Hugging Face's logo

RichardErkhov
/
v000000_-_SwallowMaid-8B-L3-SPPO-abliterated-gguf

GGUF
conversational
Model card Files Community
RichardErkhov commited on
Commit
d22a5a6
·
verified ·
1 Parent(s): a09b92a

uploaded readme

Browse files
Files changed (1) hide show
  1. README.md +169 -0
README.md ADDED
@@ -0,0 +1,169 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Quantization made by Richard Erkhov.
2
+
3
+ [Github](https://github.com/RichardErkhov)
4
+
5
+ [Discord](https://discord.gg/pvy7H8DZMG)
6
+
7
+ [Request more models](https://github.com/RichardErkhov/quant_request)
8
+
9
+
10
+ SwallowMaid-8B-L3-SPPO-abliterated - GGUF
11
+ - Model creator: https://huggingface.co/v000000/
12
+ - Original model: https://huggingface.co/v000000/SwallowMaid-8B-L3-SPPO-abliterated/
13
+
14
+
15
+ | Name | Quant method | Size |
16
+ | ---- | ---- | ---- |
17
+ | [SwallowMaid-8B-L3-SPPO-abliterated.Q2_K.gguf](https://huggingface.co/RichardErkhov/v000000_-_SwallowMaid-8B-L3-SPPO-abliterated-gguf/blob/main/SwallowMaid-8B-L3-SPPO-abliterated.Q2_K.gguf) | Q2_K | 2.96GB |
18
+ | [SwallowMaid-8B-L3-SPPO-abliterated.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/v000000_-_SwallowMaid-8B-L3-SPPO-abliterated-gguf/blob/main/SwallowMaid-8B-L3-SPPO-abliterated.IQ3_XS.gguf) | IQ3_XS | 3.28GB |
19
+ | [SwallowMaid-8B-L3-SPPO-abliterated.IQ3_S.gguf](https://huggingface.co/RichardErkhov/v000000_-_SwallowMaid-8B-L3-SPPO-abliterated-gguf/blob/main/SwallowMaid-8B-L3-SPPO-abliterated.IQ3_S.gguf) | IQ3_S | 3.43GB |
20
+ | [SwallowMaid-8B-L3-SPPO-abliterated.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/v000000_-_SwallowMaid-8B-L3-SPPO-abliterated-gguf/blob/main/SwallowMaid-8B-L3-SPPO-abliterated.Q3_K_S.gguf) | Q3_K_S | 3.41GB |
21
+ | [SwallowMaid-8B-L3-SPPO-abliterated.IQ3_M.gguf](https://huggingface.co/RichardErkhov/v000000_-_SwallowMaid-8B-L3-SPPO-abliterated-gguf/blob/main/SwallowMaid-8B-L3-SPPO-abliterated.IQ3_M.gguf) | IQ3_M | 3.52GB |
22
+ | [SwallowMaid-8B-L3-SPPO-abliterated.Q3_K.gguf](https://huggingface.co/RichardErkhov/v000000_-_SwallowMaid-8B-L3-SPPO-abliterated-gguf/blob/main/SwallowMaid-8B-L3-SPPO-abliterated.Q3_K.gguf) | Q3_K | 3.74GB |
23
+ | [SwallowMaid-8B-L3-SPPO-abliterated.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/v000000_-_SwallowMaid-8B-L3-SPPO-abliterated-gguf/blob/main/SwallowMaid-8B-L3-SPPO-abliterated.Q3_K_M.gguf) | Q3_K_M | 3.74GB |
24
+ | [SwallowMaid-8B-L3-SPPO-abliterated.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/v000000_-_SwallowMaid-8B-L3-SPPO-abliterated-gguf/blob/main/SwallowMaid-8B-L3-SPPO-abliterated.Q3_K_L.gguf) | Q3_K_L | 4.03GB |
25
+ | [SwallowMaid-8B-L3-SPPO-abliterated.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/v000000_-_SwallowMaid-8B-L3-SPPO-abliterated-gguf/blob/main/SwallowMaid-8B-L3-SPPO-abliterated.IQ4_XS.gguf) | IQ4_XS | 4.18GB |
26
+ | [SwallowMaid-8B-L3-SPPO-abliterated.Q4_0.gguf](https://huggingface.co/RichardErkhov/v000000_-_SwallowMaid-8B-L3-SPPO-abliterated-gguf/blob/main/SwallowMaid-8B-L3-SPPO-abliterated.Q4_0.gguf) | Q4_0 | 4.34GB |
27
+ | [SwallowMaid-8B-L3-SPPO-abliterated.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/v000000_-_SwallowMaid-8B-L3-SPPO-abliterated-gguf/blob/main/SwallowMaid-8B-L3-SPPO-abliterated.IQ4_NL.gguf) | IQ4_NL | 4.38GB |
28
+ | [SwallowMaid-8B-L3-SPPO-abliterated.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/v000000_-_SwallowMaid-8B-L3-SPPO-abliterated-gguf/blob/main/SwallowMaid-8B-L3-SPPO-abliterated.Q4_K_S.gguf) | Q4_K_S | 4.37GB |
29
+ | [SwallowMaid-8B-L3-SPPO-abliterated.Q4_K.gguf](https://huggingface.co/RichardErkhov/v000000_-_SwallowMaid-8B-L3-SPPO-abliterated-gguf/blob/main/SwallowMaid-8B-L3-SPPO-abliterated.Q4_K.gguf) | Q4_K | 4.58GB |
30
+ | [SwallowMaid-8B-L3-SPPO-abliterated.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/v000000_-_SwallowMaid-8B-L3-SPPO-abliterated-gguf/blob/main/SwallowMaid-8B-L3-SPPO-abliterated.Q4_K_M.gguf) | Q4_K_M | 4.58GB |
31
+ | [SwallowMaid-8B-L3-SPPO-abliterated.Q4_1.gguf](https://huggingface.co/RichardErkhov/v000000_-_SwallowMaid-8B-L3-SPPO-abliterated-gguf/blob/main/SwallowMaid-8B-L3-SPPO-abliterated.Q4_1.gguf) | Q4_1 | 4.78GB |
32
+ | [SwallowMaid-8B-L3-SPPO-abliterated.Q5_0.gguf](https://huggingface.co/RichardErkhov/v000000_-_SwallowMaid-8B-L3-SPPO-abliterated-gguf/blob/main/SwallowMaid-8B-L3-SPPO-abliterated.Q5_0.gguf) | Q5_0 | 5.21GB |
33
+ | [SwallowMaid-8B-L3-SPPO-abliterated.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/v000000_-_SwallowMaid-8B-L3-SPPO-abliterated-gguf/blob/main/SwallowMaid-8B-L3-SPPO-abliterated.Q5_K_S.gguf) | Q5_K_S | 5.21GB |
34
+ | [SwallowMaid-8B-L3-SPPO-abliterated.Q5_K.gguf](https://huggingface.co/RichardErkhov/v000000_-_SwallowMaid-8B-L3-SPPO-abliterated-gguf/blob/main/SwallowMaid-8B-L3-SPPO-abliterated.Q5_K.gguf) | Q5_K | 5.34GB |
35
+ | [SwallowMaid-8B-L3-SPPO-abliterated.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/v000000_-_SwallowMaid-8B-L3-SPPO-abliterated-gguf/blob/main/SwallowMaid-8B-L3-SPPO-abliterated.Q5_K_M.gguf) | Q5_K_M | 5.34GB |
36
+ | [SwallowMaid-8B-L3-SPPO-abliterated.Q5_1.gguf](https://huggingface.co/RichardErkhov/v000000_-_SwallowMaid-8B-L3-SPPO-abliterated-gguf/blob/main/SwallowMaid-8B-L3-SPPO-abliterated.Q5_1.gguf) | Q5_1 | 5.65GB |
37
+ | [SwallowMaid-8B-L3-SPPO-abliterated.Q6_K.gguf](https://huggingface.co/RichardErkhov/v000000_-_SwallowMaid-8B-L3-SPPO-abliterated-gguf/blob/main/SwallowMaid-8B-L3-SPPO-abliterated.Q6_K.gguf) | Q6_K | 6.14GB |
38
+ | [SwallowMaid-8B-L3-SPPO-abliterated.Q8_0.gguf](https://huggingface.co/RichardErkhov/v000000_-_SwallowMaid-8B-L3-SPPO-abliterated-gguf/blob/main/SwallowMaid-8B-L3-SPPO-abliterated.Q8_0.gguf) | Q8_0 | 7.95GB |
39
+
40
+
41
+
42
+
43
+ Original model description:
44
+ ---
45
+ base_model:
46
+ - x0000001/mergekit-task_arithmetic-vlehhex
47
+ - grimjim/Llama-3-Instruct-abliteration-LoRA-8B
48
+ library_name: transformers
49
+ tags:
50
+ - mergekit
51
+ - merge
52
+ - english
53
+ - japanese
54
+ - llama
55
+ ---
56
+ <!DOCTYPE html>
57
+ <style>
58
+
59
+ h1 {
60
+ color: #900C3F; /* Red color */
61
+ font-size: 1.25em; /* Larger font size */
62
+ text-align: left; /* Center alignment */
63
+ text-shadow: 2px 2px 4px rgba(0, 0, 0, 0.5); /* Shadow effect */
64
+ background: linear-gradient(90deg, #900C3F, #fba8a8); /* Gradient background */
65
+ -webkit-background-clip: text; /* Clipping the background to text */
66
+ -webkit-text-fill-color: transparent; /* Making the text transparent */
67
+ }
68
+
69
+ </style>
70
+ <html lang="en">
71
+ <head>
72
+ </head>
73
+ <body>
74
+ General Instruct, RP, Q&A and Storywriting.
75
+
76
+ -------------------------------------------
77
+ <h1>SwallowMaid-8B-Llama-3-SPPO-abliterated</h1>
78
+
79
+ ![image/png](https://cdn-uploads.huggingface.co/production/uploads/64f74b6e6389380c77562762/j0i4gIFnxGOdgyINopKUW.png)
80
+
81
+ "Llama-3-Instruct-8B-SPPO-Iter3" fully uncensored with 35% RP-Mix infused vector direction to gain some roleplay capabilities and prose while attempting to preserve the qualities of Meta's Llama-3-Instruct finetune.
82
+
83
+ # <a>Thank you mradermacher for the quants!</a>
84
+ * [GGUFs](https://huggingface.co/mradermacher/SwallowMaid-8B-L3-SPPO-abliterated-GGUF)
85
+ * [GGUFs imatrix](https://huggingface.co/mradermacher/SwallowMaid-8B-L3-SPPO-abliterated-i1-GGUF)
86
+
87
+ # <a>Quants</a>
88
+ * [GGUF Q8_0](https://huggingface.co/v000000/SwallowMaid-8B-L3-SPPO-abliterated-Q8_0-GGUF)
89
+
90
+ # <h1>merge</h1>
91
+
92
+ This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
93
+
94
+ # <h1>Merge Details</h1>
95
+ # <h1>Merge Method</h1>
96
+
97
+ This model was merged using a multi-step merge method.
98
+
99
+ # <h1>Models Merged</h1>
100
+
101
+ The following models were included in the merge:
102
+ * [grimjim/Llama-3-Instruct-abliteration-LoRA-8B](https://huggingface.co/grimjim/Llama-3-Instruct-abliteration-LoRA-8B)
103
+ * [UCLA-AGI/Llama-3-Instruct-8B-SPPO-Iter3](https://huggingface.co/UCLA-AGI/Llama-3-Instruct-8B-SPPO-Iter3)
104
+ * [NeverSleep/Llama-3-Lumimaid-8B-v0.1-OAS](https://huggingface.co/NeverSleep/Llama-3-Lumimaid-8B-v0.1-OAS)
105
+ * [maldv/llama-3-fantasy-writer-8b](https://huggingface.co/maldv/llama-3-fantasy-writer-8b)
106
+ * [tokyotech-llm/Llama-3-Swallow-8B-v0.1](https://huggingface.co/tokyotech-llm/Llama-3-Swallow-8B-v0.1)
107
+ * [Nitral-AI/Hathor_Respawn-L3-8B-v0.8](https://huggingface.co/Nitral-AI/Hathor_Respawn-L3-8B-v0.8)
108
+
109
+ # <h1>Configuration</h1>
110
+
111
+ The following YAML configuration was used to produce this model:
112
+
113
+ ```yaml
114
+ # Part 3, Apply abliteration (SwallowMaid-8B)
115
+ models:
116
+ - model: sppo-rpmix-part2+grimjim/Llama-3-Instruct-abliteration-LoRA-8B
117
+ parameters:
118
+ weight: 1.0
119
+ merge_method: linear
120
+ dtype: float32
121
+
122
+ # Part 2, infuse 35% swallow+rpmix to SPPO-Iter3 (sppo-rpmix-part2)
123
+ models:
124
+ - model: UCLA-AGI/Llama-3-Instruct-8B-SPPO-Iter3
125
+ parameters:
126
+ weight: 1.0
127
+ - model: rpmix-part1
128
+ parameters:
129
+ weight: 0.35
130
+ merge_method: task_arithmetic
131
+ base_model: UCLA-AGI/Llama-3-Instruct-8B-SPPO-Iter3
132
+ parameters:
133
+ normalize: false
134
+ dtype: float32
135
+
136
+ # Part 1, linear merge rpmix (rpmix-part1)
137
+ models:
138
+ - model: Nitral-AI/Hathor_Respawn-L3-8B-v0.8
139
+ parameters:
140
+ weight: 0.6
141
+ - model: maldv/llama-3-fantasy-writer-8b
142
+ parameters:
143
+ weight: 0.1
144
+ - model: NeverSleep/Llama-3-Lumimaid-8B-v0.1-OAS
145
+ parameters:
146
+ weight: 0.4
147
+ - model: tokyotech-llm/Llama-3-Swallow-8B-v0.1
148
+ parameters:
149
+ weight: 0.15
150
+ merge_method: linear
151
+ dtype: float32
152
+ ```
153
+
154
+ # <h1>Prompt Template:</h1>
155
+ ```bash
156
+ <|begin_of_text|><|start_header_id|>system<|end_header_id|>
157
+
158
+ {system_prompt}<|eot_id|><|start_header_id|>user<|end_header_id|>
159
+
160
+ {input}<|eot_id|><|start_header_id|>assistant<|end_header_id|>
161
+
162
+ {output}<|eot_id|>
163
+
164
+ ```
165
+
166
+
167
+ </body>
168
+ </html>
169
+