Upload nqlsg-14b.yaml
Browse filesAdd mergekit-multi configuration file
- nqlsg-14b.yaml +152 -0
nqlsg-14b.yaml
ADDED
@@ -0,0 +1,152 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
name: NQLSG-Qwen2.5-14B-Base1
|
2 |
+
models:
|
3 |
+
- model: sthenno/tempesthenno-nuslerp-0124
|
4 |
+
- model: sthenno/tempesthenno-ppo-ckpt40
|
5 |
+
- model: sthenno-com/miscii-14b-1225
|
6 |
+
- model: jpacifico/Chocolatine-2-14B-Instruct-v2.0b3
|
7 |
+
- model: jpacifico/Chocolatine-2-14B-Instruct-v2.0.3
|
8 |
+
- model: sometimesanotion/Lamarck-14B-v0.6
|
9 |
+
- model: sometimesanotion/Lamarck-14B-v0.7
|
10 |
+
- model: sometimesanotion/Qwen2.5-14B-Vimarckoso-v3
|
11 |
+
- model: sometimesanotion/Qwenvergence-14B-v11
|
12 |
+
- model: sometimesanotion/Qwenvergence-14B-v12-Prose-DS
|
13 |
+
- model: sometimesanotion/Qwenvergence-14B-v13-Prose-DS
|
14 |
+
base_model: jpacifico/Chocolatine-2-14B-Instruct-v2.0.3
|
15 |
+
chat_template: auto
|
16 |
+
dtype: bfloat16
|
17 |
+
merge_method: ties
|
18 |
+
parameters:
|
19 |
+
weight: 1.0
|
20 |
+
density: 1.0
|
21 |
+
int8_mask: true
|
22 |
+
tokenizer:
|
23 |
+
source: union
|
24 |
+
---
|
25 |
+
name: NQLSG-Qwen2.5-14B-Base2
|
26 |
+
models:
|
27 |
+
- model: sthenno/tempesthenno-nuslerp-0124
|
28 |
+
- model: sthenno/tempesthenno-ppo-ckpt40
|
29 |
+
- model: sthenno-com/miscii-14b-1225
|
30 |
+
- model: jpacifico/Chocolatine-2-14B-Instruct-v2.0b3
|
31 |
+
- model: jpacifico/Chocolatine-2-14B-Instruct-v2.0.3
|
32 |
+
- model: sometimesanotion/Lamarck-14B-v0.6
|
33 |
+
- model: sometimesanotion/Lamarck-14B-v0.7
|
34 |
+
- model: sometimesanotion/Qwen2.5-14B-Vimarckoso-v3
|
35 |
+
- model: sometimesanotion/Qwenvergence-14B-v11
|
36 |
+
- model: sometimesanotion/Qwenvergence-14B-v12-Prose-DS
|
37 |
+
- model: sometimesanotion/Qwenvergence-14B-v13-Prose-DS
|
38 |
+
base_model: jpacifico/Chocolatine-2-14B-Instruct-v2.0.3
|
39 |
+
chat_template: auto
|
40 |
+
dtype: bfloat16
|
41 |
+
merge_method: sce
|
42 |
+
parameters:
|
43 |
+
int8_mask: true
|
44 |
+
tokenizer:
|
45 |
+
source: union
|
46 |
+
---
|
47 |
+
name: NQLSG-Qwen2.5-14B-Base3
|
48 |
+
models:
|
49 |
+
- model: sthenno/tempesthenno-nuslerp-0124
|
50 |
+
- model: sthenno/tempesthenno-ppo-ckpt40
|
51 |
+
- model: sthenno-com/miscii-14b-1225
|
52 |
+
- model: jpacifico/Chocolatine-2-14B-Instruct-v2.0b3
|
53 |
+
- model: jpacifico/Chocolatine-2-14B-Instruct-v2.0.3
|
54 |
+
- model: sometimesanotion/Lamarck-14B-v0.6
|
55 |
+
- model: sometimesanotion/Lamarck-14B-v0.7
|
56 |
+
- model: sometimesanotion/Qwen2.5-14B-Vimarckoso-v3
|
57 |
+
- model: sometimesanotion/Qwenvergence-14B-v11
|
58 |
+
- model: sometimesanotion/Qwenvergence-14B-v12-Prose-DS
|
59 |
+
- model: sometimesanotion/Qwenvergence-14B-v13-Prose-DS
|
60 |
+
base_model: jpacifico/Chocolatine-2-14B-Instruct-v2.0.3
|
61 |
+
chat_template: auto
|
62 |
+
dtype: bfloat16
|
63 |
+
merge_method: model_stock
|
64 |
+
parameters:
|
65 |
+
int8_mask: true
|
66 |
+
tokenizer:
|
67 |
+
source: union
|
68 |
+
---
|
69 |
+
name: NQLSG-Qwen2.5-14B-Base4
|
70 |
+
models:
|
71 |
+
- model: sthenno/tempesthenno-nuslerp-0124
|
72 |
+
- model: sthenno/tempesthenno-ppo-ckpt40
|
73 |
+
- model: sthenno-com/miscii-14b-1225
|
74 |
+
- model: jpacifico/Chocolatine-2-14B-Instruct-v2.0b3
|
75 |
+
- model: jpacifico/Chocolatine-2-14B-Instruct-v2.0.3
|
76 |
+
- model: sometimesanotion/Lamarck-14B-v0.6
|
77 |
+
- model: sometimesanotion/Lamarck-14B-v0.7
|
78 |
+
- model: sometimesanotion/Qwen2.5-14B-Vimarckoso-v3
|
79 |
+
- model: sometimesanotion/Qwenvergence-14B-v11
|
80 |
+
- model: sometimesanotion/Qwenvergence-14B-v12-Prose-DS
|
81 |
+
- model: sometimesanotion/Qwenvergence-14B-v13-Prose-DS
|
82 |
+
base_model: sometimesanotion/Lamarck-14B-v0.6
|
83 |
+
chat_template: auto
|
84 |
+
dtype: bfloat16
|
85 |
+
merge_method: ties
|
86 |
+
parameters:
|
87 |
+
weight: 1.0
|
88 |
+
density: 1.0
|
89 |
+
int8_mask: true
|
90 |
+
tokenizer:
|
91 |
+
source: union
|
92 |
+
---
|
93 |
+
name: NQLSG-Qwen2.5-14B-Base5
|
94 |
+
models:
|
95 |
+
- model: sthenno/tempesthenno-nuslerp-0124
|
96 |
+
- model: sthenno/tempesthenno-ppo-ckpt40
|
97 |
+
- model: sthenno-com/miscii-14b-1225
|
98 |
+
- model: jpacifico/Chocolatine-2-14B-Instruct-v2.0b3
|
99 |
+
- model: jpacifico/Chocolatine-2-14B-Instruct-v2.0.3
|
100 |
+
- model: sometimesanotion/Lamarck-14B-v0.6
|
101 |
+
- model: sometimesanotion/Lamarck-14B-v0.7
|
102 |
+
- model: sometimesanotion/Qwen2.5-14B-Vimarckoso-v3
|
103 |
+
- model: sometimesanotion/Qwenvergence-14B-v11
|
104 |
+
- model: sometimesanotion/Qwenvergence-14B-v12-Prose-DS
|
105 |
+
- model: sometimesanotion/Qwenvergence-14B-v13-Prose-DS
|
106 |
+
base_model: sometimesanotion/Lamarck-14B-v0.6
|
107 |
+
chat_template: auto
|
108 |
+
dtype: bfloat16
|
109 |
+
merge_method: sce
|
110 |
+
parameters:
|
111 |
+
int8_mask: true
|
112 |
+
tokenizer:
|
113 |
+
source: union
|
114 |
+
---
|
115 |
+
name: NQLSG-Qwen2.5-14B-Base6
|
116 |
+
models:
|
117 |
+
- model: sthenno/tempesthenno-nuslerp-0124
|
118 |
+
- model: sthenno/tempesthenno-ppo-ckpt40
|
119 |
+
- model: sthenno-com/miscii-14b-1225
|
120 |
+
- model: jpacifico/Chocolatine-2-14B-Instruct-v2.0b3
|
121 |
+
- model: jpacifico/Chocolatine-2-14B-Instruct-v2.0.3
|
122 |
+
- model: sometimesanotion/Lamarck-14B-v0.6
|
123 |
+
- model: sometimesanotion/Lamarck-14B-v0.7
|
124 |
+
- model: sometimesanotion/Qwen2.5-14B-Vimarckoso-v3
|
125 |
+
- model: sometimesanotion/Qwenvergence-14B-v11
|
126 |
+
- model: sometimesanotion/Qwenvergence-14B-v12-Prose-DS
|
127 |
+
- model: sometimesanotion/Qwenvergence-14B-v13-Prose-DS
|
128 |
+
base_model: sometimesanotion/Lamarck-14B-v0.6
|
129 |
+
chat_template: auto
|
130 |
+
dtype: bfloat16
|
131 |
+
merge_method: model_stock
|
132 |
+
parameters:
|
133 |
+
int8_mask: true
|
134 |
+
tokenizer:
|
135 |
+
source: union
|
136 |
+
---
|
137 |
+
name: NQLSG-Qwen2.5-14B-MegaFusion
|
138 |
+
models:
|
139 |
+
- model: NQLSG-Qwen2.5-14B-Base1
|
140 |
+
- model: NQLSG-Qwen2.5-14B-Base2
|
141 |
+
- model: NQLSG-Qwen2.5-14B-Base3
|
142 |
+
- model: NQLSG-Qwen2.5-14B-Base4
|
143 |
+
- model: NQLSG-Qwen2.5-14B-Base5
|
144 |
+
- model: NQLSG-Qwen2.5-14B-Base6
|
145 |
+
base_model: NQLSG-Qwen2.5-14B-Base5
|
146 |
+
chat_template: auto
|
147 |
+
dtype: bfloat16
|
148 |
+
merge_method: sce
|
149 |
+
parameters:
|
150 |
+
int8_mask: true
|
151 |
+
tokenizer:
|
152 |
+
source: union
|