Casual-Autopsy
commited on
Commit
•
b5c32a9
1
Parent(s):
c695489
Update README.md
Browse files
README.md
CHANGED
@@ -1,40 +1,170 @@
|
|
1 |
---
|
2 |
base_model:
|
3 |
-
-
|
|
|
|
|
|
|
|
|
|
|
|
|
4 |
- ChaoticNeutrals/Hathor_RP-v.01-L3-8B
|
|
|
5 |
- Casual-Autopsy/Omelette-2
|
|
|
|
|
6 |
- Sao10K/L3-8B-Stheno-v3.1
|
|
|
7 |
- ResplendentAI/Nymph_8B
|
8 |
-
- cgato/L3-TheSpice-8b-v0.8.3
|
9 |
-
- ChaoticNeutrals/Poppy_Porpoise-1.4-L3-8B
|
10 |
library_name: transformers
|
11 |
tags:
|
12 |
- mergekit
|
13 |
- merge
|
14 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
15 |
---
|
16 |
-
#
|
17 |
|
18 |
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
|
19 |
|
20 |
## Merge Details
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
21 |
### Merge Method
|
22 |
|
23 |
-
|
24 |
|
25 |
### Models Merged
|
26 |
|
27 |
The following models were included in the merge:
|
28 |
-
* [
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
29 |
* [ChaoticNeutrals/Hathor_RP-v.01-L3-8B](https://huggingface.co/ChaoticNeutrals/Hathor_RP-v.01-L3-8B)
|
30 |
* [Sao10K/L3-8B-Stheno-v3.1](https://huggingface.co/Sao10K/L3-8B-Stheno-v3.1)
|
|
|
31 |
* [ResplendentAI/Nymph_8B](https://huggingface.co/ResplendentAI/Nymph_8B)
|
32 |
-
* [cgato/L3-TheSpice-8b-v0.8.3](https://huggingface.co/cgato/L3-TheSpice-8b-v0.8.3)
|
33 |
-
* [ChaoticNeutrals/Poppy_Porpoise-1.4-L3-8B](https://huggingface.co/ChaoticNeutrals/Poppy_Porpoise-1.4-L3-8B)
|
34 |
|
35 |
-
###
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
36 |
|
37 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
38 |
|
39 |
```yaml
|
40 |
models:
|
@@ -60,5 +190,4 @@ models:
|
|
60 |
merge_method: task_arithmetic
|
61 |
base_model: Casual-Autopsy/Omelette-2
|
62 |
dtype: bfloat16
|
63 |
-
|
64 |
-
```
|
|
|
1 |
---
|
2 |
base_model:
|
3 |
+
- Sao10K/L3-8B-Stheno-v3.2
|
4 |
+
- Casual-Autopsy/L3-Umbral-Mind-RP-v0.3-8B
|
5 |
+
- bluuwhale/L3-SthenoMaidBlackroot-8B-V1
|
6 |
+
- migtissera/Llama-3-8B-Synthia-v3.5
|
7 |
+
- tannedbum/L3-Nymeria-Maid-8B
|
8 |
+
- Casual-Autopsy/L3-Umbral-Mind-RP-v1.0-8B
|
9 |
+
- tannedbum/L3-Nymeria-8B
|
10 |
- ChaoticNeutrals/Hathor_RP-v.01-L3-8B
|
11 |
+
- ChaoticNeutrals/Poppy_Porpoise-1.4-L3-8B
|
12 |
- Casual-Autopsy/Omelette-2
|
13 |
+
- cgato/L3-TheSpice-8b-v0.8.3
|
14 |
+
- ChaoticNeutrals/Hathor_RP-v.01-L3-8B
|
15 |
- Sao10K/L3-8B-Stheno-v3.1
|
16 |
+
- aifeifei798/llama3-8B-DarkIdol-1.0
|
17 |
- ResplendentAI/Nymph_8B
|
|
|
|
|
18 |
library_name: transformers
|
19 |
tags:
|
20 |
- mergekit
|
21 |
- merge
|
22 |
+
- not-for-all-audiences
|
23 |
+
- nsfw
|
24 |
+
- rp
|
25 |
+
- roleplay
|
26 |
+
- role-play
|
27 |
+
license: llama3
|
28 |
+
language:
|
29 |
+
- en
|
30 |
---
|
31 |
+
# L3-Uncen-Merger-Omelette-RP-v0.1-8B
|
32 |
|
33 |
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
|
34 |
|
35 |
## Merge Details
|
36 |
+
|
37 |
+
A merger recipe inspired by [invisietch/EtherealRainbow-v0.3-8B](https://huggingface.co/invisietch/EtherealRainbow-v0.3-8B) combined with a merger technique known as merge densification( [grimjim/kunoichi-lemon-royale-v3-32K-7B](https://huggingface.co/grimjim/kunoichi-lemon-royale-v3-32K-7B) )
|
38 |
+
|
39 |
+
The model recipe ended up being something I can only describe as making an omelette. Hence the model name.
|
40 |
+
|
41 |
+
The models are scrambled with Dare Ties to induce a bit of randomness, then the Dare Ties merges are merged into themselves with SLERP to repair any holes cause by Dare Ties, and finally a bunch of high creativity models are thrown into the merger through merge densification(Task Arithmetic).
|
42 |
+
|
43 |
+
This model uses a bunch of the top models of the UGI Leaderboard, I picked out a few of the top 8B models of each category. Most of the high creativity models in the last step were found through Lewdiculus' account uploads
|
44 |
+
|
45 |
+
**Downgraded to Stheno v3.2 due to issues with the model**
|
46 |
+
|
47 |
### Merge Method
|
48 |
|
49 |
+
Dare Ties, SLERP, and Task Arithmetic
|
50 |
|
51 |
### Models Merged
|
52 |
|
53 |
The following models were included in the merge:
|
54 |
+
* [Sao10K/L3-8B-Stheno-v3.2](https://huggingface.co/Sao10K/L3-8B-Stheno-v3.2)
|
55 |
+
* [Casual-Autopsy/L3-Umbral-Mind-RP-v0.3-8B](https://huggingface.co/Casual-Autopsy/L3-Umbral-Mind-RP-v0.3-8B)
|
56 |
+
* [bluuwhale/L3-SthenoMaidBlackroot-8B-V1](https://huggingface.co/bluuwhale/L3-SthenoMaidBlackroot-8B-V1)
|
57 |
+
* [migtissera/Llama-3-8B-Synthia-v3.5](https://huggingface.co/migtissera/Llama-3-8B-Synthia-v3.5)
|
58 |
+
* [tannedbum/L3-Nymeria-Maid-8B](https://huggingface.co/tannedbum/L3-Nymeria-Maid-8B)
|
59 |
+
* [Casual-Autopsy/L3-Umbral-Mind-RP-v1.0-8B](https://huggingface.co/Casual-Autopsy/L3-Umbral-Mind-RP-v1.0-8B)
|
60 |
+
* [tannedbum/L3-Nymeria-8B](https://huggingface.co/tannedbum/L3-Nymeria-8B)
|
61 |
+
* [ChaoticNeutrals/Hathor_RP-v.01-L3-8B](https://huggingface.co/ChaoticNeutrals/Hathor_RP-v.01-L3-8B)
|
62 |
+
* [ChaoticNeutrals/Poppy_Porpoise-1.4-L3-8B](https://huggingface.co/ChaoticNeutrals/Poppy_Porpoise-1.4-L3-8B)
|
63 |
+
* [cgato/L3-TheSpice-8b-v0.8.3](https://huggingface.co/cgato/L3-TheSpice-8b-v0.8.3)
|
64 |
* [ChaoticNeutrals/Hathor_RP-v.01-L3-8B](https://huggingface.co/ChaoticNeutrals/Hathor_RP-v.01-L3-8B)
|
65 |
* [Sao10K/L3-8B-Stheno-v3.1](https://huggingface.co/Sao10K/L3-8B-Stheno-v3.1)
|
66 |
+
* [aifeifei798/llama3-8B-DarkIdol-1.0](https://huggingface.co/aifeifei798/llama3-8B-DarkIdol-1.0)
|
67 |
* [ResplendentAI/Nymph_8B](https://huggingface.co/ResplendentAI/Nymph_8B)
|
|
|
|
|
68 |
|
69 |
+
### Quants
|
70 |
+
|
71 |
+
[Static quants](https://huggingface.co/mradermacher/L3-Uncen-Merger-Omelette-RP-v0.1-8B-GGUF) by mradermacher
|
72 |
+
|
73 |
+
## Secret Sauce
|
74 |
+
|
75 |
+
The following YAML configurations were used to produce this model:
|
76 |
+
|
77 |
+
### Scrambled-Egg-1
|
78 |
+
|
79 |
+
```yaml
|
80 |
+
models:
|
81 |
+
- model: Sao10K/L3-8B-Stheno-v3.2
|
82 |
+
- model: Casual-Autopsy/L3-Umbral-Mind-RP-v0.3-8B
|
83 |
+
parameters:
|
84 |
+
density: 0.45
|
85 |
+
weight: 0.33
|
86 |
+
- model: bluuwhale/L3-SthenoMaidBlackroot-8B-V1
|
87 |
+
parameters:
|
88 |
+
density: 0.75
|
89 |
+
weight: 0.33
|
90 |
+
merge_method: dare_ties
|
91 |
+
base_model: Sao10K/L3-8B-Stheno-v3.2
|
92 |
+
parameters:
|
93 |
+
int8_mask: true
|
94 |
+
dtype: bfloat16
|
95 |
+
```
|
96 |
|
97 |
+
### Scrambled-Egg-2
|
98 |
+
|
99 |
+
```yaml
|
100 |
+
models:
|
101 |
+
- model: [Unreleased psychology model]
|
102 |
+
- model: migtissera/Llama-3-8B-Synthia-v3.5
|
103 |
+
parameters:
|
104 |
+
density: 0.35
|
105 |
+
weight: 0.25
|
106 |
+
- model: tannedbum/L3-Nymeria-Maid-8B
|
107 |
+
parameters:
|
108 |
+
density: 0.65
|
109 |
+
weight: 0.25
|
110 |
+
merge_method: dare_ties
|
111 |
+
base_model: [Unreleased psychology model]
|
112 |
+
parameters:
|
113 |
+
int8_mask: true
|
114 |
+
dtype: bfloat16
|
115 |
+
```
|
116 |
+
|
117 |
+
### Scrambled-Egg-3
|
118 |
+
|
119 |
+
```yaml
|
120 |
+
models:
|
121 |
+
- model: Casual-Autopsy/L3-Umbral-Mind-RP-v1.0-8B
|
122 |
+
- model: tannedbum/L3-Nymeria-8B
|
123 |
+
parameters:
|
124 |
+
density: 0.5
|
125 |
+
weight: 0.35
|
126 |
+
- model: ChaoticNeutrals/Hathor_RP-v.01-L3-8B
|
127 |
+
parameters:
|
128 |
+
density: 0.4
|
129 |
+
weight: 0.2
|
130 |
+
merge_method: dare_ties
|
131 |
+
base_model: Casual-Autopsy/L3-Umbral-Mind-RP-v1.0-8B
|
132 |
+
parameters:
|
133 |
+
int8_mask: true
|
134 |
+
dtype: bfloat16
|
135 |
+
```
|
136 |
+
|
137 |
+
### Omelette-1
|
138 |
+
|
139 |
+
```yaml
|
140 |
+
models:
|
141 |
+
- model: Casual-Autopsy/Scrambled-Egg-1
|
142 |
+
- model: Casual-Autopsy/Scrambled-Egg-3
|
143 |
+
merge_method: slerp
|
144 |
+
base_model: Casual-Autopsy/Scrambled-Egg-1
|
145 |
+
parameters:
|
146 |
+
t:
|
147 |
+
- value: [0.1, 0.15, 0.2, 0.4, 0.6, 0.4, 0.2, 0.15, 0.1]
|
148 |
+
embed_slerp: true
|
149 |
+
dtype: bfloat16
|
150 |
+
```
|
151 |
+
|
152 |
+
### Omelette-2
|
153 |
+
|
154 |
+
```yaml
|
155 |
+
models:
|
156 |
+
- model: Casual-Autopsy/Omelette-1
|
157 |
+
- model: Casual-Autopsy/Scrambled-Egg-2
|
158 |
+
merge_method: slerp
|
159 |
+
base_model: Casual-Autopsy/Omelette-1
|
160 |
+
parameters:
|
161 |
+
t:
|
162 |
+
- value: [0.7, 0.5, 0.3, 0.25, 0.2, 0.25, 0.3, 0.5, 0.7]
|
163 |
+
embed_slerp: true
|
164 |
+
dtype: bfloat16
|
165 |
+
```
|
166 |
+
|
167 |
+
### L3-Uncen-Merger-Omelette-RP-v0.1-8B
|
168 |
|
169 |
```yaml
|
170 |
models:
|
|
|
190 |
merge_method: task_arithmetic
|
191 |
base_model: Casual-Autopsy/Omelette-2
|
192 |
dtype: bfloat16
|
193 |
+
```
|
|