Update README.md
Browse files
README.md
CHANGED
@@ -10,15 +10,26 @@ language:
|
|
10 |
- en
|
11 |
library_name: transformers
|
12 |
---
|
13 |
-

|
14 |
|
15 |
-
|
16 |
|
|
|
17 |
## A SLERP merge of my two current fav 7B models
|
|
|
18 |
[macadeliccc/WestLake-7B-v2-laser-truthy-dpo](https://huggingface.co/macadeliccc/WestLake-7B-v2-laser-truthy-dpo) & [SanjiWatsuki/Kunoichi-DPO-v2-7B](https://huggingface.co/SanjiWatsuki/Kunoichi-DPO-v2-7B)
|
19 |
|
20 |
A set of GGUF quants of [Finch](https://huggingface.co/antiven0m/finch)
|
21 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
22 |
```yaml
|
23 |
base_model: macadeliccc/WestLake-7B-v2-laser-truthy-dpo
|
24 |
dtype: float16
|
|
|
10 |
- en
|
11 |
library_name: transformers
|
12 |
---
|
|
|
13 |
|
14 |
+

|
15 |
|
16 |
+
# **Finch 7b Merge**
|
17 |
## A SLERP merge of my two current fav 7B models
|
18 |
+
|
19 |
[macadeliccc/WestLake-7B-v2-laser-truthy-dpo](https://huggingface.co/macadeliccc/WestLake-7B-v2-laser-truthy-dpo) & [SanjiWatsuki/Kunoichi-DPO-v2-7B](https://huggingface.co/SanjiWatsuki/Kunoichi-DPO-v2-7B)
|
20 |
|
21 |
A set of GGUF quants of [Finch](https://huggingface.co/antiven0m/finch)
|
22 |
|
23 |
+
- [6bpw EXL2 Quant](https://huggingface.co/antiven0m/finch-6bpw-exl2)
|
24 |
+
|
25 |
+
## Settings
|
26 |
+
I reccomend using the **ChatML** format. As for samplers, I reccomend the following:
|
27 |
+
```yml
|
28 |
+
Temperature: 1.2
|
29 |
+
Min P: 0.2
|
30 |
+
Smoothing Factor: 0.2
|
31 |
+
```
|
32 |
+
## Mergekit Config
|
33 |
```yaml
|
34 |
base_model: macadeliccc/WestLake-7B-v2-laser-truthy-dpo
|
35 |
dtype: float16
|