Downloads · 30 days
10
13% of all-time downloads
ApexVOrteX-1/egyptian-qwen-adapter
egyptian-qwen-adapter is a text generation model from ApexVOrteX-1. 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.
This repository contains a LoRA adapter fine-tuned on Qwen/Qwen2.5-1.5B-Instruct to improve conversational responses in the Egyptian Arabic dialect.
Downloads · 30 days
10
13% of all-time downloads
All-time downloads
75
Public
Repo size
13.6 MB
Likes
1
Public
Click a slice to open those files.
.json11.4 MB · 84%
From the Hugging Face model README
This repository contains a LoRA adapter fine-tuned on Qwen/Qwen2.5-1.5B-Instruct to improve conversational responses in the Egyptian Arabic dialect.
The objective of this project is style adaptation, enabling the base model to generate more natural Egyptian Arabic conversations while preserving the knowledge and reasoning capabilities of the original Qwen model.
Note: This repository contains only the LoRA adapter weights, not the full base model.
The adapter was fine-tuned on a merged dataset containing approximately 57K Egyptian Arabic conversational examples collected from multiple sources, including:
The data was cleaned, unified, and converted into chat-style message format before training.
Base Model: Qwen2.5-1.5B-Instruct
PEFT: LoRA
Quantization: 4-bit NF4 (QLoRA)
Optimizer: Paged AdamW 8-bit
Frameworks:
This adapter is intended for:
from transformers import AutoTokenizer, AutoModelForCausalLM
from peft import PeftModel
base_model = AutoModelForCausalLM.from_pretrained(
"Qwen/Qwen2.5-1.5B-Instruct"
)
model = PeftModel.from_pretrained(
base_model,
"ApexVOrteX-1/egyptian-qwen-adapter"
)
tokenizer = AutoTokenizer.from_pretrained(
"Qwen/Qwen2.5-1.5B-Instruct"
)
This adapter follows the license of the base Qwen model. Please ensure compliance with the original model's license before commercial use.