jeffra commited on
Commit
f1a28d2
1 Parent(s): 8c87923

Upload config.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. config.json +75 -0
config.json ADDED
@@ -0,0 +1,75 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "ArcticForCausalLM"
4
+ ],
5
+ "attention_dropout": 0,
6
+ "bos_token_id": 1,
7
+ "enable_expert_tensor_parallelism": false,
8
+ "enc_index": [
9
+ 0,
10
+ 1,
11
+ 2,
12
+ 3,
13
+ 4,
14
+ 5,
15
+ 6,
16
+ 7,
17
+ 8,
18
+ 9,
19
+ 10,
20
+ 11,
21
+ 12,
22
+ 13,
23
+ 14,
24
+ 15,
25
+ 16,
26
+ 17,
27
+ 18,
28
+ 19,
29
+ 20,
30
+ 21,
31
+ 22,
32
+ 23,
33
+ 24,
34
+ 25,
35
+ 26,
36
+ 27,
37
+ 28,
38
+ 29,
39
+ 30,
40
+ 31,
41
+ 32,
42
+ 33,
43
+ 34
44
+ ],
45
+ "eos_token_id": 2,
46
+ "hidden_act": "silu",
47
+ "hidden_size": 7168,
48
+ "initializer_range": 0.02,
49
+ "intermediate_size": 4864,
50
+ "max_position_embeddings": 4096,
51
+ "max_sequence_length": 4096,
52
+ "model_type": "arctic",
53
+ "moe_eval_capacity_factor": 1,
54
+ "moe_layer_frequency": 1,
55
+ "moe_min_capacity": 0,
56
+ "moe_token_dropping": false,
57
+ "moe_train_capacity_factor": 1,
58
+ "num_attention_heads": 56,
59
+ "num_experts_per_tok": 2,
60
+ "num_hidden_layers": 35,
61
+ "num_key_value_heads": 56,
62
+ "num_local_experts": 128,
63
+ "parallel_attn_mlp_res": true,
64
+ "quantization": null,
65
+ "rms_norm_eps": 1e-05,
66
+ "rope_theta": 10000,
67
+ "router_aux_loss_coef": 0.001,
68
+ "sliding_window": null,
69
+ "tie_word_embeddings": false,
70
+ "torch_dtype": "bfloat16",
71
+ "transformers_version": "4.39.0.dev0",
72
+ "use_cache": true,
73
+ "use_residual": true,
74
+ "vocab_size": 32000
75
+ }