Update README.md
Browse files
README.md
CHANGED
@@ -8,11 +8,16 @@ library_name: transformers
|
|
8 |
tags:
|
9 |
- mergekit
|
10 |
- merge
|
|
|
11 |
|
12 |
---
|
|
|
|
|
|
|
|
|
13 |
# merged
|
14 |
|
15 |
-
This is a merge of pre-trained language models created using [mergekit](https://github.com/
|
16 |
|
17 |
## Merge Details
|
18 |
### Merge Method
|
@@ -39,7 +44,7 @@ slices:
|
|
39 |
- layer_range: [0, 64]
|
40 |
model: maldv/Qwentile2.5-32B-Instruct
|
41 |
parameters:
|
42 |
-
diagonal_offset: 2.0
|
43 |
random_mask: 0.666
|
44 |
random_mask_seed: 888.0
|
45 |
weight: 0.5
|
@@ -51,7 +56,7 @@ slices:
|
|
51 |
- layer_range: [0, 64]
|
52 |
model: Sao10K/32B-Qwen2.5-Kunou-v1
|
53 |
parameters:
|
54 |
-
diagonal_offset: 2.0
|
55 |
random_mask: 0.333
|
56 |
random_mask_seed: 12347888.0
|
57 |
weight: 0.5
|
|
|
8 |
tags:
|
9 |
- mergekit
|
10 |
- merge
|
11 |
+
license: apache-2.0
|
12 |
|
13 |
---
|
14 |
+
|
15 |
+
# Info
|
16 |
+
Trying to make something different, feel free to try like or dislike or leave a feedback, I'm not claiming anything about anything.
|
17 |
+
|
18 |
# merged
|
19 |
|
20 |
+
This is a merge of pre-trained language models created using my custom method in [mergekit](https://github.com/Ar57m/mergekit/tree/swapping).
|
21 |
|
22 |
## Merge Details
|
23 |
### Merge Method
|
|
|
44 |
- layer_range: [0, 64]
|
45 |
model: maldv/Qwentile2.5-32B-Instruct
|
46 |
parameters:
|
47 |
+
diagonal_offset: 2.0 # ignored here
|
48 |
random_mask: 0.666
|
49 |
random_mask_seed: 888.0
|
50 |
weight: 0.5
|
|
|
56 |
- layer_range: [0, 64]
|
57 |
model: Sao10K/32B-Qwen2.5-Kunou-v1
|
58 |
parameters:
|
59 |
+
diagonal_offset: 2.0 # ignored here
|
60 |
random_mask: 0.333
|
61 |
random_mask_seed: 12347888.0
|
62 |
weight: 0.5
|