Hugging Face's logo

Sorawiz
/
Qwen2.5-14B-Instinct-RP

Text Generation
Transformers
Safetensors
qwen2
mergekit
Merge
conversational
text-generation-inference
Model card Files
xet
Community
Sorawiz commited on
Commit
9a3d1cd
·
verified ·
1 Parent(s): d8f1623

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +152 -2
README.md CHANGED
@@ -2,7 +2,10 @@
2
  base_model:
3
  - Ttimofeyka/Tissint-14B-v1.2-128k-RP
4
  - SicariusSicariiStuff/Impish_QWEN_14B-1M
5
- - Sorawiz/Qwen2.5-14B-1M-Instinct
 
 
 
6
  - Sorawiz/Qwen2.5-14B-GCC
7
  library_name: transformers
8
  tags:
@@ -17,7 +20,7 @@ This is a merge of pre-trained language models created using [mergekit](https://
17
  ## Merge Details
18
  ### Merge Method
19
 
20
- This model was merged using the [DARE TIES](https://arxiv.org/abs/2311.03099) merge method using [Sorawiz/Qwen2.5-14B-1M-Instinct](https://huggingface.co/Sorawiz/Qwen2.5-14B-1M-Instinct) as a base.
21
 
22
  ### Models Merged
23
 
@@ -31,6 +34,153 @@ The following models were included in the merge:
31
  The following YAML configuration was used to produce this model:
32
 
33
  ```yaml
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
34
  merge_method: dare_ties
35
  base_model: Sorawiz/Qwen2.5-14B-1M-Instinct
36
  models:
 
2
  base_model:
3
  - Ttimofeyka/Tissint-14B-v1.2-128k-RP
4
  - SicariusSicariiStuff/Impish_QWEN_14B-1M
5
+ - Goekdeniz-Guelmez/Josiefied-Qwen2.5-14B-Instruct-abliterated-v4
6
+ - deepcogito/cogito-v1-preview-qwen-14B
7
+ - Sao10K/14B-Qwen2.5-Freya-x1
8
+ - Sao10K/14B-Qwen2.5-Kunou-v1
9
  - Sorawiz/Qwen2.5-14B-GCC
10
  library_name: transformers
11
  tags:
 
20
  ## Merge Details
21
  ### Merge Method
22
 
23
+ This model was merged using the [DARE TIES](https://arxiv.org/abs/2311.03099) merge method using Sorawiz/Qwen2.5-14B-1M-Instinct as a base.
24
 
25
  ### Models Merged
26
 
 
34
  The following YAML configuration was used to produce this model:
35
 
36
  ```yaml
37
+ name: Sorawiz/Qwen2.5-14B-Instinct-Base
38
+ merge_method: dare_ties
39
+ base_model: Goekdeniz-Guelmez/Josiefied-Qwen2.5-14B-Instruct-abliterated-v4
40
+ models:
41
+ - model: Goekdeniz-Guelmez/Josiefied-Qwen2.5-14B-Instruct-abliterated-v4
42
+ parameters:
43
+ weight: 0.3
44
+ - model: Ttimofeyka/Tissint-14B-v1.2-128k-RP
45
+ parameters:
46
+ weight: 0.7
47
+ parameters:
48
+ density: 1
49
+ tokenizer:
50
+ source: union
51
+ chat_template: auto
52
+
53
+ ---
54
+
55
+ name: Sorawiz/Qwen2.5-14B-Instincto
56
+ merge_method: dare_ties
57
+ base_model: deepcogito/cogito-v1-preview-qwen-14B
58
+ models:
59
+ - model: deepcogito/cogito-v1-preview-qwen-14B
60
+ parameters:
61
+ weight: 0.4
62
+ - model: Sorawiz/Qwen2.5-14B-Instinct-Base
63
+ parameters:
64
+ weight: 0.3
65
+ - model: Ttimofeyka/Tissint-14B-v1.2-128k-RP
66
+ parameters:
67
+ weight: 0.3
68
+ parameters:
69
+ density: 0.5
70
+ tokenizer:
71
+ source: union
72
+ chat_template: auto
73
+
74
+ ---
75
+
76
+ name: Sorawiz/Qwen2.5-14B-Kunousint
77
+ merge_method: dare_ties
78
+ base_model: Sao10K/14B-Qwen2.5-Kunou-v1
79
+ models:
80
+ - model: Sao10K/14B-Qwen2.5-Kunou-v1
81
+ parameters:
82
+ weight: 0.5
83
+ - model: Sorawiz/Qwen2.5-14B-Instincto
84
+ parameters:
85
+ weight: 0.3
86
+ - model: Ttimofeyka/Tissint-14B-v1.2-128k-RP
87
+ parameters:
88
+ weight: 0.2
89
+ parameters:
90
+ density: 0.5
91
+ tokenizer:
92
+ source: union
93
+ chat_template: auto
94
+
95
+ ---
96
+
97
+ name: Sorawiz/Qwen2.5-14B-Kunousint-1M
98
+ merge_method: dare_ties
99
+ base_model: Sorawiz/Qwen2.5-14B-Imstinct
100
+ models:
101
+ - model: Sorawiz/Qwen2.5-14B-Imstinct
102
+ parameters:
103
+ weight: 0.2
104
+ - model: Sorawiz/Qwen2.5-14B-Kunousint
105
+ parameters:
106
+ weight: 0.5
107
+ - model: Sao10K/14B-Qwen2.5-Kunou-v1
108
+ parameters:
109
+ weight: 0.3
110
+ parameters:
111
+ density: 0.5
112
+ tokenizer:
113
+ source: union
114
+ chat_template: auto
115
+
116
+ ---
117
+
118
+ name: Sorawiz/Qwen2.5-14B-Frayasint
119
+ merge_method: dare_ties
120
+ base_model: Sao10K/14B-Qwen2.5-Freya-x1
121
+ models:
122
+ - model: Sao10K/14B-Qwen2.5-Freya-x1
123
+ parameters:
124
+ weight: 0.5
125
+ - model: Sorawiz/Qwen2.5-14B-Instincto
126
+ parameters:
127
+ weight: 0.3
128
+ - model: Ttimofeyka/Tissint-14B-v1.2-128k-RP
129
+ parameters:
130
+ weight: 0.2
131
+ parameters:
132
+ density: 0.5
133
+ tokenizer:
134
+ source: union
135
+ chat_template: auto
136
+
137
+ ---
138
+
139
+ name: Sorawiz/Qwen2.5-14B-Frayasint-1M
140
+ merge_method: dare_ties
141
+ base_model: Sorawiz/Qwen2.5-14B-Imstinct
142
+ models:
143
+ - model: Sorawiz/Qwen2.5-14B-Imstinct
144
+ parameters:
145
+ weight: 0.2
146
+ - model: Sorawiz/Qwen2.5-14B-Frayasint
147
+ parameters:
148
+ weight: 0.5
149
+ - model: Sao10K/14B-Qwen2.5-Freya-x1
150
+ parameters:
151
+ weight: 0.3
152
+ parameters:
153
+ density: 0.5
154
+ tokenizer:
155
+ source: union
156
+ chat_template: auto
157
+
158
+ ---
159
+
160
+ name: Sorawiz/Qwen2.5-14B-1M-Instinct
161
+ merge_method: dare_ties
162
+ base_model: Sorawiz/Qwen2.5-14B-Imstinct
163
+ models:
164
+ - model: Sorawiz/Qwen2.5-14B-Imstinct
165
+ parameters:
166
+ weight: 0.25
167
+ - model: Sorawiz/Qwen2.5-14B-1M-Kunousint-1M
168
+ parameters:
169
+ weight: 0.25
170
+ - model: Sorawiz/Qwen2.5-14B-Frayasint-1M
171
+ parameters:
172
+ weight: 0.25
173
+ - model: Ttimofeyka/Tissint-14B-v1.2-128k-RP
174
+ parameters:
175
+ weight: 0.25
176
+ parameters:
177
+ density: 1
178
+ tokenizer:
179
+ source: union
180
+ chat_template: auto
181
+
182
+ ---
183
+
184
  merge_method: dare_ties
185
  base_model: Sorawiz/Qwen2.5-14B-1M-Instinct
186
  models: