xiaol commited on
Commit
a704cd2
1 Parent(s): 378293f

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +6 -3
README.md CHANGED
@@ -4,9 +4,9 @@ license: apache-2.0
4
 
5
  This is full finetuned model from RWKV 4 world 7B CHNTuned model using data from Readflow tech (readflow.com.cn) ,
6
 
7
- finetuned for 32k context length, used to summary news article
8
 
9
- using inf-ctx training https://github.com/SynthiaDL/TrainChatGalRWKV
10
 
11
  you can test summary prompt using RWKV runner(https://github.com/josStorer/RWKV-Runner) in chat mode , and check conversation files in examples folders.
12
 
@@ -17,7 +17,10 @@ https://discord.gg/pWH5MkvtNR
17
 
18
  这是和会读Readflow合作的模型( readflow.com.cn),用于超长微信文章的摘要,训练了32k的context长度,能一次性输入整篇文章进行摘要,如下图以及在example文件夹中的例子,
19
 
20
- 23k tokens一次性输入摘要,“Assistant:"是模型输出内容
 
 
 
21
 
22
  可以使用runner进行测试 https://github.com/josStorer/RWKV-Runner
23
 
 
4
 
5
  This is full finetuned model from RWKV 4 world 7B CHNTuned model using data from Readflow tech (readflow.com.cn) ,
6
 
7
+ finetuned for 32k context length, used to summary news article
8
 
9
+ using inf-ctx training https://github.com/SynthiaDL/TrainChatGalRWKV with fixed VRAM
10
 
11
  you can test summary prompt using RWKV runner(https://github.com/josStorer/RWKV-Runner) in chat mode , and check conversation files in examples folders.
12
 
 
17
 
18
  这是和会读Readflow合作的模型( readflow.com.cn),用于超长微信文章的摘要,训练了32k的context长度,能一次性输入整篇文章进行摘要,如下图以及在example文件夹中的例子,
19
 
20
+ 23k tokens一次性输入摘要,不增显存,如果有state状态,可以秒出,可以反复重试多种可能性,
21
+
22
+ “Assistant:"是模型输出内容
23
+
24
 
25
  可以使用runner进行测试 https://github.com/josStorer/RWKV-Runner
26