Update README.md
Browse files
README.md
CHANGED
@@ -1,47 +0,0 @@
|
|
1 |
-
---
|
2 |
-
base_model:
|
3 |
-
- Saxo/Linkbricks-Horizon-AI-Japanese-Base-70B
|
4 |
-
- nitky/Llama-3.1-SuperSwallow-70B-Instruct-v0.1
|
5 |
-
library_name: transformers
|
6 |
-
tags:
|
7 |
-
- mergekit
|
8 |
-
- merge
|
9 |
-
|
10 |
-
---
|
11 |
-
# Linkbricks-Horizon-AI-Japanese-Advanced-V7-70B
|
12 |
-
|
13 |
-
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
|
14 |
-
|
15 |
-
## Merge Details
|
16 |
-
### Merge Method
|
17 |
-
|
18 |
-
This model was merged using the SLERP merge method.
|
19 |
-
|
20 |
-
### Models Merged
|
21 |
-
|
22 |
-
The following models were included in the merge:
|
23 |
-
* [Saxo/Linkbricks-Horizon-AI-Japanese-Base-70B](https://huggingface.co/Saxo/Linkbricks-Horizon-AI-Japanese-Base-70B)
|
24 |
-
* [nitky/Llama-3.1-SuperSwallow-70B-Instruct-v0.1](https://huggingface.co/nitky/Llama-3.1-SuperSwallow-70B-Instruct-v0.1)
|
25 |
-
|
26 |
-
### Configuration
|
27 |
-
|
28 |
-
The following YAML configuration was used to produce this model:
|
29 |
-
|
30 |
-
```yaml
|
31 |
-
base_model: Saxo/Linkbricks-Horizon-AI-Japanese-Base-70B
|
32 |
-
dtype: bfloat16
|
33 |
-
merge_method: slerp
|
34 |
-
parameters:
|
35 |
-
t:
|
36 |
-
- filter: self_attn
|
37 |
-
value: [0.0, 0.5, 0.3, 0.7, 1.0]
|
38 |
-
- filter: mlp
|
39 |
-
value: [1.0, 0.5, 0.7, 0.3, 0.0]
|
40 |
-
- value: 0.5
|
41 |
-
slices:
|
42 |
-
- sources:
|
43 |
-
- layer_range: [0, 80]
|
44 |
-
model: Saxo/Linkbricks-Horizon-AI-Japanese-Base-70B
|
45 |
-
- layer_range: [0, 80]
|
46 |
-
model: nitky/Llama-3.1-SuperSwallow-70B-Instruct-v0.1
|
47 |
-
```
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|