xuan luo
commited on
Upload config.json with huggingface_hub
Browse files- config.json +54 -0
config.json
ADDED
@@ -0,0 +1,54 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "./",
|
3 |
+
"architectures": [
|
4 |
+
"DDLlamaForCausalLM"
|
5 |
+
],
|
6 |
+
"attention_bias": false,
|
7 |
+
"attention_dropout": 0.0,
|
8 |
+
"auto_map": {
|
9 |
+
"AutoConfig": "configuration_ddllama.DDLlamaConfig",
|
10 |
+
"AutoModelForCausalLM": "modeling_ddllama.DDLlamaForCausalLM"
|
11 |
+
},
|
12 |
+
"bos_token_id": 128000,
|
13 |
+
"eos_token_id": 128009,
|
14 |
+
"head_dim": 128,
|
15 |
+
"hidden_act": "silu",
|
16 |
+
"hidden_size": 4096,
|
17 |
+
"initializer_range": 0.02,
|
18 |
+
"intermediate_size": 14336,
|
19 |
+
"max_position_embeddings": 8192,
|
20 |
+
"mlp_bias": false,
|
21 |
+
"model_type": "ddllama",
|
22 |
+
"num_attention_heads": 32,
|
23 |
+
"num_hidden_layers": 32,
|
24 |
+
"num_key_value_heads": 8,
|
25 |
+
"pretraining_tp": 1,
|
26 |
+
"proj_reduction_factor": 16,
|
27 |
+
"rms_norm_eps": 1e-05,
|
28 |
+
"rope_scaling": null,
|
29 |
+
"rope_theta": 500000.0,
|
30 |
+
"router_layers": [
|
31 |
+
16,
|
32 |
+
17,
|
33 |
+
18,
|
34 |
+
19,
|
35 |
+
20,
|
36 |
+
21,
|
37 |
+
22,
|
38 |
+
23,
|
39 |
+
24,
|
40 |
+
25,
|
41 |
+
26,
|
42 |
+
27,
|
43 |
+
28,
|
44 |
+
29,
|
45 |
+
30,
|
46 |
+
31
|
47 |
+
],
|
48 |
+
"router_reduction_factor": 16,
|
49 |
+
"tie_word_embeddings": false,
|
50 |
+
"torch_dtype": "bfloat16",
|
51 |
+
"transformers_version": "4.48.3",
|
52 |
+
"use_cache": false,
|
53 |
+
"vocab_size": 128256
|
54 |
+
}
|