Downloads · 30 days
9
18% of all-time downloads
dpo-qwen-expG-lengthfix-adapter
Umezaki/dpo-qwen-expG-lengthfix-adapter
dpo-qwen-expG-lengthfix-adapter is a text generation model from Umezaki. 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.
SFT(ExpG) + DPO, r=8, alpha=8 - SFT adapter merged from Exp G - DPO r=8, alpha=8 - LR=5e-7, epoch=1, beta=0.1 - Length bias fix: rejected truncated to match chosen token length - maxlength=2048
All-time downloads
49
Public
Repo size
77.6 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors66.1 MB · 81%
At a glance
- Task
- Text Generation
- Library
- peft
- License
- apache-2.0
- Access
- Public
- Created
- Feb 7, 2026
- Updated
- Feb 7, 2026
- SHA
- b2da6757
Try a prompt
Base models
Model card
From the Hugging Face model README
DPO LoRA Adapter (ExpL) for Qwen3-4B (StructEval)
SFT(ExpG) + DPO, r=8, alpha=8
- SFT adapter merged from Exp G
- DPO r=8, alpha=8
- LR=5e-7, epoch=1, beta=0.1
- Length bias fix: rejected truncated to match chosen token length
- max_length=2048
- Task
- Text Generation
- Library
- peft
- License
- apache-2.0
- Languages
- en
- Created
- Feb 7, 2026
- Updated
- Feb 7, 2026