Commit
•
112a067
1
Parent(s):
9a5e7b8
Update README.md
Browse files
README.md
CHANGED
@@ -1,34 +1,125 @@
|
|
1 |
---
|
2 |
base_model:
|
3 |
-
-
|
4 |
-
-
|
5 |
- TheBloke/Llama-2-13B-fp16
|
6 |
-
-
|
|
|
|
|
|
|
|
|
7 |
tags:
|
8 |
- mergekit
|
9 |
- merge
|
10 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
11 |
---
|
12 |
-
#
|
|
|
|
|
|
|
13 |
|
14 |
-
|
|
|
|
|
15 |
|
16 |
## Merge Details
|
|
|
|
|
|
|
|
|
17 |
### Merge Method
|
18 |
|
19 |
-
This model was merged using the [
|
20 |
|
21 |
### Models Merged
|
22 |
|
23 |
The following models were included in the merge:
|
24 |
-
* [
|
25 |
-
* [
|
26 |
-
* [
|
|
|
|
|
|
|
|
|
27 |
|
28 |
### Configuration
|
29 |
|
30 |
The following YAML configuration was used to produce this model:
|
31 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
32 |
```yaml
|
33 |
base_model:
|
34 |
model:
|
@@ -68,4 +159,4 @@ slices:
|
|
68 |
model:
|
69 |
model:
|
70 |
path: TheBloke/Llama-2-13B-fp16
|
71 |
-
```
|
|
|
1 |
---
|
2 |
base_model:
|
3 |
+
- Riiid/sheep-duck-llama-2-13b
|
4 |
+
- IkariDev/Athena-v4
|
5 |
- TheBloke/Llama-2-13B-fp16
|
6 |
+
- KoboldAI/LLaMA2-13B-Psyfighter2
|
7 |
+
- KoboldAI/LLaMA2-13B-Erebus-v3
|
8 |
+
- Henk717/echidna-tiefigther-25
|
9 |
+
- Undi95/Unholy-v2-13B
|
10 |
+
- ddh0/EstopianOrcaMaid-13b
|
11 |
tags:
|
12 |
- mergekit
|
13 |
- merge
|
14 |
+
- not-for-all-audiences
|
15 |
+
- ERP
|
16 |
+
- RP
|
17 |
+
- Roleplay
|
18 |
+
- uncensored
|
19 |
+
license: llama2
|
20 |
+
language:
|
21 |
+
- en
|
22 |
---
|
23 |
+
# Model
|
24 |
+
This is the GPTQ 4bit quantized version of SnowyRP
|
25 |
+
|
26 |
+
[FP16](https://huggingface.co/Masterjp123/SnowyRP-FinalV1-L2-13B)
|
27 |
|
28 |
+
[GPTQ](https://huggingface.co/Masterjp123/SnowyRP-FinalV1-L2-13B-GPTQ)
|
29 |
+
|
30 |
+
Any Future Quantizations I am made aware of will be added.
|
31 |
|
32 |
## Merge Details
|
33 |
+
just used highly ranked modles to try and get a better result, Also I made sure that Model incest would not be a BIG problem by merging models that are pretty pure.
|
34 |
+
|
35 |
+
These models CAN and WILL produce X rated or harmful content, due to being heavily uncensored in a attempt to not limit it
|
36 |
+
|
37 |
### Merge Method
|
38 |
|
39 |
+
This model was merged using the [ties](https://arxiv.org/abs/2306.01708) merge method using [TheBloke/Llama-2-13B-fp16](https://huggingface.co/TheBloke/Llama-2-13B-fp16) as a base.
|
40 |
|
41 |
### Models Merged
|
42 |
|
43 |
The following models were included in the merge:
|
44 |
+
* [Riiid/sheep-duck-llama-2-13b](https://huggingface.co/Riiid/sheep-duck-llama-2-13b)
|
45 |
+
* [IkariDev/Athena-v4](https://huggingface.co/IkariDev/Athena-v4)
|
46 |
+
* [KoboldAI/LLaMA2-13B-Psyfighter2](https://huggingface.co/KoboldAI/LLaMA2-13B-Psyfighter2)
|
47 |
+
* [KoboldAI/LLaMA2-13B-Erebus-v3](https://huggingface.co/KoboldAI/LLaMA2-13B-Erebus-v3)
|
48 |
+
* [Henk717/echidna-tiefigther-25](https://huggingface.co/Henk717/echidna-tiefigther-25)
|
49 |
+
* [Undi95/Unholy-v2-13B](https://huggingface.co/Undi95/Unholy-v2-13B)
|
50 |
+
* [EstopianOrcaMaid](https://huggingface.co/ddh0/EstopianOrcaMaid-13b)
|
51 |
|
52 |
### Configuration
|
53 |
|
54 |
The following YAML configuration was used to produce this model:
|
55 |
|
56 |
+
for P1
|
57 |
+
```yaml
|
58 |
+
base_model:
|
59 |
+
model:
|
60 |
+
path: TheBloke/Llama-2-13B-fp16
|
61 |
+
dtype: bfloat16
|
62 |
+
merge_method: task_arithmetic
|
63 |
+
slices:
|
64 |
+
- sources:
|
65 |
+
- layer_range: [0, 40]
|
66 |
+
model:
|
67 |
+
model:
|
68 |
+
path: TheBloke/Llama-2-13B-fp16
|
69 |
+
- layer_range: [0, 40]
|
70 |
+
model:
|
71 |
+
model:
|
72 |
+
path: Undi95/Unholy-v2-13B
|
73 |
+
parameters:
|
74 |
+
weight: 1.0
|
75 |
+
- layer_range: [0, 40]
|
76 |
+
model:
|
77 |
+
model:
|
78 |
+
path: Henk717/echidna-tiefigther-25
|
79 |
+
parameters:
|
80 |
+
weight: 0.45
|
81 |
+
- layer_range: [0, 40]
|
82 |
+
model:
|
83 |
+
model:
|
84 |
+
path: KoboldAI/LLaMA2-13B-Erebus-v3
|
85 |
+
parameters:
|
86 |
+
weight: 0.33
|
87 |
+
```
|
88 |
+
|
89 |
+
for P2
|
90 |
+
```yaml
|
91 |
+
base_model:
|
92 |
+
model:
|
93 |
+
path: TheBloke/Llama-2-13B-fp16
|
94 |
+
dtype: bfloat16
|
95 |
+
merge_method: task_arithmetic
|
96 |
+
slices:
|
97 |
+
- sources:
|
98 |
+
- layer_range: [0, 40]
|
99 |
+
model:
|
100 |
+
model:
|
101 |
+
path: TheBloke/Llama-2-13B-fp16
|
102 |
+
- layer_range: [0, 40]
|
103 |
+
model:
|
104 |
+
model:
|
105 |
+
path: KoboldAI/LLaMA2-13B-Psyfighter2
|
106 |
+
parameters:
|
107 |
+
weight: 1.0
|
108 |
+
- layer_range: [0, 40]
|
109 |
+
model:
|
110 |
+
model:
|
111 |
+
path: Riiid/sheep-duck-llama-2-13b
|
112 |
+
parameters:
|
113 |
+
weight: 0.45
|
114 |
+
- layer_range: [0, 40]
|
115 |
+
model:
|
116 |
+
model:
|
117 |
+
path: IkariDev/Athena-v4
|
118 |
+
parameters:
|
119 |
+
weight: 0.33
|
120 |
+
```
|
121 |
+
|
122 |
+
for the final merge
|
123 |
```yaml
|
124 |
base_model:
|
125 |
model:
|
|
|
159 |
model:
|
160 |
model:
|
161 |
path: TheBloke/Llama-2-13B-fp16
|
162 |
+
```
|