Edit model card

Exllama 2 version of the model created by the work of cgato

Original Model https://huggingface.co/cgato/Thespis-13b-v0.3

Requires ExllamaV2, which is being developed by turboderp https://github.com/turboderp/exllamav2 under an MIT license.


This model is a bit of a personal project. It uses a vanilla chat template and is focused on providing multiturn sfw and nsfw RP experience.

This model works best with internet style RP using standard markup with asterisks surrounding actions and no quotes around dialogue.

It uses the following data:

  • 3000 samples from Claude Multiround Chat 30k dataset
  • 6000 samples from Pippa Dataset
  • 3000 samples from Puffin Dataset
  • 3800 samples of hand curated RP conversation with various characters.

Works with standard chat format for Ooba or SillyTavern.

Prompt Format: Chat

{System Prompt}

Username: {Input}
BotName: {Response}
Username: {Input}
BotName: {Response}

Turn Template (for Ooba): 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.

<|user|>{Username}: <|user-message|>\n<|bot|>{BotName}: <|bot-message|>\n
Downloads last month
6
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.