kernel_enabled": false,
88
+ "mlp_kernel_fuse_residual_add": false,
89
+ "n_active_tokens": 4096,
90
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
91
+ "num_cores_per_group": 1,
92
+ "on_device_sampling": true,
93
+ "optimum_neuron_version": "0.3.1.dev4",
94
+ "output_logits": false,
95
+ "pp_degree": 1,
96
+ "qkv_kernel_enabled": false,
97
+ "sequence_length": 4096,
98
+ "sequence_parallel_enabled": false,
99
+ "speculation_length": 0,
100
+ "start_rank_id": 0,
101
+ "target": null,
102
+ "torch_dtype": "bfloat16",
103
+ "tp_degree": 8,
104
+ "vocab_parallel": false
105
+ },
106
+ "num_attention_heads": 40,
107
+ "num_hidden_layers": 48,
108
+ "num_key_value_heads": 8,
109
+ "rms_norm_eps": 1e-05,
110
+ "rope_scaling": null,
111
+ "rope_theta": 1000000.0,
112
+ "sliding_window": null,
113
+ "tie_word_embeddings": false,
114
+ "use_cache": true,
115
+ "use_sliding_window": false,
116
+ "vocab_size": 152064
117
+ }