Update README.md
Browse files
README.md
CHANGED
@@ -4,12 +4,16 @@ datasets:
|
|
4 |
- Norquinal/claude_multiround_chat_30k
|
5 |
- OpenLeecher/Teatime
|
6 |
---
|
7 |
-
|
|
|
|
|
|
|
8 |
|
9 |
With RWKV world tokenizer,multi-langs have 1:1 tokenization ratio ,one word to one token.
|
10 |
(https://github.com/BlinkDL/ChatRWKV/blob/2a13ddecd81f8fd615b6da3a8f1091a594689e30/tokenizer/rwkv_tokenizer.py#L163)
|
11 |
|
12 |
|
|
|
13 |
|
14 |
This model trained with instructions datasets and chinese web novel and tradition wuxia,
|
15 |
more trainning details would be updated.
|
@@ -21,6 +25,8 @@ https://github.com/SynthiaDL/TrainChatGalRWKV/blob/main/train_world.sh
|
|
21 |
|
22 |
![QQ图片20230810153529.jpg](https://cdn-uploads.huggingface.co/production/uploads/6176b32847ee6431f632981e/d8ekmc4Lfhy2lYEdrRKXz.jpeg)
|
23 |
|
|
|
|
|
24 |
Using RWKV Runner https://github.com/josStorer/RWKV-Runner to test this model, only need 16G vram to run fp16 or 8G vram fp16i8, use temp 0.1-0.2 topp 0.7 for more precise answer ,temp between 1-2.x is more creatively.
|
25 |
![微信截图_20230810162303.png](https://cdn-uploads.huggingface.co/production/uploads/6176b32847ee6431f632981e/Ww45-WMngl4Jyt1OZDAa_.png)
|
26 |
|
|
|
4 |
- Norquinal/claude_multiround_chat_30k
|
5 |
- OpenLeecher/Teatime
|
6 |
---
|
7 |
+
|
8 |
+
# RWKV 7B World 128k for novel writing
|
9 |
+
|
10 |
+
We proudly announce this is the world first **128k context** model based on RWKV architecture today, 2023-08-10.
|
11 |
|
12 |
With RWKV world tokenizer,multi-langs have 1:1 tokenization ratio ,one word to one token.
|
13 |
(https://github.com/BlinkDL/ChatRWKV/blob/2a13ddecd81f8fd615b6da3a8f1091a594689e30/tokenizer/rwkv_tokenizer.py#L163)
|
14 |
|
15 |
|
16 |
+
# How to train infinte context model?
|
17 |
|
18 |
This model trained with instructions datasets and chinese web novel and tradition wuxia,
|
19 |
more trainning details would be updated.
|
|
|
25 |
|
26 |
![QQ图片20230810153529.jpg](https://cdn-uploads.huggingface.co/production/uploads/6176b32847ee6431f632981e/d8ekmc4Lfhy2lYEdrRKXz.jpeg)
|
27 |
|
28 |
+
# How to Test?
|
29 |
+
|
30 |
Using RWKV Runner https://github.com/josStorer/RWKV-Runner to test this model, only need 16G vram to run fp16 or 8G vram fp16i8, use temp 0.1-0.2 topp 0.7 for more precise answer ,temp between 1-2.x is more creatively.
|
31 |
![微信截图_20230810162303.png](https://cdn-uploads.huggingface.co/production/uploads/6176b32847ee6431f632981e/Ww45-WMngl4Jyt1OZDAa_.png)
|
32 |
|