Jan
commited on
Commit
·
55d1889
1
Parent(s):
662c68e
Update README.md
Browse files
README.md
CHANGED
@@ -17,22 +17,22 @@ language:
|
|
17 |
|
18 |
# Model Description
|
19 |
This model uses the `Slerp` merge method from 2 models:
|
20 |
-
1. [
|
21 |
-
2. [
|
22 |
|
23 |
-
- base model: [
|
24 |
|
25 |
The yaml config file for this model is here:
|
26 |
|
27 |
```yaml
|
28 |
slices:
|
29 |
- sources:
|
30 |
-
- model:
|
31 |
layer_range: [0, 32]
|
32 |
-
- model:
|
33 |
layer_range: [0, 32]
|
34 |
merge_method: slerp
|
35 |
-
base_model:
|
36 |
parameters:
|
37 |
t:
|
38 |
- filter: self_attn
|
@@ -68,4 +68,5 @@ Detailed results can be found here.
|
|
68 |
# Acknowlegement
|
69 |
- [mergekit](https://github.com/cg123/mergekit)
|
70 |
- [DARE](https://github.com/yule-BUAA/MergeLM/blob/main/README.md)
|
|
|
71 |
- [lm-evaluation-harness](https://github.com/EleutherAI/lm-evaluation-harness)
|
|
|
17 |
|
18 |
# Model Description
|
19 |
This model uses the `Slerp` merge method from 2 models:
|
20 |
+
1. [v1olet_marcoroni-go-bruins-merge-7B](https://huggingface.co/v1olet/v1olet_marcoroni-go-bruins-merge-7B)
|
21 |
+
2. [juanako-7b-UNA](https://huggingface.co/fblgit/juanako-7b-UNA)
|
22 |
|
23 |
+
- base model: [v1olet_marcoroni-go-bruins-merge-7B](https://huggingface.co/v1olet/v1olet_marcoroni-go-bruins-merge-7B)
|
24 |
|
25 |
The yaml config file for this model is here:
|
26 |
|
27 |
```yaml
|
28 |
slices:
|
29 |
- sources:
|
30 |
+
- model: v1olet/v1olet_marcoroni-go-bruins-merge-7B
|
31 |
layer_range: [0, 32]
|
32 |
+
- model: fblgit/juanako-7b-UNA
|
33 |
layer_range: [0, 32]
|
34 |
merge_method: slerp
|
35 |
+
base_model: v1olet/v1olet_marcoroni-go-bruins-merge-7B
|
36 |
parameters:
|
37 |
t:
|
38 |
- filter: self_attn
|
|
|
68 |
# Acknowlegement
|
69 |
- [mergekit](https://github.com/cg123/mergekit)
|
70 |
- [DARE](https://github.com/yule-BUAA/MergeLM/blob/main/README.md)
|
71 |
+
- [SLERP](https://github.com/Digitous/LLM-SLERP-Merge)
|
72 |
- [lm-evaluation-harness](https://github.com/EleutherAI/lm-evaluation-harness)
|