Downloads · 30 days
3
11% of all-time downloads
haofue2i1z3/Qwen3-8B-LegalConsult-Gold
Qwen3-8B-LegalConsult-Gold is a text generation model from haofue2i1z3. Use it when you need the model to write or continue text. It is set up for peft. The card lists the license as apache-2.0.
基于 Qwen/Qwen3-8B-Base 的中文法律咨询 LoRA。模型使用 86 条经过审核的咨询样本进行 2 epoch、all-linear rank-64 QLoRA 微调。
Downloads · 30 days
3
11% of all-time downloads
All-time downloads
27
Public
Repo size
710 MB
Likes
0
Public
Click a slice to open those files.
.safetensors698 MB · 98%
From the Hugging Face model README
基于 Qwen/Qwen3-8B-Base 的中文法律咨询 LoRA。模型使用 86 条经过审核的咨询样本进行 2 epoch、all-linear rank-64 QLoRA 微调。
在项目历史 Legal-50 语义口径下,均分由 6.81 提升至 7.33,成对结果为 29 胜、11 平、10 负。这是研究阳性结果,不代表生产级法律可靠性;模型仍可能给出错误、过时或无依据的法律判断与法条引用。
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
adapter = "haofue2i1z3/Qwen3-8B-LegalConsult-Gold"
base = "Qwen/Qwen3-8B-Base"
tokenizer = AutoTokenizer.from_pretrained(adapter, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(base, device_map="auto", torch_dtype="auto")
model = PeftModel.from_pretrained(model, adapter)