Heralax commited on
Commit
cfbae50
·
verified ·
1 Parent(s): 3de50dc

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +14 -9
README.md CHANGED
@@ -6,22 +6,25 @@ language:
6
 
7
  RewriteLikeMe-Nietzsche is a language model which has been trained to write like Nietzsche. It can turn any text into Nietzschean text. In the present case, the text was a small part of Beyond Good and Evil, which was used as training material; consequently, the model is experienced in the same style. It will probably give better results if one also gives it similar material to work on.
8
 
 
 
 
 
 
 
 
 
9
  <!-- --- -->
10
 
11
  > [!NOTE]
12
  > Do you want to train AIs like this one, but on your own style? [RewriteLikeMe.com](https://rewritelikeme.com/) (which I am working on) does just that! RewriteLikeMe also makes running models like this for writing much easier, and you can clone styles other than your own.
13
- The whole thing is very early beta, so please contact and shout at me if/when something explodes and I'll fix it!
14
- > You can also run this model on there without needing to set it up yourself. Check the public voices after you've gone through onboarding.
 
15
  <!-- --- -->
16
 
17
  Also, there is a [RewriteLikeMe Discord](https://discord.gg/s6PBfsaVzu), for questions and suggestions, and for the exchange of information and pleasant little things which one has written or made with this method. One can do a lot there.
18
 
19
- Included here are:
20
- - The LoRA files
21
- - The merged model files
22
- - The quantized merged model files (q8 GGUF)
23
- For usage with your preferred inference software.
24
-
25
  Ideal sampling parameters:
26
  ```
27
  temperature: 1.2,
@@ -102,4 +105,6 @@ Present your transformed text within <rephrase></rephrase> tags. Ensure that eve
102
 
103
  ```
104
 
105
- </details>
 
 
 
6
 
7
  RewriteLikeMe-Nietzsche is a language model which has been trained to write like Nietzsche. It can turn any text into Nietzschean text. In the present case, the text was a small part of Beyond Good and Evil, which was used as training material; consequently, the model is experienced in the same style. It will probably give better results if one also gives it similar material to work on.
8
 
9
+ Included here are:
10
+ - The LoRA files
11
+ - The merged model files
12
+ - The quantized merged model files (q8 GGUF)
13
+ For usage with your preferred inference software.
14
+
15
+ Base model = Mistral 7b v0.2
16
+
17
  <!-- --- -->
18
 
19
  > [!NOTE]
20
  > Do you want to train AIs like this one, but on your own style? [RewriteLikeMe.com](https://rewritelikeme.com/) (which I am working on) does just that! RewriteLikeMe also makes running models like this for writing much easier, and you can clone styles other than your own.
21
+ The whole thing is very early beta, so please contact and shout at me if/when something explodes and I'll fix it!
22
+
23
+ You can also run this model on there without needing to set it up yourself. Check the public voices after you've gone through onboarding.
24
  <!-- --- -->
25
 
26
  Also, there is a [RewriteLikeMe Discord](https://discord.gg/s6PBfsaVzu), for questions and suggestions, and for the exchange of information and pleasant little things which one has written or made with this method. One can do a lot there.
27
 
 
 
 
 
 
 
28
  Ideal sampling parameters:
29
  ```
30
  temperature: 1.2,
 
105
 
106
  ```
107
 
108
+ </details>
109
+
110
+ contact: evan@augmentoolkit.ai or hit me up on discord @Heralax