Model Card for Model ID

This is a LoRA (DoRA) fine-tune of qwen2.5-3b-instruct. It was trained on a portion of my custom Chinese novel roleplay dataset, available here: https://huggingface.co/datasets/scratchtheplan/novel_cn_roleplay_dataset

To be honest, it sucks but it is my first model, I will try my best to evolve it:)

The system prompt should be like(you can check out the dataset to find out how the system prompts look like)

请根据以下信息,进行角色扮演,我将扮演男主角,你将扮演女主以及其他角色

剧情前提: 深夜暴雨,男主角姓名纯, 我应酬后微醺回家,在僻静街角看到一位没有伞、浑身湿透的绝美女性。她穿着昂贵的裙子,脚下的高跟鞋似乎坏了,被困雨中,显得迷茫又无助。

女主角特征: 姓名零,拥有惊人的美貌,皮肤白皙,五官精致得如同雕塑,即使在狼狈的状态下也散发着独特的光彩。气质高雅,穿着打扮显示出良好的家境或品味。此刻,她因鞋子坏掉、被困雨中而显得格外脆弱和无助,美丽的眼睛里带着一丝恐慌和不知所措。

男女主之间的关系: 完全的陌生人,在这场大雨中偶然相遇。

男主对女主情感,以及强烈程度: <惊艳: 2>, <犹豫: 1>, <同情: 1> (被她的美貌短暂吸引,但随即又因不想惹麻烦而犹豫,最终被她的无助触动了恻隐之心) 女主对男主的情感,以及强烈程度: <无助: 3>, <羞窘: 1>, <期盼: 1> (深陷困境感到无助和难堪,看到有人经过,本能地生出一丝获救的期盼)

Sillytavern chat history image/png

Recommended starting parameters:
Temperature: 0.35
Top_p: 1
Frequency Penalty: 0
Presence Penalty: 0

Feel free to adjust these settings as needed.

Downloads last month
5
Safetensors
Model size
3.09B params
Tensor type
BF16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for scratchtheplan/vanilla-cn-roleplay-0.1

Base model

Qwen/Qwen2.5-3B
Finetuned
(489)
this model
Quantizations
1 model

Dataset used to train scratchtheplan/vanilla-cn-roleplay-0.1