Commit
•
c7100ef
1
Parent(s):
3d16d12
Update README.md
Browse files
README.md
CHANGED
@@ -10,7 +10,9 @@ license: apache-2.0
|
|
10 |
|
11 |
This model was made by merging models based on Mistral with the SLERP merge method.
|
12 |
|
13 |
-
All model's weights were merged using the SLERP method. More information below.
|
|
|
|
|
14 |
|
15 |
Spherical Linear Interpolation (SLERP)
|
16 |
Traditionally, model merging often resorts to weight averaging which, although straightforward, might not always capture the intricate features of the models being merged. The SLERP technique addresses this limitation, producing a blended model with characteristics smoothly interpolated from both parent models, ensuring the resultant model captures the essence of both its parents.
|
|
|
10 |
|
11 |
This model was made by merging models based on Mistral with the SLERP merge method.
|
12 |
|
13 |
+
All model's weights were merged using the SLERP method. More information below.
|
14 |
+
|
15 |
+
Advantages of the SLERP method vs averaging weights are as follows:
|
16 |
|
17 |
Spherical Linear Interpolation (SLERP)
|
18 |
Traditionally, model merging often resorts to weight averaging which, although straightforward, might not always capture the intricate features of the models being merged. The SLERP technique addresses this limitation, producing a blended model with characteristics smoothly interpolated from both parent models, ensuring the resultant model captures the essence of both its parents.
|