limarp-alpaca-instruct-lora

limarp-alpaca-instruct-lora

Experimental qlora/8bit lora trained using axolotl on the limarp dataset with the following Alpaca instruct format:

### Instruction:
Play the role of Character. You must engage in a roleplaying chat with User below this line. Do not write dialogues and narration for User. Character should respond with messages of medium length.

Character's Persona: {bot character description}

User's Persona: {user character description}

Scenario: {what happens in the story}

### Input:
...
{chat history}
...
User: {utterance}

### Response:
Character: {utterance}

Note that this is different from the default Alpaca instruct preset in SillyTavern and the one used for lemonilia/limarp-llama2-v2.

The system prompt and character template are provided under ### Instruction:

The chat context and user message are provided under ### Input:

The model's reply is under ### Response:

Model should be considered experimental, and YMMV.

Bias, Risks, and Limitations

The model will show biases similar to those observed in niche roleplaying forums on the Internet, besides those exhibited by the base model. It is not intended for supplying factual information or advice in any form.