Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,38 @@
|
|
1 |
---
|
2 |
license: cc-by-sa-4.0
|
|
|
|
|
|
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
license: cc-by-sa-4.0
|
3 |
+
language:
|
4 |
+
- zh
|
5 |
+
- en
|
6 |
---
|
7 |
+
|
8 |
+
|
9 |
+
# Zero凉宫春日
|
10 |
+
|
11 |
+
# Haruhi-Zero: Zero-Shot Role-Playing Model tuned on Yi-6B
|
12 |
+
|
13 |
+
主项目链接 https://github.com/LC1332/Chat-Haruhi-Suzumiya
|
14 |
+
|
15 |
+
过往的ChatHaruhi模型需要角色库来完成角色的构建,而Pygmalion,CharacterGLM,CharacterBaichuan等开源/闭源模型都开始支持zero-shot的角色卡片创建
|
16 |
+
|
17 |
+
我们构造以及收集了105k个中英文的conversation,以2500的token长度重新切到了120k左右个conversation,再结合小说数据进行了训练
|
18 |
+
|
19 |
+
- [李鲁鲁](https://github.com/LC1332)完成了数据的收集,搭建了gradio雏形
|
20 |
+
- [刘崇寒](https://github.com/khazic)完成了Yi-6B模型的sft训练并且上传
|
21 |
+
- [豆角](https://github.com/goodnessSZW)完成了qwen-1.8B Lora和Yi-6B Lora训练,我们会在之后上传
|
22 |
+
- [米唯实](https://github.com/hhhwmws0117)测试并上传模型到hugging face
|
23 |
+
|
24 |
+
## inference code
|
25 |
+
|
26 |
+
(搭建中)
|
27 |
+
|
28 |
+
https://github.com/LC1332/Zero-Haruhi/blob/main/notebook/HaruhiZeroGradio.ipynb
|
29 |
+
|
30 |
+
## TODO
|
31 |
+
|
32 |
+
|
33 |
+
|
34 |
+
## 鸣谢
|
35 |
+
|
36 |
+
樟树的ClaudeAPI
|
37 |
+
|
38 |
+
|