tsukasa

https://huggingface.co/ludis/tsukasa-13b-qlora-limarp
https://huggingface.co/ludis/tsukasa-13b-qlora-limarp-gptq
https://huggingface.co/ludis/tsukasa-13b-qlora-limarp-gguf

if u wanna host it urself on collab put ludis for the org tsukasa-13b-qlora-limarp-gptq for the repo and main for the branch, normal llamabro notebook is slightly bugged since my repo name has the gptq in lowercase use the notebook posted below if you have issues
https://0x0.st/HfVJ.ipynb

silly tavern prompts

story string:

1
2
3
4

{{#if system}}{{system}}{{/if}}
{{#if description}}{{description}}{{/if}}
{{char}}'s Persona: {{persona}}
{{#if scenario}}This scenario of the conversation is: {{scenario}}{{/if}}

system:

1

Enter roleplay mode. You must act as {{char}}, whose persona follows:

agnai prompts

system:

1

You are now in roleplay conversation mode. This is your character persona:

gaslight:

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17

<|system|>{{#if system_prompt}}{{system_prompt}}{{/if}}


{{char}}'s Persona: {{personality}}

{{#if example_dialogue}}
{{char}} talks like: {{example_dialogue}}
{{/if}}

{{#if scenario}}
This scenario of the conversation is: {{scenario}}
{{/if}}

{{#each msg}}{{#if .isbot}}<|model|>{{/if}}{{#if .isuser}}<|user|>{{/if}}{{.name}}: {{.msg}}
{{/each}}
{{#if ujb}}<|system|>{{ujb}}{{/if}}
<|model|>{{post}}

ujb:

1

Play the role of {{char}}. You must engage in a roleplaying chat with {{user}} below this line. Do not write dialogues and narration for {{user}}. {{char}} should respond with messages of medium length.

gen settings

add <|user|>, <|system|>, and <|model|> to custom stopping strings

tail-free sampling:
temperature=0.75, Repetition penalty=~1.18, top-p=1, top-k=0, Tail Free Sampling=0.85, Repetition penalty range=0, Microstat Tau=0, Microstat LR (ETA)=1

i find temp, tfs, and rep pen has greatest impact on output quality
this is just what i found works nice, if you find better gen settings lmk at no1namialuvr@proton.me logs also welcome at same email

limarp

since the last dataset the model was tuned on (limarp) includes persona's for both characters, you might get better results if you include a persona for the character (you) are roleplaying as not just the bot, also limarp data doesn't use asteriks for actions and has dialogue in quotes.