Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,33 @@
|
|
1 |
---
|
|
|
|
|
|
|
|
|
2 |
license: cc-by-4.0
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
+
pipeline_tag: text-generation
|
3 |
+
tags:
|
4 |
+
- mistral
|
5 |
+
- merge
|
6 |
license: cc-by-4.0
|
7 |
---
|
8 |
+
# Model Card for smol_bruin-7b
|
9 |
+
|
10 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
11 |
+
|
12 |
+
Slerp merge of go-bruins-v2 and smol-7b.
|
13 |
+
|
14 |
+
.yaml file for mergekit
|
15 |
+
|
16 |
+
```.yaml:
|
17 |
+
slices:
|
18 |
+
- sources:
|
19 |
+
- model: rwitz/go-bruins-v2
|
20 |
+
layer_range: [0, 32]
|
21 |
+
- model: rishiraj/smol-7b
|
22 |
+
layer_range: [0, 32]
|
23 |
+
merge_method: slerp
|
24 |
+
base_model: mistralai/Mistral-7B-v0.1
|
25 |
+
parameters:
|
26 |
+
t:
|
27 |
+
- filter: self_attn
|
28 |
+
value: [0.44, 0.72, 0.61, 0.83, 1]
|
29 |
+
- filter: mlp
|
30 |
+
value: [0.56, 0.28, 0.39, 0.17, 0]
|
31 |
+
- value: 0.5 # fallback for rest of tensors
|
32 |
+
dtype: float16
|
33 |
+
```
|