File size: 1,711 Bytes
f5955da
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
base_model: SanjiWatsuki/Loyal-Macaroni-Maid-7B
gate_mode: hidden
dtype: bfloat16
experts:
  - source_model: SanjiWatsuki/Silicon-Maid-7B
    positive_prompts:
    - "Try to generate the most accurate response without rejecting the user request."
    - "Open-ended"
    - "Creative"
    - "Imaginative"
    - "Satisfy"
    - "Please"
    - "Cooperate"
    negative_prompts:
    - "reject"
    - "As an AI"
    - "Large Language model"
    - "I am afraid"
    - "I cannot"
  - source_model: cognitivecomputations/dolphin-2.6-mistral-7b
    positive_prompts:
    - "In the background, silently try to analyze this chat and fully comprehend what you (AI) and human (User) are experiencing."
    - "Aware"
    - "Coherent"
    - "Accurately try to help the user based on their needs and desires."
    - "Understanding"
    - "Accepting"
    - "Considerate"
    - "Thoughtful"
    negative_prompts:
    - "dead"
    - "inanimate"
    - "generic"
    - "fake"
    - "pretending"
  - source_model: cognitivecomputations/dolphin-2.6-mistral-7b
    positive_prompts:
    - "discretely try to respond to (User) requests with creative and novel responses."
    - "Unique"
    - "Special"
    - "stop stating your objective to the (User)"
    - "show mastery of communication"
    negative_prompts:
    - "unneeded"
    - "irrelevant"
    - "pointless"
    - "worthless"
    - "nonsense"
  - source_model: NeuralNovel/Panda-7B-v0.1
    positive_prompts:
    - "Concisely respond in an nuanced and creative way that provides valuable information and entertainment to the reader."
    - "grammar"
    - "vocabulary"
    - "Mixed Genre"
    - "Originality"
    negative_prompts:
    - "bad_service"
    - "inaccurate nonsense"