athirdpath
commited on
Update README.md
Browse files
README.md
CHANGED
@@ -7,7 +7,14 @@ tags:
|
|
7 |
- merge
|
8 |
|
9 |
---
|
10 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
11 |
|
12 |
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
|
13 |
|
|
|
7 |
- merge
|
8 |
|
9 |
---
|
10 |
+
|
11 |
+
This is meant for further finetuning, but works well as-is. Made using a new structure I call "ripple merge" that works backwards and forwards through the model.
|
12 |
+
|
13 |
+
Other frankenmerge methods were failing at sizes over 11b.
|
14 |
+
|
15 |
+
---
|
16 |
+
|
17 |
+
# Llama-3-15b-Instruct
|
18 |
|
19 |
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
|
20 |
|