bamec66557
commited on
Update README.md
Browse files
README.md
CHANGED
@@ -1,66 +1,66 @@
|
|
1 |
-
---
|
2 |
-
|
3 |
-
|
4 |
-
|
5 |
-
-
|
6 |
-
-
|
7 |
-
|
8 |
-
---
|
9 |
-
# Z-2-A.TEST-TEMP-MODEL
|
10 |
-
|
11 |
-
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
|
12 |
-
|
13 |
-
## Merge Details
|
14 |
-
### Merge Method
|
15 |
-
|
16 |
-
This model was merged using the SLERP merge method.
|
17 |
-
|
18 |
-
### Models Merged
|
19 |
-
|
20 |
-
The following models were included in the merge:
|
21 |
-
* D:\VICIOUS_MESH-12B-OMEGA
|
22 |
-
* D:\jetreessence
|
23 |
-
|
24 |
-
### Configuration
|
25 |
-
|
26 |
-
The following YAML configuration was used to produce this model:
|
27 |
-
|
28 |
-
```yaml
|
29 |
-
models:
|
30 |
-
- model: "D:\\VICIOUS_MESH-12B-OMEGA"
|
31 |
-
- model: "D:\\jetreessence"
|
32 |
-
merge_method: slerp
|
33 |
-
base_model: "D:\\VICIOUS_MESH-12B-OMEGA"
|
34 |
-
dtype: bfloat16
|
35 |
-
parameters:
|
36 |
-
t: [0, 0.5, 1, 0.5, 0]
|
37 |
-
|
38 |
-
regularization:
|
39 |
-
- method: gradient_penalty
|
40 |
-
scale: 0.05
|
41 |
-
- method: weight_clipping
|
42 |
-
clip_range: [-0.15, 0.15]
|
43 |
-
- method: random_noise
|
44 |
-
scale: 0.01
|
45 |
-
- method: attention_dropout
|
46 |
-
scale: 0.02
|
47 |
-
|
48 |
-
postprocessing:
|
49 |
-
- operation: entropy_regularization
|
50 |
-
scale: 0.05
|
51 |
-
- operation: non_linear_scaling
|
52 |
-
parameters:
|
53 |
-
function: relu
|
54 |
-
- operation: sharpening
|
55 |
-
intensity: 0.6
|
56 |
-
- operation: gaussian_smoothing
|
57 |
-
sigma: 0.3
|
58 |
-
- operation: normalize
|
59 |
-
- operation: dynamic_scaling
|
60 |
-
scale_range: [0.98, 1.02]
|
61 |
-
- operation: smoothing
|
62 |
-
parameters:
|
63 |
-
adaptive: true
|
64 |
-
range: [0.98, 1.02]
|
65 |
-
kernel_size: 3
|
66 |
-
```
|
|
|
1 |
+
---
|
2 |
+
library_name: transformers
|
3 |
+
tags:
|
4 |
+
- mergekit
|
5 |
+
- merge
|
6 |
+
- not-for-all-audiences
|
7 |
+
license: apache-2.0
|
8 |
+
---
|
9 |
+
# Z-2-A.TEST-TEMP-MODEL
|
10 |
+
|
11 |
+
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
|
12 |
+
|
13 |
+
## Merge Details
|
14 |
+
### Merge Method
|
15 |
+
|
16 |
+
This model was merged using the SLERP merge method.
|
17 |
+
|
18 |
+
### Models Merged
|
19 |
+
|
20 |
+
The following models were included in the merge:
|
21 |
+
* D:\VICIOUS_MESH-12B-OMEGA
|
22 |
+
* D:\jetreessence
|
23 |
+
|
24 |
+
### Configuration
|
25 |
+
|
26 |
+
The following YAML configuration was used to produce this model:
|
27 |
+
|
28 |
+
```yaml
|
29 |
+
models:
|
30 |
+
- model: "D:\\VICIOUS_MESH-12B-OMEGA"
|
31 |
+
- model: "D:\\jetreessence"
|
32 |
+
merge_method: slerp
|
33 |
+
base_model: "D:\\VICIOUS_MESH-12B-OMEGA"
|
34 |
+
dtype: bfloat16
|
35 |
+
parameters:
|
36 |
+
t: [0, 0.5, 1, 0.5, 0]
|
37 |
+
|
38 |
+
regularization:
|
39 |
+
- method: gradient_penalty
|
40 |
+
scale: 0.05
|
41 |
+
- method: weight_clipping
|
42 |
+
clip_range: [-0.15, 0.15]
|
43 |
+
- method: random_noise
|
44 |
+
scale: 0.01
|
45 |
+
- method: attention_dropout
|
46 |
+
scale: 0.02
|
47 |
+
|
48 |
+
postprocessing:
|
49 |
+
- operation: entropy_regularization
|
50 |
+
scale: 0.05
|
51 |
+
- operation: non_linear_scaling
|
52 |
+
parameters:
|
53 |
+
function: relu
|
54 |
+
- operation: sharpening
|
55 |
+
intensity: 0.6
|
56 |
+
- operation: gaussian_smoothing
|
57 |
+
sigma: 0.3
|
58 |
+
- operation: normalize
|
59 |
+
- operation: dynamic_scaling
|
60 |
+
scale_range: [0.98, 1.02]
|
61 |
+
- operation: smoothing
|
62 |
+
parameters:
|
63 |
+
adaptive: true
|
64 |
+
range: [0.98, 1.02]
|
65 |
+
kernel_size: 3
|
66 |
+
```
|