46 lines
1.1 KiB
Markdown
46 lines
1.1 KiB
Markdown
|
---
|
||
|
|
tags:
|
||
|
|
- not-for-all-audiences
|
||
|
|
license: apache-2.0
|
||
|
|
---
|
||
|
|
|
||
|
|
|
||
|
|
Test model, still cleaning the dataset up. Expect repetition.
|
||
|
|
|
||
|
|
|
||
|
|
|
||
|
|
|
||
|
|
|
||
|
|
## Thespis Test
|
||
|
|
|
||
|
|
A test model, I make no guarantees about the performance. Let me know if its any good. Seems more verbose than my previous models from my testing.
|
||
|
|
|
||
|
|
External Datasets Used:
|
||
|
|
|
||
|
|
* Pure-Dove Dataset
|
||
|
|
* Claude Multiround 30k
|
||
|
|
* OpenOrcaSlim
|
||
|
|
* Augmental Dataset
|
||
|
|
* ToxicQA
|
||
|
|
* yahoo_answers_qa
|
||
|
|
|
||
|
|
Special thanks to kubernetes-bad for providing me a cleaned chub.ai characters dataset.
|
||
|
|
|
||
|
|
## Prompt Format: Chat ( The default Ooba template and Silly Tavern Template )
|
||
|
|
```
|
||
|
|
{System Prompt}
|
||
|
|
|
||
|
|
Username: {Input}
|
||
|
|
BotName: {Response}
|
||
|
|
Username: {Input}
|
||
|
|
BotName: {Response}
|
||
|
|
|
||
|
|
```
|
||
|
|
## Ooba ( Set it to Chat, select a character and go. )
|
||
|
|
|
||
|
|
|
||
|
|
## Silly Tavern Settings ( Default )
|
||
|
|
|
||
|
|
```
|
||
|
|
|