Update README.md
Browse files
README.md
CHANGED
@@ -1,7 +1,7 @@
|
|
1 |
---
|
2 |
language:
|
3 |
- en
|
4 |
-
license:
|
5 |
tags:
|
6 |
- text-generation-inference
|
7 |
- transformers
|
@@ -9,15 +9,29 @@ tags:
|
|
9 |
- mistral
|
10 |
- trl
|
11 |
- sft
|
|
|
|
|
12 |
base_model: SanjiWatsuki/Kunoichi-DPO-v2-7B
|
13 |
---
|
|
|
|
|
|
|
14 |
|
15 |
-
|
16 |
|
17 |
-
- **
|
18 |
-
- **
|
19 |
-
- **Finetuned from model :** SanjiWatsuki/Kunoichi-DPO-v2-7B
|
20 |
|
21 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
22 |
|
23 |
-
|
|
|
|
|
|
1 |
---
|
2 |
language:
|
3 |
- en
|
4 |
+
license: cc-by-nc-4.0
|
5 |
tags:
|
6 |
- text-generation-inference
|
7 |
- transformers
|
|
|
9 |
- mistral
|
10 |
- trl
|
11 |
- sft
|
12 |
+
- Roleplay
|
13 |
+
- roleplay
|
14 |
base_model: SanjiWatsuki/Kunoichi-DPO-v2-7B
|
15 |
---
|
16 |
+
> [!Important]
|
17 |
+
> Still in experiment
|
18 |
+
# About this model
|
19 |
|
20 |
+
Remake [version 2](https://huggingface.co/Alsebay/NarumashiRTS-V2) with safetensor format, more safety and stable method, nothing change too much (base on the model hash). But to be real, in the previous version 2, I have use unsafety method to save pretrain model, which could lead apply Lora layer twice to model, that make model have terrible performance.
|
21 |
|
22 |
+
- **Finetuned with rough translate dataset, to increase the accuracy in TSF theme, which is not quite popular. (lewd dataset)**
|
23 |
+
- **Finetuned from model :** SanjiWatsuki/Kunoichi-DPO-v2-7B . Thank SanjiWatsuki a lot :)
|
|
|
24 |
|
25 |
+
## V2 have more epochs.
|
26 |
+
## Dataset
|
27 |
+
```
|
28 |
+
Dataset(all are novels):
|
29 |
+
30% skinsuit
|
30 |
+
30% possession
|
31 |
+
35% transform(shapeshift)
|
32 |
+
5% other
|
33 |
+
```
|
34 |
|
35 |
+
# Thank Unsloth for good finetuning tool. This mistral model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
|
36 |
+
|
37 |
+
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
|