90 lines
3.3 KiB
Markdown
90 lines
3.3 KiB
Markdown
---
|
||
base_model: Qwen/Qwen3-1.7B
|
||
library_name: transformers
|
||
tags:
|
||
- qlora
|
||
- roleplay
|
||
- qwen
|
||
- sft
|
||
- touhou
|
||
- character-roleplay
|
||
- synthetic-data
|
||
- generated_from_trainer
|
||
language:
|
||
- zh
|
||
license: apache-2.0
|
||
---
|
||
|
||
# Qwen3-1.7B-Yukari-SFT-v2
|
||
|
||
八云紫 (Yukari Yakumo) 角色扮演模型 — SFT 改进版。
|
||
|
||
v1 (r=8) 的基础上提高 LoRA rank 至 32，改用 1,233 条纯原文 SFT 数据解决攻击性输出问题。
|
||
|
||
## 模型信息
|
||
|
||
| 项目 | 值 |
|
||
|------|-----|
|
||
| 基座 | [Qwen/Qwen3-1.7B](https://huggingface.co/Qwen/Qwen3-1.7B) |
|
||
| 方法 | QLoRA 4-bit NF4 (双重量化), bf16 |
|
||
| LoRA | **r=32, alpha=64**, target: q/k/v/o/gate/up/down proj |
|
||
| 数据 | 1,233 条合成 SFT 对话 (MiMo-V2.5-Pro, 纯原文无情绪变体) |
|
||
| 训练 | 1 epoch, seed=478 |
|
||
| 格式 | Merged bf16 safetensors + GGUF F16 |
|
||
|
||
## 系列模型
|
||
|
||
| 模型 | LoRA | 方法 | 版本 |
|
||
|------|------|------|------|
|
||
| [Qwen3-1.7B-Yukari-SFT](https://hf.co/zzoceanpie/Qwen3-1.7B-Yukari-SFT) | r=8, a=8 | SFT | v1 |
|
||
| **Qwen3-1.7B-Yukari-SFT-v2** (本模型) | r=32, a=64 | SFT | v2 |
|
||
| [Qwen3-1.7B-Yukari-DPO](https://hf.co/zzoceanpie/Qwen3-1.7B-Yukari-DPO) | r=32, a=64 | SFT+DPO | v2+DPO |
|
||
|
||
## 用法
|
||
|
||
模型输入格式为 `[情绪标签]\n用户输入`，标签控制八云紫的回复语气。8 维 Plutchik 情绪向量离散化：喜悦/愤怒/悲伤/恐惧/厌恶/惊讶/信任/期待，每维取「无」「极轻微」「轻微」「中轻度」「中度」「较强烈」「强烈」「极强烈」。
|
||
|
||
```python
|
||
from transformers import AutoModelForCausalLM, AutoTokenizer
|
||
|
||
model = AutoModelForCausalLM.from_pretrained(
|
||
"zzoceanpie/Qwen3-1.7B-Yukari-SFT-v2",
|
||
torch_dtype="auto", device_map="auto", trust_remote_code=True
|
||
)
|
||
tokenizer = AutoTokenizer.from_pretrained("zzoceanpie/Qwen3-1.7B-Yukari-SFT-v2")
|
||
|
||
tags = "[<|中度喜悦|><|轻微愤怒|><|轻微悲伤|><|无恐惧|><|无厌恶|><|无惊讶|><|轻微信任|><|中度期待|>]"
|
||
prompt = f"{tags}\n你觉得人类最有趣的地方是什么？"
|
||
|
||
messages = [{"role": "user", "content": prompt}]
|
||
text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
|
||
inputs = tokenizer(text, return_tensors="pt").to(model.device)
|
||
outputs = model.generate(**inputs, max_new_tokens=128, temperature=0.8)
|
||
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
|
||
```
|
||
### GGUF
|
||
|
||
```python
|
||
from llama_cpp import Llama
|
||
|
||
model = Llama(model_path="Qwen3-1.7B-Yukari-SFT-f16.gguf", n_ctx=2048, n_gpu_layers=-1)
|
||
response = model.create_chat_completion(
|
||
messages=[{"role": "user", "content": prompt}],
|
||
max_tokens=200, temperature=0.8
|
||
)
|
||
print(response["choices"][0]["message"]["content"])
|
||
```
|
||
|
||
## 许可 / License
|
||
|
||
- 模型权重: Apache 2.0
|
||
- 角色「八云紫」版权: (C) 上海爱丽丝幻乐团 (Team Shanghai Alice) / ZUN
|
||
- 本模型是东方Project的二次创作/粉丝作品。遵循 [东方Project二次创作指南](https://touhou-project.news/guidelines_en/)。
|
||
- 训练数据: [yukari-synthetic](https://hf.co/datasets/zzoceanpie/yukari-synthetic) (Apache 2.0)
|
||
- *This is a fan-made derivative work based on Touhou Project. Touhou Project is (C) Team Shanghai Alice / ZUN.*
|
||
|
||
> This model is a fan-made derivative work based on Touhou Project.
|
||
> Touhou Project is (C) Team Shanghai Alice / ZUN.
|
||
> Please refer to the [official guidelines](https://touhou-project.news/guidelines_en/) for Touhou fan content.
|
||
|