Qwen2.5-3B-Instruct-writing-medium-heavy

🎯 WRITING-optimized | 📦 Medium Heavy pruning | ⚡ 18% weights pruned

This model is a moderate-heavyly pruned version of Qwen/Qwen2.5-3B-Instruct, specialized for WRITING tasks using activation-aware weight pruning (Wanda-style).

✨ Key Features

• Specialization: Optimized for Writing tasks
• Pruning Method: Wanda-style (|W| × |activation|) importance scoring
• Size Reduction: 18% weights pruned
• Use Case: Significant compression, some accuracy trade-off

📊 Performance Comparison

Category	Original	Pruned	Change
Python	100.0%	100.0%	→
Html	6.7%	6.7%	→
Trivia	100.0%	93.3%	↓ 6.7%
Math	60.0%	66.7%	↑ 6.7%
Reasoning	N/A	N/A
Medical	100.0%	100.0%	→
Linux	100.0%	100.0%	→
Writing	73.3%	86.7% ⭐	↑ 13.3%

Average: 77.1% → 79.0% (+1.9%)

Writing Retention: 118.2% of original performance

🚀 Quick Start

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("CompactAI/Qwen2.5-3B-Instruct-writing-medium-heavy")
tokenizer = AutoTokenizer.from_pretrained("CompactAI/Qwen2.5-3B-Instruct-writing-medium-heavy")

# Example usage
inputs = tokenizer("Your prompt here", return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=100)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

📋 Technical Details

Property	Value
Base Model	Qwen/Qwen2.5-3B-Instruct
Specialization	Writing
Prune Mode	Medium Heavy
Pruning Method	Activation-based weight pruning (Wanda)
Weight Reduction	18% weights pruned

🔗 Related Models

This model is part of the Qwen2.5-3B-Instruct pruned model collection. Other variants:

• Extra-light (minimal pruning)
• Light
• Medium-light
• Medium
• Medium-heavy
• Heavy
• Extra-heavy (maximum compression)

📜 License

This model inherits the license from the base model Qwen/Qwen2.5-3B-Instruct.

---

Generated by ZANNPS [Zeto Automatic Neural Network Pruning System]