Update config.json
Browse files- config.json +47 -2
config.json
CHANGED
@@ -1,5 +1,47 @@
|
|
1 |
{
|
2 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
"vocab_size": 50257,
|
4 |
"max_position_embeddings": 1024,
|
5 |
"encoder_layers": 24,
|
@@ -10,5 +52,8 @@
|
|
10 |
"decoder_attention_heads": 16,
|
11 |
"dropout": 0.2,
|
12 |
"activation_function": "gelu",
|
13 |
-
"initializer_range": 0.02
|
|
|
|
|
|
|
14 |
}
|
|
|
1 |
{
|
2 |
+
"model_name": "Daedalus_1: The Forge of Visionary Innovation",
|
3 |
+
"description": "A cutting-edge AI model blending CodeBERT, Codex, T5, SAM, Gemini, and Megatron for transformative innovation.",
|
4 |
+
"version": "1.0",
|
5 |
+
"author": "Or4cl3 AI Solutions",
|
6 |
+
"framework": "PyTorch",
|
7 |
+
"python_version": "3.x",
|
8 |
+
"deep_learning_framework": "PyTorch",
|
9 |
+
"constituent_models": [
|
10 |
+
{
|
11 |
+
"name": "Or4cl3-1/code-slerp",
|
12 |
+
"description": "Empowers Daedalus_1 with expertise in software engineering, code generation, and task-oriented language understanding."
|
13 |
+
},
|
14 |
+
{
|
15 |
+
"name": "Or4cl3-1/SAM-Gemini-BLOOM-OPT-Gopher-Megatron-slerp",
|
16 |
+
"description": "Provides Daedalus_1 with a deep, multifaceted understanding of complex concepts, human-like reasoning, and creativity."
|
17 |
+
}
|
18 |
+
],
|
19 |
+
"architectural_design": {
|
20 |
+
"merge_method": "SLERP (Spherical Linear Interpolation)",
|
21 |
+
"selected_layers": 32
|
22 |
+
},
|
23 |
+
"capabilities": [
|
24 |
+
"Rapid Prototyping and Code Generation",
|
25 |
+
"Multidisciplinary Understanding",
|
26 |
+
"Adaptability and Continuous Improvement",
|
27 |
+
"Ethical Considerations"
|
28 |
+
],
|
29 |
+
"applications": [
|
30 |
+
"Software Development",
|
31 |
+
"Scientific Research",
|
32 |
+
"Creative Problem-Solving"
|
33 |
+
],
|
34 |
+
"training_data": "Internal and External Datasets",
|
35 |
+
"training_steps": 200000,
|
36 |
+
"batch_size": 32,
|
37 |
+
"learning_rate": 0.0001,
|
38 |
+
"max_sequence_length": 1024,
|
39 |
+
"num_layers": 24,
|
40 |
+
"num_heads": 16,
|
41 |
+
"hidden_size": 1024,
|
42 |
+
"dropout_rate": 0.2,
|
43 |
+
"num_epochs": 20,
|
44 |
+
"model_type": "blenderbot",
|
45 |
"vocab_size": 50257,
|
46 |
"max_position_embeddings": 1024,
|
47 |
"encoder_layers": 24,
|
|
|
52 |
"decoder_attention_heads": 16,
|
53 |
"dropout": 0.2,
|
54 |
"activation_function": "gelu",
|
55 |
+
"initializer_range": 0.02,
|
56 |
+
"tie_encoder_decoder": True,
|
57 |
+
"tie_word_embeddings": True,
|
58 |
+
"output_past": True
|
59 |
}
|