Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,34 @@
|
|
1 |
---
|
2 |
license: apache-2.0
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
license: apache-2.0
|
3 |
---
|
4 |
+
Merge viethq188/Rabbit-7B-v2-DPO-Chat and v1olet/v1olet_marcoroni-go-bruins-merge-7B using slerp merge from https://github.com/cg123/mergekit.
|
5 |
+
|
6 |
+
*config.yaml*
|
7 |
+
```
|
8 |
+
slices:
|
9 |
+
- sources:
|
10 |
+
- model: AIDC-ai-business/Marcoroni-7B-v3
|
11 |
+
layer_range: [0, 32]
|
12 |
+
- model: Q-bert/MetaMath-Cybertron-Starling
|
13 |
+
layer_range: [0, 32]
|
14 |
+
merge_method: slerp
|
15 |
+
base_model: AIDC-ai-business/Marcoroni-7B-v3
|
16 |
+
parameters:
|
17 |
+
t:
|
18 |
+
- filter: self_attn
|
19 |
+
value: [0, 0.5, 0.3, 0.7, 1]
|
20 |
+
- filter: mlp
|
21 |
+
value: [1, 0.5, 0.7, 0.3, 0]
|
22 |
+
- value: 0.5
|
23 |
+
dtype: float16
|
24 |
+
```
|
25 |
+
|
26 |
+
You can use alpaca template.
|
27 |
+
```
|
28 |
+
template_format = """{system}
|
29 |
+
### Instruction:
|
30 |
+
{prompt}
|
31 |
+
|
32 |
+
### Response:
|
33 |
+
"""
|
34 |
+
```
|