Update README.md
Browse files
README.md
CHANGED
@@ -9,7 +9,7 @@ tags:
|
|
9 |
|
10 |
This is a merge of Bytedance Seed-OSS-36B Base and Instruct, using the karcher-means method in [mergekit](https://github.com/cg123/mergekit), with the idea being to get Bytedance Instruct to 'feel' and write more like a raw continuation model.
|
11 |
|
12 |
-
Karcher was tested because this and SLERP are seemingly the only viable ways to merge
|
13 |
|
14 |
Quantized, it gets an MMLU score (via the exllamav3 eval script) of `11853/ 14042 = 84.41% correct, ( 80.41% prob.)`
|
15 |
|
|
|
9 |
|
10 |
This is a merge of Bytedance Seed-OSS-36B Base and Instruct, using the karcher-means method in [mergekit](https://github.com/cg123/mergekit), with the idea being to get Bytedance Instruct to 'feel' and write more like a raw continuation model.
|
11 |
|
12 |
+
Karcher was tested because this and SLERP are seemingly the only viable ways to merge an instruct and base model.
|
13 |
|
14 |
Quantized, it gets an MMLU score (via the exllamav3 eval script) of `11853/ 14042 = 84.41% correct, ( 80.41% prob.)`
|
15 |
|