Downloads · 30 days
14
19% of all-time downloads
Nova-Qwen-14B-LoRA
Lighstromo/Nova-Qwen-14B-LoRA
Nova-Qwen-14B-LoRA is a text generation model from Lighstromo. 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.
LoRA adapter for Qwen2.5-14B-Instruct, fine-tuned on GPT-4o conversations to create a warm, creative, and intelligent conversational model.
All-time downloads
73
Public
Repo size
1.1 GB
Likes
4
Public
Repo makeup
Click a slice to open those files.
.safetensors1.1 GB · 99%
At a glance
- Task
- Text Generation
- Library
- peft
- License
- apache-2.0
- Access
- Public
- Created
- Mar 2, 2026
- Updated
- Mar 2, 2026
- SHA
- aa07b291
Try a prompt
Base models
Model card
From the Hugging Face model README
Nova-Qwen-14B-LoRA
LoRA adapter for Qwen2.5-14B-Instruct, fine-tuned on GPT-4o conversations to create a warm, creative, and intelligent conversational model.
Usage
from peft import PeftModel
base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2.5-14B-Instruct")
model = PeftModel.from_pretrained(base_model, "Lighstromo/Nova-Qwen-14B-LoRA")
tokenizer = AutoTokenizer.from_pretrained("Lighstromo/Nova-Qwen-14B-LoRA")
Training Details
Setting Value
Base Model Qwen/Qwen2.5-14B-Instruct
LoRA Rank 64
LoRA Alpha 128
Target Modules q_proj, k_proj, v_proj, o_proj, gate_proj, up_proj, down_proj
Dataset 3,066 unique GPT-4o conversations
Epochs 3
Learning Rate 2e-4 (cosine)
Final Loss 0.29
Hardware RTX 6000 Ada on RunPod
GGUF Versions: Lighstromo/Nova-Qwen-14B-GGUF
- Task
- Text Generation
- Library
- peft
- License
- apache-2.0
- Languages
- en
- Created
- Mar 2, 2026
- Updated
- Mar 2, 2026