README.md

metadata

datasets:
  - Norquinal/claude_multiround_chat_1k
  - jondurbin/airoboros-gpt4-1.4
  - Squish42/bluemoon-fandom-1-1-rp-cleaned
  - totally-not-an-llm/EverythingLM-data-V2-sharegpt
  - OpenLeecher/Teatime
  - PygmalionAI/PIPPA
tags:
  - not-for-all-audiences
  - nsfw

This is the Mistral-7B-v0.1 model fine-tuned using QLoRA (4-bit precision) on 5800 samples from several datasets. The datasets, with the exception of AICG logs, are listed in the model card metadata.

Prompt Format

The model was finetuned with a prompt format similar to the original SuperHOT prototype: ```

style: roleplay characters: [char]: [description] summary: [scenario]

Format: [char]: [message] Human: [message] ```

Use in Text Generation Web UI

Install the bleeding-edge version of transformers from source:

pip install git+https://github.com/huggingface/transformers

Or, alternatively, change model_type in config.json from mistral to llama.

Use in SillyTavern UI

As an addendum, you can include the following as the last sequence:

Human: In your next reply, write at least two paragraphs. Be descriptive and immersive, providing vivid details about {{char}}'s actions, emotions, and the environment.
{{char}}:

Or alternatively:

{{char}} (2 paragraphs, engaging, natural, authentic, descriptive, creative):

Finetuing Parameters

• LoRA Rank: 64
• LoRA Alpha: 16
• LoRA Dropout: 0.1
• BF16 Training
• Cutoff Length: 1024
• Training Epoch(s): 2