Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,57 @@
|
|
1 |
---
|
2 |
license: apache-2.0
|
|
|
|
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
license: apache-2.0
|
3 |
+
tags:
|
4 |
+
- not-for-all-audiences
|
5 |
---
|
6 |
+
|
7 |
+
![image/png](https://cdn-uploads.huggingface.co/production/uploads/64dd7cda3d6b954bf7cdd922/ZXmxNKGaHUrqjdS1I3GkL.png)
|
8 |
+
|
9 |
+
## Mistral 7b Version of v0.6 ( with some extras )
|
10 |
+
|
11 |
+
This version has some additional data vs the llama2 13b version. Making it something of a 0.6 and a half or something. But that's confusing.
|
12 |
+
|
13 |
+
The 7b version of my Thespis finetune. From my testing it seems to perform somewhere between llama 7b and llama 13b.
|
14 |
+
|
15 |
+
This model is a personal project. It uses a vanilla chat template and is focused on providing multiturn sfw and nsfw RP experience.
|
16 |
+
|
17 |
+
This model works best with internet style RP using standard markup with asterisks surrounding actions and no quotes around dialogue.
|
18 |
+
|
19 |
+
It uses the following data:
|
20 |
+
|
21 |
+
* 2200 samples from Pure-Dove Dataset ( 90 token length or greater. )
|
22 |
+
* 2200 samples from Claude Multiround 30k ( 90 token length or greater. )
|
23 |
+
* 700 samples from Airoboros 3.1 ( Writing samples longer than 1500 tokens only. )
|
24 |
+
* 900 samples from the Augmental Dataset ( 90 token length or greater )
|
25 |
+
* 6000 samples of hand curated RP conversation with various characters.
|
26 |
+
|
27 |
+
Works with standard chat format for Ooba or SillyTavern.
|
28 |
+
|
29 |
+
## Prompt Format: Chat ( The default Ooba template and Silly Tavern Template )
|
30 |
+
```
|
31 |
+
{System Prompt}
|
32 |
+
|
33 |
+
Username: {Input}
|
34 |
+
BotName: {Response}
|
35 |
+
Username: {Input}
|
36 |
+
BotName: {Response}
|
37 |
+
|
38 |
+
```
|
39 |
+
## Ooba ( Set it to Chat, select a character and go. )
|
40 |
+
![image/png](https://cdn-uploads.huggingface.co/production/uploads/64dd7cda3d6b954bf7cdd922/HTl7QlAZcqe2hV8rwh4DG.png)
|
41 |
+
|
42 |
+
## Silly Tavern Settings ( Default )
|
43 |
+
![image/png](https://cdn-uploads.huggingface.co/production/uploads/64dd7cda3d6b954bf7cdd922/ajny8P0LdW0nFtghpPbfB.png)
|
44 |
+
|
45 |
+
|
46 |
+
|
47 |
+
## Turn Template (for Ooba Instruct if making a Discord bot or Some other Many to one Chat):
|
48 |
+
|
49 |
+
You can either bake usernames into the prompt directly for ease of use or programatically add them if running through the API to use as a chatbot.
|
50 |
+
|
51 |
+
```
|
52 |
+
User string: ( Leave empty if populating username into prompt through a script. Put in your username if its a 1 on 1 convo.) Ex. "DiscordUser1: "
|
53 |
+
Bot String: ( The bots name, followed by a colon and a space.) Ex. "Mayo: "
|
54 |
+
Context: ( Your bots system prompt, follow by a newline. )
|
55 |
+
|
56 |
+
<|user|><|user-message|>\n<|bot|><|bot-message|>\n
|
57 |
+
```
|