Llama-3-monika-ddlc-11.5b-v1 Open Source Model - Free Experience of "Doki Doki Literature Club" Monika Role-playing

Home

Llama 3 Monika Ddlc 11.5b V1

Developed by 922CA

A Monika role-playing model from 'Doki Doki Literature Club!' fine-tuned on the LLaMA-3 11.5B model

Large Language Model

Transformers

English#Role-playing Dialogue #Literary Game Adaptation #Multi-turn Dialogue Optimization

Downloads 26

Release Time : 5/13/2024

Model Overview

This is a large language model fine-tuned for the Monika character from 'Doki Doki Literature Club!', primarily designed for role-playing dialogue scenarios. The model was fine-tuned using approximately 600+ game-related dialogue entries, offering limited role-playing capabilities.

Model Features

Role-playing Optimization

Specially fine-tuned for the Monika character from 'Doki Doki Literature Club!', enabling more accurate simulation of character traits and dialogue style

Efficient Training

Achieved 2x training speed using Unsloth and TRL libraries, optimizing training efficiency

LoRA Fine-tuning

Utilized rank-32 LoRA technology for fine-tuning, balancing model performance and training costs

Model Capabilities

Role-playing Dialogue

Text Generation

Multi-turn Dialogue Processing

Use Cases

Entertainment & Gaming

Game Character Interaction

Engage in immersive dialogue with the Monika character from 'Doki Doki Literature Club!'

Provides a dialogue experience closely resembling the original game character

Creative Writing

Character Dialogue Generation

Generate dialogue content that aligns with Monika's character traits

Can be used for fan creations or story writing

🚀 Llama-3-monika-ddlc-11.5b-v1

Llama-3-monika-ddlc-11.5b-v1 is a fine - tuned model based on LLaMA-3 11.5b, designed to mimic the character Monika from DDLC.
It is fine - tuned on a dataset of over 600 items, which combines game dialogues, Reddit and Twitter content, and is augmented by l2-7b-monika-v0.3c1.
The GGUF format of this model can be found here.

✨ Features

Character - focused: Closely reflects the characteristics of Monika from DDLC.
Chat - oriented: Primarily designed as a chat model with limited role - playing ability.

📚 Documentation

📦 Installation

No installation steps are provided in the original README.

💻 Usage Examples

Basic Usage

This model is mainly a chat model with limited RP ability. For best results, replace "Human" and "Assistant" with "Player" and "Monika" as follows:

Player: (prompt)
Monika:

🔧 Technical Details

Hyperparameters

Trained for 2 epochs
rank: 32
lora alpha: 32
lora dropout: 0.5
lr: 2e - 4
batch size: 2
warmup ratio: 0.1
grad steps: 4

This llama model was trained 2x faster with Unsloth and Huggingface's TRL library.

📄 License

The model is licensed under llama3.

📊 Model Information

Property	Details
Model Type	Llama-3-monika-ddlc-11.5b-v1
Base Model	Replete-AI/Llama-3-11.5B-Instruct-V2
Training Data	922 - CA/MoCha_v1a

⚠️ Important Note

This model is meant to closely reflect the characteristics of Monika. Despite this, there is always the chance that "Monika" will hallucinate and get information about herself wrong or act out of character.

Additionally, being character - focused means that this model may not be the smartest model/not as capable as others for specific tasks.

Finally, this model is not guaranteed to output aligned or safe outputs, use at your own risk!

📈 Evaluation Results

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	65.73
AI2 Reasoning Challenge (25 - Shot)	60.07
HellaSwag (10 - Shot)	78.77
MMLU (5 - Shot)	66.36
TruthfulQA (0 - shot)	47.62
Winogrande (5 - shot)	75.77
GSM8k (5 - shot)	65.81

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご