Create config.json
Browse files- config.json +70 -0
config.json
ADDED
@@ -0,0 +1,70 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "Deci/DeciLM-7b",
|
3 |
+
"architectures": [
|
4 |
+
"DeciLMForCausalLM"
|
5 |
+
],
|
6 |
+
"attention_bias": false,
|
7 |
+
"auto_map": {
|
8 |
+
"AutoConfig": "Deci/DeciLM-7b--configuration_decilm.DeciLMConfig",
|
9 |
+
"AutoModelForCausalLM": "Deci/DeciLM-7b--modeling_decilm.DeciLMForCausalLM"
|
10 |
+
},
|
11 |
+
"bos_token_id": 1,
|
12 |
+
"eos_token_id": 2,
|
13 |
+
"hidden_act": "silu",
|
14 |
+
"hidden_size": 4096,
|
15 |
+
"initializer_range": 0.02,
|
16 |
+
"intermediate_size": 14336,
|
17 |
+
"max_position_embeddings": 8192,
|
18 |
+
"model_type": "deci",
|
19 |
+
"num_attention_heads": 32,
|
20 |
+
"num_hidden_layers": 32,
|
21 |
+
"num_key_value_heads": 32,
|
22 |
+
"num_key_value_heads_per_layer": [
|
23 |
+
4,
|
24 |
+
4,
|
25 |
+
4,
|
26 |
+
4,
|
27 |
+
4,
|
28 |
+
2,
|
29 |
+
2,
|
30 |
+
2,
|
31 |
+
2,
|
32 |
+
2,
|
33 |
+
4,
|
34 |
+
2,
|
35 |
+
2,
|
36 |
+
2,
|
37 |
+
2,
|
38 |
+
2,
|
39 |
+
2,
|
40 |
+
2,
|
41 |
+
2,
|
42 |
+
2,
|
43 |
+
1,
|
44 |
+
1,
|
45 |
+
1,
|
46 |
+
1,
|
47 |
+
1,
|
48 |
+
1,
|
49 |
+
1,
|
50 |
+
1,
|
51 |
+
1,
|
52 |
+
1,
|
53 |
+
1,
|
54 |
+
4
|
55 |
+
],
|
56 |
+
"pretraining_tp": 1,
|
57 |
+
"rms_norm_eps": 1e-05,
|
58 |
+
"rope_scaling": {
|
59 |
+
"factor": 2.0,
|
60 |
+
"type": "dynamic"
|
61 |
+
},
|
62 |
+
"rope_theta": 10000.0,
|
63 |
+
"tie_word_embeddings": false,
|
64 |
+
"tokenizer_class": "LlamaTokenizer",
|
65 |
+
"torch_dtype": "bfloat16",
|
66 |
+
"transformers_version": "4.38.2",
|
67 |
+
"use_bfloat16": true,
|
68 |
+
"use_cache": true,
|
69 |
+
"vocab_size": 32000
|
70 |
+
}
|