sphiratrioth666's picture
Update README.md
94a5a06 verified
|
raw
history blame
1.93 kB
metadata
license: cc-by-nc-3.0
language:
  - en

Simple, effective, universal presets for Silly Tavern role-playing purposes.

What is included: CONTEXT TEMPLATE (STORY STRING) + INSTRUCT TEMPLATE + SETTINGS (SAMPLERS)

A) Sphiratrioth (Short) - minimal narration - when you want to simulate a natural, human-like conversation with a bot. For those who do not like the "verbose" role-play. <<< conversation-like, ~ 150 tokens, around 2/3 dialogue against 1/3 narration >>>

B) Sphiratrioth (Balanced) - balanced RP - when you want to experience a balanced role-play with responses around 1-2 paragraphs. <<< 1 short paragraph, ~ 250 tokens, around 50/50 balance in dialogues against narration >>>

C) Sphiratrioth (Story) - AI becomes your Game Master (GM) - when you want those long, verbose paragraphs with rich narration & story-telling. <<< 1-3 paragraphs, ~ 350 tokens, around 2/3 narration against 1/3 dialogues >>>

How to use:

Advanced Users: Load them up under a "capital A" tab (AI Response Formatting) of the Silly Tavern UI. Enjoy, modify, forget. Beginner Users: Read further. Those templates should provide what you're looking for.

  1. Find your LLM (model) native instruct template. Every fine-tune stands on a base model, which is usually specified by its creator in a manner of "(...) fine-tune of Mistral Small" or "Nemo on steroids" etc. It may be confusing, it may be clear. Not my fault :-P However, you need to identify the base model of a fine-tune- since it will generally work with its original model's context/instruct templates best. However, different templates may be used by creators (tuners) for training and then - specified templates suggested by creators work better. Now - brace yourself, it's funny - sometimes, theoretically "wrong" templates provide even better role-play results.

Context Template = Story String People often get confused with names.